209
ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD PNEUMONIA AND DIARRHOEA Zhiwei Xu BMed, MMed Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy School of Public Health and Social Work Faculty of Health Queensland University of Technology April 2015

ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

  • Upload
    others

  • View
    6

  • Download
    0

Embed Size (px)

Citation preview

Page 1: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD PNEUMONIA AND

DIARRHOEA

Zhiwei Xu BMed, MMed

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Public Health and Social Work

Faculty of Health

Queensland University of Technology

April 2015

n8022411
Typewritten Text
Page 2: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects
Page 3: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

i

Keywords

Children

Climate change

Cold spell

Diarrhoea

Geographic co-distribution

Heat wave

Pneumonia

Remote sensing

Temperature

Page 4: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

ii

Abstract

Pneumonia and diarrhoea are the two leading killers of children under five years, and

climate change may impact the burden of these two paediatric diseases, even though

the strength of the association between climate change and childhood pneumonia and

diarrhoea remains largely unclear, especially in subtropical regions.

This thesis identified the risk areas of childhood pneumonia and diarrhoea in

Queensland and quantified the effects of temperature on emergency department visits

(EDVs) for childhood pneumonia and diarrhoea. Specifically, three research

questions were answered in this thesis: (I) What are the spatial and temporal patterns

of EDVs for childhood pneumonia and diarrhoea in Queensland, and is there any

geographic co-distribution of these two diseases? (II) What is the relationship

between extreme temperatures and EDVs for childhood pneumonia and diarrhoea?

(III) What is the relationship between temperature variability and EDVs for

childhood pneumonia and diarrhoea?

In a systematic review, I discussed how climatic factors may impact childhood

pneumonia and diarrhoea, and found that the symbolic parameter of climate change,

namely, increasing temperature, is reported as the most likely climatic factor

associated with childhood pneumonia and diarrhoea. I also identified six knowledge

gaps in the existing body of knowledge, two of which are filled by this thesis.

Page 5: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

iii

As the spatiotemporal patterns of childhood pneumonia and diarrhoea in Queensland

remain unknown, I attempted to address this issue. The results of this study showed

that childhood pneumonia and diarrhoea mainly distributed in northwest of

Queensland, and Mount Isa had a high-risk cluster where childhood pneumonia and

diarrhoea co-distributed.

To assess the impact of extreme temperatures and prolonged extreme temperatures

(i.e., heat waves and cold spells) on EDVs for childhood pneumonia and diarrhoea in

Brisbane from 2001 to 2010, I conducted two time-series studies, and observed that

both high and low temperatures were associated with increases in EDVs for

childhood pneumonia and diarrhoea. During the study period, there was a decreasing

trend in the high temperature effect on childhood pneumonia, while the low

temperature effect on childhood pneumonia experienced an increasing trend. Heat

waves had significant added effects on childhood pneumonia and diarrhoea, and the

magnitude of these effects increased with intensity and duration. Added effects of

cold spells on childhood pneumonia were also detected.

Data on the effects of temperature variability on childhood pneumonia and diarrhoea

are scarce, even though a big temperature change may pose a threat to the less-

developed immune system of children and thus may trigger their underlying

respiratory or intestinal health condition. In light of this, I did the other two studies

looking at the impacts of temperature variability (defined as diurnal temperature

range (DTR), and temperature change between two neighbouring days (TCN)) on

EDVs for childhood pneumonia and diarrhoea. The results of these two studies

Page 6: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

iv

suggested that big TCN may increase EDVs for childhood pneumonia and diarrhoea.

Great DTR was associated with the increase in EDVs for childhood diarrhoea.

Indigenous children were particularly vulnerable to the impact of temperature

variability.

In summary, this thesis adds to the large body of literature on climate variability

impact on children’s health and may have significant implications for developing

climate change adaptation and paediatric care policies. Children’s health in Mount

Isa, the city where childhood pneumonia and diarrhoea co-distributed, requires more

attention from scientific and policy communities. This study improves our

understanding on how increasing temperature may affect the burden of two important

childhood diseases, pneumonia and diarrhoea, as climate change progresses. It also

calls for attention to the possible adverse impact of large temperature variability on

children’s health.

Page 7: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

v

Table of Contents

Keywords .................................................................................................................................................i

Abstract .................................................................................................................................................. ii

Table of Contents .................................................................................................................................... v

List of Figures ........................................................................................................................................vi

List of Tables .........................................................................................................................................ix

List of Abbreviations ............................................................................................................................ xii Statement of Original Authorship ....................................................................................................... xiii

Acknowledgements .............................................................................................................................. xiv

CHAPTER 1: INTRODUCTION ....................................................................................................... 1

CHAPTER 2: LITERATURE REVIEW ......................................................................................... 11

CHAPTER 3: RESULTS PAPER ONE ........................................................................................... 43

CHAPTER 4: RESULTS PAPER TWO .......................................................................................... 65 CHAPTER 5: RESULTS PAPER THREE .................................................................................... 103

CHAPTER 6: RESULTS PAPER FOUR ....................................................................................... 133

CHAPTER 7: RESULTS PAPER FIVE ........................................................................................ 157

CHAPTER 8: DISCUSSION AND CONCLUSIONS ................................................................... 175

APPENDICES ................................................................................................................................... 189 Appendix A Publications .................................................................................................................... 189

Page 8: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

vi

List of Figures

Figure 1-1. Conceptual framework of the thesis…………………………. 4

Figure 2-1. The flow chart of literature selection process………………... 16

Figure 3-1. The spatial distribution of EDVs for childhood pneumonia and diarrhoea

in Queensland, from 2007 to 2011………………………………………… 50

Figure 3-2. The spatial distribution of EDVs for childhood pneumonia by age in

Queensland, from 2007 to 2011.…………………………………………... 51

Figure 3-3. The spatial distribution of EDVs for childhood pneumonia by gender in

Queensland, from 2007 to 2011…………………………………..……….. 52

Figure 3-4. The spatial distribution of EDVs for childhood diarrhoea by age in

Queensland, from 2007 to 2011…………………………………..……….. 53

Figure 3-5. The spatial distribution of EDVs for childhood diarrhoea by gender in

Queensland, from 2007 to 2011…………………………………..……….. 54

Figure 3-6. The change of EDVs for childhood pneumonia and diarrhoea in

Queensland…………………………………..…………………………...... 55

Figure 3-7. The daily distribution of EDVs for childhood pneumonia and diarrhoea

in Queensland…………………………………..………………………….. 56

Figure 3-8. The spatial clusters of EDVs for childhood pneumonia and diarrhoea in

Queensland.……………………………………..………...............................57

Figure 3-9. The spatial patterns of mean temperature and rainfall in Queensland,

from 2007 to 2011.……………………………………..…………………... 58

Figure 4-1. The areas where satellite remote sensing temperature data were

collected…………………………………..………………………………... 71

Page 9: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

vii

Figure 4-2a. The decomposed distribution of EDVs for paediatric pneumonia in

Brisbane, from 2001 to 2010………………………………………………. 76

Figure 4-2b. The daily distributions of climate variables in Brisbane, from 2001 to

2010…………………………………..…………………………………… 77

Figure 4-2c. The daily distributions of air pollutants in Brisbane, from 2001 to

2010…………………………………..…………………………………… 78

Figure 4-3. The pairwise plot of paediatric pneumonia, mean temperature, rainfall

and relative humidity in Brisbane, from 2001 to 2010……………………. 80

Figure 4-4. The overall effect of mean temperature on paediatric pneumonia in

Brisbane, from 2001 to 2010………………………………………………. 81

Figure 4-5. The change over time in the temperature effect on childhood pneumonia;

left hand side: hot effect; right hand side: cold effect; p1= 2001-2005, p2=2002-

2006, p3=2003-2007, p4=2004-2008, p5=2005-2009, p6=2006-2010……. 85

Figure 5-1. The daily distributions of EDVs for paediatric diarrhoea and climatic

factors in Brisbane, from 2001 to 2010……………………………………. 110

Figure 5-2. The daily distribution of diarrhoea caused by different

pathogens…………………………………..……………………………… 111

Figure 5-3. The overall effect of mean temperature on paediatric diarrhoea in

Brisbane, from 2001 to 2010……………………………………………… 114

Figure 5-4. The change over time in the temperature effect on childhood diarrhoea;

left hand side: hot effect; right hand side: cold effect; p1= 2001-2005, p2=2002-

2006, p3=2003-2007, p4=2004-2008, p5=2005-2009, p6=2006-2010…… 118

Figure 6-1. The daily distributions of EDVs for paediatric pneumonia, mean

temperature, DTR and TCN in Brisbane, from 2001 to 2010…………….. 139

Page 10: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

viii

Figure 6-2. The overall effects of DTR and TCN on paediatric pneumonia in

Brisbane, from 2001 to 2010………………………………………………. 142

Figure 6-3. Monthly average number of days with TCN < -2 °C………… 143

Figure 6-4. The lagged effect of TCN on childhood pneumonia…………. 144

Figure 6-5. The effect of TCN on the total-, age-, gender- and ethnic-specific

childhood pneumonia in Brisbane, from 2001 to 2010 (Relative risk: The risk of

EDVs for childhood pneumonia on days with temperature drop =5.7 °C relative to

days with temperature drop= 2.0 °C) …………………………………….. 145

Figure 6-6. The effects of TCN on childhood pneumonia in summer and

winter…………………………………..…………………………………. 146

Figure 6-7. The overall effects of TCN on childhood pneumonia during two

periods…………………………………..………………………………… 147

Figure 6-8. The overall effect of TCN on childhood pneumonia in Brisbane, from

2001 to 2010 (excluding 2009) …………………………………………… 148

Figure 7-1. The daily distributions of EDVs for paediatric diarrhoea, mean

temperature, DTR and TCN in Brisbane, from 2001 to 2010……………... 163

Figure 7-2. The overall effects of DTR and TCN on paediatric diarrhoea in Brisbane,

from 2001 to 2010…………………………………..……………………... 166

Figure 7-3. The monthly distribution of days when DTR > 17 °C and TCN < -2

°C.…………………………………..………… ………………………….. 167

Figure 7-4a. The effect of DTR on the total-, age-, gender- and ethnic-specific

childhood diarrhoea in Brisbane, from 2001 to 2010……………………… 168

Figure 7-4b. The effect of TCN on the total-, age-, gender- and ethnic-specific

childhood diarrhoea in Brisbane, from 2001 to 2010. …………………….. 168

Page 11: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

ix

List of Tables

Table 2-1. Characteristics of studies about ambient temperature and childhood

pneumonia…………………………………..…………………………….. 21

Table 2-2. Characteristics of studies about ambient temperature and childhood

diarrhoea…………………………………..……………………………… 23

Table 2-3. Characteristics of studies about rainfall and childhood pneumonia

…………………………………..………………………..………………. 26

Table 2-4. Characteristics of studies about rainfall and childhood diarrhoea

…………………………………..………………………..………………. 27

Table 2-5. Characteristics of studies about relative humidity and childhood

pneumonia …………………………………..……………………………. 29

Table 2-6. Characteristics of studies about relative humidity and childhood

diarrhoea…………………………………..……………………………… 30

Table 3-1. Summary statistics for EDVs for childhood pneumonia and diarrhoea by

postcode in Queensland, Australia, during 2007-2011………..………….. 47

Table 4-1. Summary statistics for climatic variables, air pollutants and paediatric

pneumonia in Brisbane, Australia, 2001–2010…………………………… 75

Table 4-2. Spearman’s correlation between daily weather conditions, air pollutants

and paediatric pneumonia in Brisbane, Australia, from 2001–

2010…………………………………..…………………………………… 79

Table 4-3. The cumulative effect of high and low temperatures on EDVs for

paediatric pneumonia, with 99th percentile (29.6 °C) and 1st percentile (10.4 °C) of

temperature relative to reference temperature (23°C) …………………… 82

Page 12: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

x

Table 4-4. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2001 to 2010…………………………. 84

Table 4-5a. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2001 to 2005 ………………………… 87

Table 4-5b. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2002 to 2006………………………… 88

Table 4-5c. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2003 to 2007………………………… 89

Table 4-5d. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2004 to 2008………………………… 90

Table 4-5e. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2005 to 2009………………………… 91

Table 4-5f. Paediatric pneumonia due to the added effect of heat waves and cold

spells in Brisbane, Australia, from 2006 to 2010………………………… 92

Table 5-1. Summary statistics for climatic variables and paediatric diarrhoea in

Brisbane, Australia, 2001–2010………………………………………….. 109

Table 5-2. Spearman’s correlation between daily weather conditions, air pollutants

and paediatric diarrhoea in Brisbane, Australia, from 2001–2010………. 112

Table 5-3. The cumulative effect of high and low temperatures on EDVs for

paediatric diarrhoea in Brisbane, with 99th percentile (29.6 °C) and 1st (10.4°C) of

temperature relative to reference temperature (16 °C) …………………… 115

Table 5-4. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2001 to 2010……………………………………………… 117

Table 5-5a. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2001 to 2005…………………………………..…………. 119

Page 13: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

xi

Table 5-5b. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2002 to 2006…………………………………………….. 120

Table 5-5c. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2003 to 2007…………………………………………….. 121

Table 5-5d. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2004 to 2008…………………………………..………… 122

Table 5-5e. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2005 to 2009…………………………………..…………. 123

Table 5-5f. Paediatric diarrhoea due to the added effect of heat waves in Brisbane,

Australia, from 2006 to 2010…………………………………………….. 124

Table 6-1. Spearman’s correlation between daily weather conditions, air pollutants

and paediatric pneumonia in Brisbane, Australia, from 2001–2010…….. 140

Table 7-1. Summary statistics for climatic variables, air pollutants and paediatric

diarrhoea in Brisbane, Australia, 2001–2010……………………………. 162

Table 7-2. Spearman’s correlation between daily weather conditions, air pollutants

and paediatric diarrhoea in Brisbane, Australia, from 2001–

2010…………………………………..………………………………….. 164

Page 14: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

xii

List of Abbreviations

ARIMA Autoregressive integrated moving average

CI Confidence interval

DLNM Distributed lag non-linear model

DOW Day of week

DTR Diurnal temperature range

EDV Emergency department visit

ICD9 International classification of disease, ninth revision

ICD10 International classification of disease, tenth revision

IPCC Intergovernmental Panel on Climate Change

MeSH Medicine’s Medical Subject Headings

NASA National Aerospace and Space Administration

NO2 Nitrogen dioxide

O3 Ozone

OR Odds ratio

PM10 Particulate matter less than 10 μm in aerodynamic diameter

RR Relative risk

SD Standard deviation

TCN Temperature change between two neighbouring days.

Page 15: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

QUT Verified Signature

Page 16: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

xiv

Acknowledgements

I am very grateful to many people who have contributed in various ways to the

completion of this thesis. First and foremost, I would like to thank my supervisory

team, Prof. Shilu Tong, Dr. Wenbiao Hu, Dr. Weiwei Yu and Prof. Hong Su.

I would like to express my heartfelt gratitude to Shilu, my dear principal supervisor,

for his support in my whole PhD journey. He provided me with an opportunity to do

my PhD in QUT, and encouraged me to pursue a promising research topic. He has

always supported me without any reservation despite his busy schedule, teaching me

not only how to be a good researcher but also how to be a righteous person.

Particularly, he has put a lot of efforts to the design, conceptualisation and revision

of my manuscripts. I am so blessed to be a PhD student of Shilu and I always think

that he is one of the best supervisors ever. I am sure that the invaluable things which

I’ve learned from him will always guide me in my future career.

I am deeply grateful to Wenbiao, my associate supervisor, who has also made big

contributions to my research. He was exceptionally generous with his time, teaching

me how to do spatial and temporal analysis. Without his help, most goals are

unachievable in my PhD.

I am sincerely thankful to Dr. Weiwei Yu and Prof. Hong Su for their contributions

to my research.

I would also like to acknowledge the generous financial support for the completion

of my PhD program, including a China Scholarship Council Postgraduate

Scholarship, Queensland University of Technology Fee Waiving Scholarship, and a

CSIRO Top-up Scholarship from the Climate Adaptation Flagship Collaboration

Fund.

Page 17: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

xv

Special thanks to my dear friends and colleagues who went through the journey with

me (especially Xiaoyu Wang, Xin Qi, Xiaofang Ye, Yan Bi, Cunrui Huang, Yuming

Guo, Jiajia Wang, Lyle R. Turner, Shahera Banu, Suchithra Naish, and Sam Toloo).

My deepest expression of appreciation is to my dear family for their patience and

support. Without their love, any achievement is meaningless.

Page 18: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects
Page 19: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 1: Introduction 1

Chapter 1: Introduction

1.1 Background

Pneumonia and diarrhoea are the two major killers in children younger than five

years (Walker et al. 2013). It is estimated that in 2010, there were more than 1.7

billion episodes of diarrhoea and 120 million episodes of pneumonia in children

under five years (Walker et al. 2013). Although mortality due to pneumonia and

diarrhoea has been declining in the past decades in developed countries, morbidity

from pneumonia and diarrhoea in these regions still remains high (Podewils et al.

2004). For example, pneumonia and diarrhoea are the common causes of

hospitalization for Australian children (Rudan et al. 2013; Scallan et al. 2005).

Pneumonia and diarrhoea are usually preventable (Bhutta et al. 2013). Prior studies

have put emphasis on identifying some individual-level risk factors, such as under

nutrition (Black et al. 2008) and not exclusively breastfeeding (Walker et al. 2013).

Some of these factors are associated with both pneumonia and diarrhoea, which may

result in the geographic co-distribution of pneumonia and diarrhoea (Walker et al.

2013), even though the geographic co-distribution of pneumonia and diarrhoea has

not been explored in previous studies.

Existing evidence also suggests that climatic factors, such as temperature (Checkley

et al. 2000; Green et al. 2010) and rainfall (Garcia-Vidal et al. 2013; Hashizume et al.

2007), may be associated with the occurrence of pneumonia and diarrhoea,

highlighting that increasing temperature may increase the incidence of pneumonia

and diarrhoea. However, prior studies looking at the impacts of temperature on

pneumonia and diarrhoea mainly used the temperature data collected from ground

Page 20: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

2 Chapter 1: Introduction

monitoring sites (Checkley et al. 2000; Green et al. 2010), which may cause

measurement bias as most of these monitoring sites were in or nearby urban areas but

temperature usually varies across one city (Estes et al. 2009; Laaidi et al. 2012).

Satellite remote sensing technology may largely solve this problem, given its broad

spatial coverage (Anderson et al. 2012; Estes et al. 2009; Evans et al. 2013). To the

best of my knowledge, no study has used satellite remote sensing data to look at the

effects of temperature on childhood diseases so far.

There is a widespread consensus that climate is changing (IPCC 2013), and climate

change has posed a huge threat to children’s health (Xu et al. 2012b). Particularly,

global burden of childhood pneumonia and diarrhoea may continue to rise due to the

Earth’s increasing average surface temperature (Walker et al. 2013). Further, the

frequency of unstable weather patterns (e.g., sharp increase/decrease in temperature)

will also increase (Epstein 2005), and children are particularly vulnerable to big

temperature variation (Xu et al. 2013), due to their relatively less-developed

thermoregulation capability (Xu et al. 2012a).

Understanding the spatial and temporal patterns of childhood pneumonia and

diarrhoea (especially their geographic co-distribution), and exploring the climatic

drivers behind them using accurate exposure data will shed new light on future effect

to control and prevent these two leading childhood diseases. However, to date, no

studies have assessed the spatiotemporal patterns of childhood pneumonia and

diarrhoea in Queensland. Also, no studies have identified the climatic drivers of

childhood pneumonia and diarrhoea using satellite remote sensing data. This thesis

aimed to fill these knowledge gaps. The conceptual framework for this thesis is

shown in Figure 1-1.

Page 21: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 1: Introduction 3

1.2 Purposes

In this thesis, I used the data on emergency department visits for childhood

pneumonia and diarrhoea in Queensland, Australia, from January 1st 2001 to

December 31st 2011 and addressed three key issues:

I. What are the spatial and temporal patterns of EDVs for childhood pneumonia

and diarrhoea in Queensland, and is there any geographic co-distribution

of these two diseases?

II. What is the relationship between extreme temperatures and EDVs for

childhood pneumonia and diarrhoea?

III. What is the relationship between temperature variations and EDVs for

childhood pneumonia and diarrhoea?

Page 22: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

4 Chapter 1: Introduction

Figure 1-1. Conceptual framework of the thesis

Chapter 5: Extreme temperature effect on

childhood diarrhoea in Brisbane

Climate change

Temperature and weather extremes increase

Climate variability may impact the burden of childhood

pneumonia and diarrhoea

Chapter 2: Identify the

knowledge gaps

Chapter 3: Explore the spatiotemporal patterns of childhood pneumonia

and diarrhoea in Queensland

Chapter 4: Extreme temperature effect on childhood pneumonia

in Brisbane

Chapter 6: Temperature

variability effect on childhood pneumonia

in Brisbane

Chapter 7: Temperature

variability effect on childhood diarrhoea in

Brisbane

Chapter 5: Extreme temperature effect on

childhood diarrhoea in Brisbane

Page 23: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 1: Introduction 5

1.3 Thesis Outline

This thesis is presented in a publication style. As such, each chapter is designed to

stand alone. In Chapter two, I reviewed the existing literature regarding the

association between climatic factors and childhood pneumonia and diarrhoea, and

found that temperature is the most likely climatic driver of childhood pneumonia and

diarrhoea. As climate change continues, burden of childhood pneumonia and

diarrhoea may vary, and the variation in specific regions may be largely driven by

climatic conditions, and the proportion of vulnerable children.

In Chapter three, I explored the spatiotemporal patterns and geographic co-

distribution of childhood pneumonia and diarrhoea in Queensland. A distinct

seasonality of childhood pneumonia and diarrhoea was found. Childhood pneumonia

and diarrhoea mainly distributed in northwest of Queensland, and Mount Isa had a

high-risk cluster where childhood pneumonia and diarrhoea co-distributed.

Chapter four is a time-series study looking at the impacts of extreme temperatures

and persistent extreme temperatures (i.e., heat waves and cold spells) on EDVs for

childhood pneumonia in Brisbane. It is observed that both high and low temperatures

were associated with an increase in EDVs for childhood pneumonia. Heat waves and

cold spells had significant added effects on childhood pneumonia, and the magnitude

of these effects increased with intensity and duration. However, there were changes

over time in both the main and added effects of temperature on childhood

pneumonia.

In Chapter five, I examined the effects of extreme temperatures and heat waves on

EDVs for childhood diarrhoea in Brisbane. We found both low and high

temperatures had significant impacts on childhood diarrhoea. Heat waves had an

Page 24: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

6 Chapter 1: Introduction

added effect on childhood diarrhoea, and this effect increased with intensity and

duration of heat waves. There was a decreasing trend of heat effect on childhood

diarrhoea in Brisbane across the study period.

Chapter six is designed to assess the impact of temperature variability (diurnal

temperature range (DTR) and temperature change between two neighbouring days

(TCN)) on EDVs for childhood pneumonia in Brisbane. An adverse impact of TCN

on EDVs for childhood pneumonia was observed, and the magnitude of this impact

increased from the first five years (2001–2005) to the second five years (2006–2010).

Children aged 5–14 years, female children and Indigenous children were particularly

vulnerable to TCN impact. However, there was no significant association between

DTR and EDVs for childhood pneumonia.

Chapter seven is a time-series study quantifying the effect of temperature variability

on EDVs for childhood diarrhoea. It was observed that high DTR and TCN were

significantly associated with an increase in EDVs for childhood diarrhoea in

Brisbane. Every year, from May to September, especially July, children were at a

high risk posed by high DTR and low TCN, and male children were particularly

vulnerable to the adverse impact of DTR and TCN on diarrhoea.

Chapter eight highlighted the key findings of this thesis, compared these findings

with prior studies, discussed the mechanisms underlying these findings and

implications behind these finding, and explored how to manage the effects of

temperature on childhood pneumonia and diarrhoea.

Page 25: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 1: Introduction 7

1.4 References

Anderson H, Butland B, Donkelaar AV, Brauer M, Strachan D, Clayton T, et al.

2012. Satellite-based estimates of ambient air pollution and global variations

in childhood asthma prevalence. Environ Health Perspect 120(9):1333-1339.

Bhutta ZA, Das JK, Walker N, Rizvi A, Campbell H, Rudan I, et al. 2013.

Interventions to address deaths from childhood pneumonia and diarrhoea

equitably: what works and at what cost? Lancet 381(9875):1417-1429.

Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. 2008.

Maternal and child undernutrition: global and regional exposures and health

consequences. Lancet 371(9608):243-260.

Checkley W, Epstein L, Gilman R, Figueroa D, Cama R. Patz J, et al. 2000. Effect of

El Niño and ambient temperature on hospital admissions for diarrhoeal

diseases in Peruvian children. Lancet 355(9202):442-450.

Epstein PR. 2005. Climate change and human health. N Engl J Med 353(14):1433-

1436.

Estes M, Al-Hamdan M, Crosson W, Estes S, Quattrochi D, Kent S, et al. 2009. Use

of remotely sensed data to evaluate the relationship between living

environment and blood pressure. Environ Health Perspect 117(12):1832-

1838.

Evans J, van Donkelaar A, Martin RV, Burnett R, Rainham DG, Birkett NJ, et al.

2013. Estimates of global mortality attributable to particulate air pollution

using satellite imagery. Environ Res 120:33-42.

Garcia-Vidal C, Labori M, Viasus D, Simonetti A, Garcia-Somoza D, Dorca J, et al.

2013. Rainfall is a risk factor for sporadic cases of Legionella pneumophila

Pneumonia. PLoS ONE 8(4):e61036.

Page 26: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

8 Chapter 1: Introduction

Green R, Basu R, Malig B, Broadwin R, Kim J, Ostro B. 2010. The effect of

temperature on hospital admissions in nine California counties. Int J Public

Health 55(2):113-121.

Hashizume M, Armstrong B, Hajat S, Wagatsuma Y, Faruque AS, Hayashi T, et al.

2007. Association between climate variability and hospital visits for non-

cholera diarrhoea in Bangladesh: effects and vulnerable groups. Int J

Epidemiol 36(5):1030-1037.

IPCC. 2013. Summary for policymakers. In: Climate change 2013: the physical

science basis. Contribution of Working Group I to the Fifth Assessment

Report of the Intergovernmental Panel on Climate Change. Cambridge

University Press, Cambridge.

Laaidi K, Zeghnoun A, Dousset B, Bretin P, Vandentorren S, Giraudet E, et al. 2012.

The impact of heat islands on mortality in Paris during the August 2003 heat

wave. Environ Health Perspect 120(2):254-259.

Podewils LJ, Mintz ED, Nataro JP, Parashar UD. 2004. Acute, infectious diarrhea

among children in developing countries. Semin Pediatr Infect Dis 15(3):155-

168.

Rudan I, O'Brien K, Nair H, Liu L, Theodoratou E, Qazi S, et al. 2013.

Epidemiology and etiology of childhood pneumonia in 2010: estimates of

incidence, severe morbidity, mortality, underlying risk factors and causative

pathogens for 192 countries. J Glob Health 3(1):10401.

Scallan E, Majowicz SE, Hall G, Banerjee A, Bowman CL, Daly L, et al. 2005.

Prevalence of diarrhoea in the community in Australia, Canada, Ireland, and

the United States. Int J Epidemiol 34(2), 454-460.

Page 27: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 1: Introduction 9

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Xu Z, Etzel RA, Su H, Huang C, Guo Y, Tong S. 2012a. Impact of ambient

temperature on children's health: A systematic review. Environ Res 117:120-

131.

Xu Z, Huang C, Su H, Turner L, Qiao Z, Tong S. 2013. Diurnal temperature range

and childhood asthma: a time-series study. Environ Health 12(1):12.

Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, et al. 2012b. Climate change and

children’s health—A call for research on what works to protect children. Int J

Environ Res Public Health 9(9):3298-3316.

Page 28: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects
Page 29: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 11

Chapter 2: Literature Review

Impact of climatic factors on childhood pneumonia and diarrhoea: a

review of literature

Zhiwei Xu, Wenbiao Hu, Shilu Tong

Page 30: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

12 Chapter 2: Literature Review

Abstract

Climate change is affecting and will continue to impact children’s health. To assess the

relationship between climatic factors and childhood pneumonia and diarrhoea, a literature

search was conducted using the databases PubMed, ProQuest, ScienceDirect, Scopus and

Web of Science. Empirical papers looking at the impact of climatic factors (defined as

temperature, rainfall and relative humidity) on childhood pneumonia or diarrhoea published

up to April 1st 2014 were included. Existing literature suggests that temperature is an

important climate driver of childhood pneumonia and diarrhoea. High or low rainfall, and

low relative humidity, may also increase the occurrence of childhood diarrhoea. Little

evidence regarding the effects of rainfall and relative humidity on childhood pneumonia was

found. As climate change continues, burden of childhood pneumonia and diarrhoea may vary,

and the variation in specific regions may be due largely to the major aetiological agent,

climate zones, and the proportion of vulnerable children. Future research should focus on

assessing temperature effects on childhood pneumonia and diarrhoea using more accurate

temperature data, exploring the relationship between temperature variation and childhood

pneumonia and diarrhoea, quantifying the vulnerability of different aetiological agents of

childhood pneumonia and diarrhoea to climatic factors, elucidating the modified effects of

climate types on the climate-pneumonia and climate-diarrhoea relationships, projecting the

future burden of childhood pneumonia and diarrhoea attributable to climate change, and

developing cost-effective adaptation measures to protect children from the adverse impact of

climate change.

Keywords: climate change; children; pneumonia; diarrhoea

Page 31: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 13

2.1 Introduction

Pneumonia and diarrhoea are the two major killers of children (Walker et al. 2013). In 2010,

there were more than 1.7 billion diarrhoea episodes and around 120 million pneumonia

episodes in children aged under five years, worldwide, causing approximately two million

deaths in 2011 (Walker et al. 2013).

Identifying the risk factors and taking preventive measures to block children from these risk

factors are urgently needed, given the huge burden of childhood pneumonia and diarrhoea.

Existing body of knowledge has acknowledged that individual-level biological and behaviour

factors, including suboptimum breastfeeding, underweight, stunting and zinc deficiency, may

largely contribute to the occurrence of pneumonia and diarrhoea (Walker et al. 2013). Some

prior studies have reported that climatic factors, such as temperature (Checkley et al. 2000;

Green et al. 2010), rainfall (Garcia-Vidal et al. 2013; Hashizume et al. 2007) and relative

humidity (D'Souza et al. 2008; Onozuka et al. 2009), may also be associated with the

transmission of pneumonia and diarrhoea. However, regional differences and contrasting

effects of climate on pneumonia and diarrhoea due to different aetiological agents are evident

(Ebi et al. 2001; Green et al. 2010; Hashizume et al. 2007; Paynter et al. 2013; Sumi et al.

2013).

As projected by Intergovernmental Panel on Climate Change (IPCC), the global surface

average temperature may increase by 1.8 to 4.0 °C relative to the 1961–1990 level by the end

of this century (IPCC 2007b), and more intense rainy seasons are likely to occur in Africa

and Asia (IPCC 2007a), the two regions where the burden of childhood pneumonia and

diarrhoea are the greatest. Climate change may affect the burden of childhood pneumonia and

diarrhoea (Walker et al. 2013), especially in some low-income countries. It is essential to

systematically review the relationship between climatic factors and childhood pneumonia and

diarrhoea so as to find out how climate change may impact burden of these two children

Page 32: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

14 Chapter 2: Literature Review

killers in the future. This review aimed to elucidate the effects of climate (defined as

temperature, rainfall and relative humidity) on childhood pneumonia and diarrhoea, and to

propose future research directions. Herein, children were defined as humans under 18 years

old (American Academy of Pediatrics Committee on Environmental Health 2003).

2.2 Methods

Data sources

Empirical studies regarding climate and childhood pneumonia and diarrhoea published up to

April 1st 2014 were retrieved from the electronic databases PubMed, ProQuest,

ScienceDirect, Scopus, and Web of Science. References of the identified papers were

manually checked to make sure all relevant papers were included.

Inclusion criteria

We restricted our search to peer-reviewed articles written in English. The following U.S.

National Library of Medicine’s Medical Subject Headings (MeSH terms) and keywords were

used in the primary search: “climate change”, “climate”, “temperature”, “rainfall”,

“precipitation”, “humidity”, “child”, “pneumonia”, “diarrhea”, “diarrhoea” and “rotavirus”.

Eligibility included any empirical studies which used original data and appropriate effect

estimates (e.g., regression coefficient, relative risk, odds ratio, or percentage change in

morbidity for pneumonia or diarrhoea); where temperature, rainfall and/or relative humidity

was a main exposure of interest, and where childhood pneumonia and diarrhoea were

analysed. The effect estimates (e.g., relative risks, confidence intervals) were recorded from

the papers identified. In this review, we discussed the impact of climatic factors on rotavirus

diarrhoea separately as there are too many papers focusing on this topic.

Page 33: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 15

2.3 Results

We identified 8275 papers in the initial search (excluding papers talking about climate

variability and rotavirus), and 14 papers were in the final review (Figure 2-1). Five papers

have looked at the impact of climate on childhood pneumonia, and nine papers have

examined the association between climate and childhood diarrhoea.

Page 34: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

16 Chapter 2: Literature Review

Figure 2-1. The flow chart of literature selection process

Potentially relevant studies in the initial searching

(n= 8275)

7894 excluded due to irrelevant titles

Studies after reviewing the titles

(n=381)

344 did not meet inclusion criteria according to abstract

Studies retrieved for more detailed evaluation

(n=37)

26 articles excluded (5 no appropriate effect estimate; 3

no original data; 18 no specific data for children)

Studies met inclusion criteria (n=11)

3 articles added by inspecting reference lists

Studies included in final review (n=14)

Page 35: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 17

Temperature

Five studies, conducted in Brazil (Souza et al. 2012), China (Xu et al. 2011), Philippine

(Paynter et al. 2013), and the USA (Ebi et al. 2001; Green et al. 2010), have looked at the

impact of temperature on childhood pneumonia (Table 2-1). Ebi et al. examined the effect of

temperature on hospitalizations for viral pneumonia in six counties of California, the USA,

from 1993 to 1998, and found hospitalizations for viral pneumonia increased significantly

with the decrease of temperature in children aged under 18 years, though the percent changes

varied from 21.6% to 30.7% across different counties (Ebi et al. 2001). In the other nine

counties of California, Green et al. quantified the effect of high temperature on hospital

admissions for pneumonia from 1999 to 2005, and found hospital admissions for pneumonia

in children aged under five years increased by 5.9% per 10 oF (5.6 ºC) increase in mean

apparent temperature (Green et al. 2010). In Hangzhou, China, Xu et al. investigated the

relationship between temperature and childhood pneumonia and found Mycoplasma

pneumoniaie pneumonia rate in children under 18 years increased by 0.83% per 1ºC increase

in monthly mean air temperature (Xu et al. 2011). In Campo Grande, Brazil, Souza et al.

looked at the association between temperature and outpatient visits in children aged 5 to 14

years, and found minimum temperature was positively associated with pneumonia (relative

risk (RR): 1.12), and maximum temperature was negatively associated with pneumonia (RR:

0.92) (Souza et al. 2012). One note-worthy study conducted in Japan examined the impact of

mean temperature on reported Mycoplasma pneumonia in the total population, finding that

93.7% of the reported Mycoplasma pneumonia occurred in children under 15 years, and the

weekly number of Mycoplasma pneumonia cases increased by 16.9% for every 1% increase

in mean temperature (Onozuka et al. 2009). Recently, Paynter et al. used both time-series and

case-crossover designs to examine the effect of temperature on hospitalizations for

pneumonia in children aged under three years from 2000 to 2004 in Philippine, but did not

Page 36: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

18 Chapter 2: Literature Review

find any significant relationship between temperature and childhood pneumonia (RR: 0.93

(95% confidence interval (CI): 0.63 to 1.35)) (Paynter et al. 2013).

Seven studies from Australia (Lam 2007), Bangladesh (Hashizume et al. 2007), Japan

(Onozuka and Hashizume 2010), Peru (Checkley et al. 2000), Sub-Saharan Africa

(Bandyopadhyay et al. 2012), Taiwan (Chou et al. 2010), and the USA (Green et al. 2010)

have examined the relationship between temperature and childhood diarrhoea (Table 2-2).

Bandyopadhyay et al. explored the effect of monthly temperature on diarrhoea cases in 14

Sub-Saharan African countries from 1992 to 2001, finding that an increase in monthly

average maximum temperature raises the prevalence of diarrhoea while an increase in

monthly minimum temperature reduces diarrhoea in children under three years of age

(Bandyopadhyay et al. 2012). Two studies used maximum temperature as temperature

indicator to examine the temperature-diarrhoea relationship (Chou et al. 2010; Lam 2007). In

Sydney, Australia, Lam investigated the association between maximum temperature and

hospital emergency department visits for gastroenteritis in children aged under six years and

found hospital emergency department visits for gastroenteritis increased by 11% per 1ºC

increase in maximum temperature (Lam 2007). In Taiwan, Chou et al. explored the impact of

maximum temperature on hospital admissions for diarrhoea in children aged under 15 years

from 1996 to 2007, and found hospital admissions for diarrhoea increased by 4% per 1 °C

increase in maximum temperature (Chou, et al. 2010). Another four studies have explored the

effect of mean temperature on diarrhoea (Checkley et al. 2000; Green et al. 2010; Hashizume

et al. 2007; Onozuka and Hashizume 2010), and three of them reported increased diarrhoeal

cases associated with increasing temperature (Checkley et al. 2000; Green et al. 2010;

Hashizume et al. 2007). While, Onozuka et al. found that there was a temperature threshold,

either above or below which, hospital admissions for infectious gastroenteritis in children

aged under 15 years increased (Onozuka and Hashizume 2010).

Page 37: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 19

Rainfall

So far, only two studies in Philippines (Paynter et al. 2013) and the USA (Ebi et al. 2001)

have quantified the effect of rainfall on childhood pneumonia (Table 2-3). Ebi et al. explored

the effect of rainfall on viral pneumonia in six California counties from 1983 to 1998, but

found there was no significant relationship between rainfall and hospitalizations for viral

pneumonia in children aged under 18 years (Ebi et al. 2001). In Bohol Province, Philippines,

Paynter et al. investigated the association between rainfall and hospitalizations for pneumonia

from 2000 to 2004 in children aged under three years, and also found there was no significant

relationship between them (RR:2.37 (95% CI: 0.85 to 6.60)).

Four studies, conducted in Australia (Lam 2007), Bangladesh (Hashizume et al. 2007), Brazil

(Andrade et al. 2009), and the USA (Drayna et al. 2010), have looked at the impact of rainfall

on childhood diarrhoea (Table 2-4). In Rio Grande do Norte, Brazil, Andrade et al. examined

the effect of rainfall on hospitalizations for diarrhoea in infants from 1992 to 2001, and found

the diarrhoea hospitalizations increased with the increase of rainfall (Andrade et al. 2009).

Drayna et al. found that any rainfall four days prior was significantly associated with an 11%

increase in paediatric emergency department visits for acute gastrointestinal in Wisconsin, the

USA (Drayna et al. 2010). Hashizume et al. found that in Dhaka, Bangladesh, there was a

non-linear relationship between rainfall and childhood diarrhoea. Below or above a rainfall

threshold, hospital visits for non-cholera diarrhoea in children under 15 years increased

rapidly (Hashizume et al. 2007). However, in Sydney, Australia, Lam did not find any

significant relationship between rainfall and hospital emergency department visits for

gastroenteritis in children under six years of age (RR:0.98 (P=0.09)) (Lam 2007).

Relative humidity

There are two studies looking at the effect of relative humidity on pneumonia in children

(Paynter et al. 2013; Souza et al. 2012) and both of them did not find any significant

Page 38: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

20 Chapter 2: Literature Review

relationship between relative humidity and childhood pneumonia (Table 2-5). However,

Onozuka et al. examined the impact of relative humidity on reported Mycoplasma

pneumoniae pneumonia in not just children but also adults in Fukuoka, Japan, and found that

93.7% of the reported Mycoplasma pneumoniae pneumonia were in children under 15 years,

and the weekly number of Mycoplasma pneumoniae pneumonia cases increased by 4.1% for

every 1% increase in relative humidity (Onozuka et al. 2009).

Three studies have formally explored the impact of relative humidity on childhood diarrhoea

(D'Souza et al. 2008; Lam 2007; Onozuka and Hashizume 2010) (Table 2-6). Onozuka and

Hashizume quantified the effect of relative humidity on hospital admission for infectious

gastroenteritis in children under 15 years in Japan, and found the increase in diarrhoeal cases

per 1% drop in relative humidity was 3.9% (Onozuka and Hashizume 2010). While, Lam did

not find any significant relationship between relative humidity and hospital emergency

department visits for gastroenteritis in children aged under six years in Sydney, Australia

(Lam 2007).

Climate variability and rotavirus diarrhoea

Numerous studies have examined the impacts of temperature, rainfall and relative humidity

on rotavirus diarrhoea in children. Levy et al. reviewed the seasonality of rotavirus in the

tropics and its relationship with climate variability (Levy et al. 2009). They found rotavirus

incidence reduces by 10%, 1%, and 3%, respectively, for every 1 ºC increase in mean

temperature, 1 cm increase in mean monthly rainfall, and 1% increase in relative humidity.

Jagai et al. also quantitatively reviewed the impacts of meteorological factors on rotavirus in

South Asia, and found A 1 ºC decrease in monthly ambient temperature and a decrease of 10

mm in precipitation are associated with 1.3% and 0.3% increase above the annual level in

rotavirus infections, respectively (Jagai et al. 2012)

Page 39: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 21

Table 2-1. Characteristics of studies about ambient temperature and childhood pneumonia Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Ebi et al. 2001

Six California counties, the USA,

January 1983 to June 1998, children aged 0-

17 years

Time-series; Poisson generalized estimating equations

model

Sea surface temperature

Hospitalizations for viral

pneumonia in females

Hospitalizations for viral pneumonia increased

significantly with temperature decreased

Percent change in Sacramento and Yolo

Counties: 30.7 (95% CI: 29.0 to 32.4)

Percent change in San Francisco and San Mateo Counties: 24.6 (95% CI:

9.9 to41.2 ) Percent change in Los Angeles and Orange

Counties: 21.6 (95% CI: 16.4 to 27.1)

Green et al. 2010

Nine California counties, the USA, May to September,

1999 to 2005, children under five years

Case-crossover; Conditional

logistic regression,

meta-analysis

Daily mean apparent temperature

Hospital admissions for

pneumonia

Hospital admissions for pneumonia in children aged under five years

increased with the increase of mean

temperature

Percent change: 5.9% (95% CI: -1.7% to 14.1%)

Xu et al. 2011

Hangzhou, China, January 2007 to December 2009, children under 18

years

Multiple linear regression

Monthly mean air temperature

Lab-confirmed hospitalizations for pneumonia

Mycoplasma pneumonia rate increased by 0.83%

with an increase of 1ºC in monthly mean air

temperature

Percent change: 0.83%

Page 40: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

22 Chapter 2: Literature Review

Table 2-1. Characteristics of studies about ambient temperature and childhood pneumonia (continued)

Studya Location and time Research design

and statistical analysis

Main temperature exposure

variable(s) Outcome(s) Key findings Effect estimates

Souza et al. 2012

Campo Grande, Brazil, 2004 to

2008, children aged 5-14 years

Time-series: Poisson regression

Daily maximum and minimum temperatures

Outpatient visits for pneumonia

Minimum temperature was positively associated with pneumonia, and

maximum temperature was negatively associated with

pneumonia

1). Minimum temperature:

RR:1.12; 2). Maximum

temperature: RR: 0.92

Paynter et al. 2013

Bohol Province, Philippines, 2000 to 2004, children under three years

Time-series; Poisson regression

Case-crossover; Conditional logistic

regression

Weekly mean temperature

Hospitalizations for pneumonia

No significant relationship between

temperature and hospitalizations for

pneumonia was found

RR: 0.93 (95% CI: 0.63 to 1.35)

aThese studies are ordered by the date of publication and the first author. Abbreviations: CI, confidence interval; OR, odds ratio; RR, relative risk.

Page 41: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 23

Table 2-2. Characteristics of studies about ambient temperature and childhood diarrhoea

Studya Location and time Research design and statistical analysis

Main temperature exposure variable(s) Outcome(s) Key findings Effect estimates

Checkley et al. 2000

Lima, Peru, January 1993 to November

1998, children under 10 years

Time-series; Generalized additive

model Mean temperature

Hospital admissions for

diarrhoea

Admissions for diarrhoea increased by 8% per 1ºC increase in mean ambient

temperature

RR:1.08

Hashizume et al. 2007

Dhaka, Bangladesh, January 1996-

December 2002, children aged under

15 years

Time-series; Poisson generalized

linear model

Daily maximum and minimum

temperature

Weekly hospital visits for non-

cholera diarrhoea

Percentage change in the number of non-cholera

diarrhoeal cases per week for 1ºC increase in

temperature at lag 0–4 weeks was only

statistically significant in children ≤ 14 years old

Percent change: 5.7%; (95% CI: 2.9% – 8.6%)

Lam 2007

Sydney, Australia, January 2001 and December 2002,

children under six years

Time-series; ARIMA

Daily maximum and minimum

temperature

Hospital emergency

department visits for gastroenteritis

Hospital emergency department visits for

gastroenteritis increased by 11% per 1ºC increase in maximum temperature

RR:1.11 (P=0.007)

Page 42: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

24 Chapter 2: Literature Review

Table 2-2. Characteristics of studies about ambient temperature and childhood diarrhoea (continued) Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Chou et al. 2010

Taiwan, 1996-2007, children aged under

15 years

Time-series; Poisson regression

model

Monthly maximum temperature

Hospital admissions for

diarrhoea

Hospital admissions for diarrhoea among children aged 0–14 years increased by 4% per 1 °C increase in maximum temperature

RR:1.04 (P=0.012)

Green et al. 2010

Nine California counties, the USA, May to September

1999–2005, children aged under five years

Case-crossover; Conditional

logistic regression,

meta-analysis

Daily mean apparent temperature

Hospital admissions for

intestinal infectious diseases

The highest effects of ambient temperature on

intestinal infectious disease were seen in

children aged 5-18 years

Percent change: 21.3% (95% CI: 5.2% –39.8%)

Onozuka and

Hashizume 2011

Fukuoka, Japan, 2000-2008, children aged under 15 years

Time-series; Negative binomial

regression

Daily mean temperature

Hospital admission for

infectious gastroenteritis

Among children aged under 15 years, every 1

°C increase in temperature below 13 °C

was associated with a 23.2% infectious

gastroenteritis increase, while every 1 °C increase in temperature above 13

°C was associated with an 11.8% infectious

gastroenteritis decrease

1). Hot: Percent change:11.8% (95% CI:

6.6% – 17.3%)

2). Cold: Percent change: 23.2% (95% CI:16.6%–

30.2%)

Page 43: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 25

Table 2-2. Characteristics of studies about ambient temperature and childhood diarrhoea (continued) Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Bandyopadhyay et al. 2012

14 Sub-Saharan African countries,

1992-2001, children aged under three

years

Time-series; Random effect

model

Monthly average maximum and

minimum temperature

Diarrhoea cases from a survey

An increase in monthly average maximum

temperature raises the prevalence of diarrhoea

while an increase in monthly minimum

temperature reduces diarrhoea in children

under three years of age

1). Maximum temperature: Coefficient:

1.013 (P<0.01); 2). Minimum

temperature: Coefficient: -0.475 (P<0.01)

aThese studies are ordered by the date of publication and the first author. Abbreviations: ARIMA, autoregressive integrated moving average; CI, confidence interval; RR, relative risk.

Page 44: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

26 Chapter 2: Literature Review

Table 2-3. Characteristics of studies about rainfall and childhood pneumonia Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Ebi et al. 2001

Six California counties, January

1983 to June 1998, children aged 0-17

years

Time-series; Poisson generalized estimating

equations model Daily rainfall

Hospitalizations for viral

pneumonia in females

There was no significant relationship between

rainfall and hospitalizations for viral

pneumonia

RR: 1.5 (95% CI: -1.8 to

4.8)

Paynter et al. 2013

Bohol Province, Philippines, July 2000

to December 2004, children aged under

three years

1). Time-series; Distributed lag

Poisson regression 2). Case-crossover; Conditional logistic

regression

Weekly rainfall Hospitalizations for pneumonia

No significant relationship between

rainy days and hospitalizations for

pneumonia was found

RR:2.37 (95% CI: 0.85 to 6.60)

aThese studies are ordered by the date of publication and the first author. Abbreviations: CI, confidence interval; RR, relative risk.

Page 45: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 27

Table 2-4. Characteristics of studies about rainfall and childhood diarrhoea Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Hashizume et al. 2007

Dhaka, Bangladesh, January 1996-

December 2002, children aged under

15 years

Time-series; Poisson generalized

linear model

Daily maximum and minimum

temperature

Weekly hospital visits for non-

cholera diarrhea

The number of non-

cholera diarrhoea cases per week increased both

above and below the threshold of 52mm of

average rainfall over lags 0–8 weeks.

1). Above the rainfall

threshold: Percent change: 5.1% (95% CI:

3.3%–6.8%) 2). Below the rainfall

threshold: Percent change: 3.9% (95% CI:

0.6%–7.2%)

Lam 2007

Sydney, Australia, January 2001 and December 2002,

children aged under six years

Time-series; ARIMA Daily rainfall

Hospital emergency

department visits for gastroenteritis

No significant relationship was found between rainfall and hospital emergency department visits for

gastroenteritis

RR:0.98 (P=0.09)

Andrade et al. 2009

Rio Grande do Norte, Brazil, 1992 to 2001, children under 1 year

of age

Time-series; Distributed lag model Monthly rainfall Hospitalizations

for infant diarrhea

There was a positive relationship between

rainfall and infant diarrhea

RR:1.002 (P<0.001)

Drayna et al. 2010

Wauwatosa,

Wisconsin, the USA, January 2002 to December 2007, children (age not

given)

Time-series; ARMA Daily rainfall

Emergency department visits

for acute gastrointestinal

illness

Any rainfall 4 days prior

was significantly associated with an 11%

increase in acute gastrointestinal illness

visits.

RR: 1.11

Page 46: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

28 Chapter 2: Literature Review

aThese studies are ordered by the date of publication and the first author. Abbreviations: ARIMA, autoregressive integrated moving average; ARMA, autoregressive moving average; CI, confidence interval; RR, relative risk.

Page 47: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 29

Table 2-5. Characteristics of studies about relative humidity and childhood pneumonia Studya Location and time Research design and

statistical analysis Main temperature

exposure variable(s) Outcome(s) Key findings Effect estimates

Souza et al. 2012

Campo Grande,

Brazil, 2004 to 2008, children aged 5-14

years

Time-series: Poisson regression Daily humidity Outpatient visits

for pneumonia

No significant relationship between

humidity and outpatient visits for pneumonia was

found

RR: 1.00

Paynter et al. 2013

Bohol Province, Philippines, July 2000

to December 2004, children aged under

three years

Time-series; Distributed lag

Poisson regression Case-crossover;

Conditional logistic regression

Weekly relative humidity

Hospitalizations for pneumonia

No significant relationship between relative humidity and hospitalizations for

pneumonia was found

RR: 1.01 (95% CI: 0.63 to 1.35)

aThese studies are ordered by the date of publication and the first author. Abbreviations: CI, confidence interval; OR, odds ratio; RR, relative risk.

Page 48: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

30 Chapter 2: Literature Review

Table 2-6. Characteristics of studies about relative humidity and childhood diarrhoea

Studya Location and time Research design and statistical analysis

Main temperature exposure variable(s) Outcome(s) Key findings Effect estimates

D’Souza et al. 2008

Three cities of Australia, 1993 to

2003, children aged under five years

Time-series; Negative binomial

log-linear regression Weekly relative

humidity Hospital

admissions for rotavirus diarrhoea

Higher humidity in the previous week was

associated with a decrease in rotavirus diarrhoeal

admissions in three cities

1). Canberra: RR:0.98; (95% CI:0.97 to 0.99) 2). Brisbane: RR: 0.98; (95% CI: 0.97 to 0.99)

3). Melbourne: RR: 0.99; (95% CI: 0.99 to 1.00)

Onozuka and

Hashizume 2011

Fukuoka, Japan, 2000-2008, children aged

under 15 years

Time-series; Negative binomial

regression

Daily relative humidity

Hospital admission for

infectious gastroenteritis

The increase in cases per 1% drop in relative humidity

was 3.9%

Percent change: 3.9% (95% CI:2.8 to 5.0)

Lam 2007

Sydney, Australia, January 2001 and December 2002,

children aged under six years

Time-series; ARIMA

Daily relative humidity

Hospital emergency

department visits for gastroenteritis

No significant relationship was found between

relative humidity and hospital emergency department visits for

gastroenteritis

RR:1.01 (P=0.165)

aThese studies are ordered by the date of publication and the first author. Abbreviations: CI, confidence interval; RR, relative risk.

Page 49: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 31

2.4 Discussion

This review systematically examined the association between climatic factors and

childhood pneumonia and diarrhoea. Majority of prior studies have found significant

impact of temperature on childhood pneumonia and diarrhoea, although the

magnitude of temperature effect varies with regions. Additionally, the difference in

climatic effects between temperate and tropical/subtropical regions was observed.

This difference can be attributable to many factors. One of the fundamental reasons

is the different relative importance of climatic factors in affecting the pathological

agents of childhood pneumonia and diarrhoea between temperate and

tropical/subtropical regions (Haynes et al. 2013; Patel et al. 2013). The popular

opinion is that viruses causing childhood pneumonia or diarrhoea (e.g., RSV and

rotavirus) had a distinct seasonal peak in regions with temperate climates but was

year-round in the tropical/subtropical areas (Patel et al. 2013; Pica and Bouvier

2014). Some data suggested that diarrhoeal diseases caused by rotavirus increased

during cool and dry seasons (Brandt et al. 1982; Haffejee 1995). At regional level,

many factors may interact and explain the seasonality of childhood infectious

diseases, including climate, transmission patterns, host behaviour and susceptibility.

The impact of these factors may also be context specific, and normally no factor

alone can fully capture the complexity of seasonality of childhood infectious

diseases.

It is not easy to do a meta-analysis quantifying the temperature impact on childhood

diarrhoea because of the different study designs, various temperature indicators and

heterogeneous types of diarrhoeal cases researchers used. No significant effect of

rainfall on childhood pneumonia was found. Existing literature suggests there is a

significant (most likely non-linear) relationship between rainfall and childhood

Page 50: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

32 Chapter 2: Literature Review

diarrhoea. Also, little evidence regarding the association between relative humidity

and childhood pneumonia was found, while some studies imply that there might be a

reverse association between relative humidity and childhood diarrhoea (D'Souza et

al. 2008; Onozuka and Hashizume 2010).

Mechanisms underlying temperature impacts on childhood pneumonia and

diarrhoea

Temperature is the most likely climate driver of childhood pneumonia, in temperate

settings. The discrepancy in the results of studies regarding the temperature-

pneumonia relationship may be attributable to different factors, including the age of

children (Souza et al. 2012), geography, study design (Ebi et al. 2001; Green et al.

2010), the genotypes of pathogens, and the dominant pathogens etc., among which,

the different prevailing aetiological agents in the study populations may be the most

important factor. The existing body of knowledge suggests that low temperature is

associated with peaks of respiratory syncytial virus (RSV) (Yusuf et al. 2007) and

Streptococcus pneumoniae (Herrera-Lara et al. 2013; Watson et al. 2006), and high

temperature may increase the pneumonia caused by Mycoplasma pneumoniae

(Onozuka et al. 2009; Xu et al. 2011), Pneumocystis (Djawe et al. 2013), and

Legionella pneumophila (Herrera-Lara et al. 2013). Streptococcus pneumoniae,

Haemophilus influenzae type B, RSV and influenza virus are the most common

pathogens worldwide (Rudan et al. 2013), and thus it is likely that the increasing

global surface temperature may to some extent decrease burden of childhood

pneumonia, though discrepancy may exist between different regions.

Most studies looking at the impact of temperature on childhood diarrhoea reported a

linear relationship, with a constant increase for childhood diarrhoeal cases by one

unit increase or decrease in temperature (Bandyopadhyay et al. 2012; Checkley et al.

Page 51: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 33

2000; Chou et al. 2010; D'Souza et al. 2008; Hashizume et al. 2007; Lam 2007).

Temperature may impact the transmission of childhood diarrhoea mainly through

three pathways. Firstly, high temperature promotes the growth of bacteria

(Hashizume et al. 2007), and low temperature increases the replication and survival

of virus, e.g., rotavirus (D'Souza et al. 2008). Secondly, food chain, from food

preparation stage to production process (D'Souza et al. 2004), may be affected by

temperature. Studies in England and Wales reported that food poisoning occurred

more in high temperatures (Bentham and Langford 1995; Bentham and Langford,

2001), indicating that there might be more childhood diarrhoea caused by food

poisoning in this area as climate change continues. Thirdly, high and cold

temperatures may alter people’s hygiene behaviour. For example, in hot days, people

are more likely to have cold water without disinfection, which expose them more to

bacteria. Under climate change context, it is pivotal to explore the distribution of

aetiological agents of childhood diarrhoea globally and further to project the future

childhood diarrhoea burden variation attributable to climate change.

Mechanisms underlying rainfall impact on childhood pneumonia and diarrhoea

No study, so far, has found a significant relation between rainfall and childhood

pneumonia, even though in monsoon seasons, children are more likely to spend time

indoors, which may increase crowding and their exposure to biomass fuel smoke, as

well as decrease their sunlight exposure, possibly resulting in a higher risk of getting

pneumonia (Paynter et al. 2010). In adults, results regarding the impact of rainfall on

pneumonia are not convincing either (Murdoch and Jennings 2009; Yusuf et al.

2007). Both low and high rainfall is reported to be associated with increase in

childhood diarrhoea. The mechanisms explaining the impact of rainfall on diarrhoea

are complex. The popular opinion is: Low rainfall/drought may result in water

Page 52: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

34 Chapter 2: Literature Review

scarcity, leading to more use of unprotected water sources and reducing hygiene

practices, and high rainfall/extreme precipitation may flush faecal contaminants from

dwellings into water supplies (Jofre et al. 2010).

Mechanisms underlying relative humidity impact on childhood pneumonia and

diarrhoea

RSV infections were reported positively associated with relative humidity in China,

Indonesia, Malaysia, Mexico, and Singapore (Chan et al. 2002; Loh et al. 2011;

Omer et al. 2008; Tang et al. 2010; Yusuf et al. 2007). A lab-based study suggested

that the survival of airborne Mycoplasma pneumoniae was a function of temperature

and relative humidity, and temperature response was mediated by relative humidity

(Wright et al. 1969). However, to date, very few studies have formally quantified the

association between relative humidity and childhood pneumonia. In terms of

humidity-diarrhoea relation, existing knowledge suggests that low relative humidity

may increase childhood diarrhoeal cases. One explanation is that low humidity

facilitates the survival and replication of rotavirus (D'Souza et al. 2008). Recently,

some researchers argued that rotavirus can be aerosolized, and low relative humidity

may facilitate the development of virus-laden dust and droplet nuclei, increasing

aerial transport (Levy et al. 2009; Shaman and Kohn 2009; Shaman et al. 2011).

Mechanism underlying climate variability impact on rotavirus diarrhoea

Strong evidence suggests that rotavirus normally peaks in dry and cold conditions.

Rotavirus can retain its infectivity for several days in aqueous environments, and

waterborne spread has been implicated in a number of rotavirus outbreaks (Ansari et

al. 1991). Some researchers argued that rotavirus spreads through air, although the

rotavirus has not been isolated from the respiratory tract (Bishop 1996; Haffejee

1995). The mechanisms underlying why rotavirus favours dry and cold environment

Page 53: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 35

is currently unclear. It was postulated that relative drop in humidity and rainfall

combined with drying of soils in lower temperatures might increase the aerial

transport of dried, contaminated fecal material, and might also lead to increased

formation of dust, which could provide a substrate for the virus particles (Levy et al.

2009).

2.5 Knowledge Gaps

The impact of climate on childhood pneumonia and diarrhoea has become an

important public health issue. However, there are still some knowledge gaps. First,

existing studies assessing the impact of temperature on pneumonia or diarrhoea

basically used the temperature data collected from limited ground monitors, which

may result in measurement bias because temperature across one city is spatially

variable (Zhang et al. 2011). It is urgently needed to apply some advanced

approaches (e.g., satellite remote sensing) to assessing temperature effect on

childhood pneumonia and diarrhoea. Second, seldom studies have examined the

relationship between temperature variation (diurnal temperature range (DTR) and

temperature change between two neighbouring days) and childhood pneumonia and

diarrhoea, even though a sudden temperature change may pressure on children’s

immune system (Bull 1980). Third, previous studies looking at the effect of climate

on childhood pneumonia or diarrhoea mainly used hospital admissions or outpatient

visits as the outcome variable, and the results of these studies depend largely on the

distribution of major pathogens causing pneumonia or diarrhoea in hospitalized

children. It is essential to use more accurate (e.g., lab-confirmed) data to investigate

the association between climate and a variety of specific pathogens using solid

statistical approaches. Forth, most previous studies were conducted in the settings of

Page 54: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

36 Chapter 2: Literature Review

the same climate type (either in temperate or tropical), and few multi-city studies

have been done so far (D'Souza et al. 2004), even though it is widely accepted that

the major climate drivers of pneumonia and diarrhoea vary across different climate

regions. Future research may use the data in cities of different climates and explore

the variation of the association between climate and pneumonia or diarrhoea in

different climates, using a consistent statistical method. Fifth, as climate change

continues, burden of childhood pneumonia and diarrhoea may change accordingly.

However, no study has specifically project the future burden of childhood pneumonia

and diarrhoea under climate change scenarios. Finally, so far, little research attention

has been paid to developing climate change adaptation measures, and it is important

to develop cost-effective adaptation measures to relieve the burden of childhood

pneumonia and diarrhoea due to climate change.

2.6 Conclusions

Existing scientific literature suggests a very-likely association between temperature

and childhood pneumonia and diarrhoea. There also appears to be a non-linear

relationship between rainfall and childhood diarrhoea, with both very low or high

rainfall increasing diarrhoeal cases, and high relative humidity may decrease the

incidences of childhood diarrhoea. Limited evidence on the impacts of rainfall and

relative humidity on childhood pneumonia was offered by prior studies.

2.7 References

American Academy of Pediatrics Committee on Environmental Health. 2003. In:

Etzel RA (ed) Pediatric Environmental Health, 2nd edn. Elk Grove Village,

IL: American Academy of Pediatrics.

Page 55: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 37

Andrade IG, Queiroz JW, Cabral AP, Lieberman JA, Jeronimo SMB. 2009.

Improved sanitation and income are associated with decreased rates of

hospitalization for diarrhoea in Brazilian infants. Trans R Soc Trop Med Hyg

103(5):506-511.

Ansari S, Springthorpe V, Sattar S. 1991. Survival and vehicular spread of human

rotaviruses: possible relation to seasonality of outbreaks. Rev Infect Dis

13(3):448-461.

Bandyopadhyay S, Kanji S, Wang L. 2012. The impact of rainfall and temperature

variation on diarrheal prevalence in Sub-Saharan Africa. Appl Geogr 33:63-

72.

Bentham G, Langford I. 1995. Climate change and the incidence of food poisoning

in England and Wales. Int J Biometeorol 39(2):81-86.

Bentham G, Langford I. 2001. Environmental temperatures and the incidence of food

poisoning in England and Wales. Int J Biometeorol 45(1):22-26.

Bishop RF. 1996. Natural history of human rotavirus infection. In S. Chiba, M.

Estes, S. Nakata & C. Calisher (Eds.), Viral Gastroenteritis (Vol. 12, pp. 119-

128): Springer Vienna.

Brandt CD, Kim HW, Rodriguez WJ, Arrobio JO, Jeffries BC, Parrott RH. 1982.

Rotavirus gastroenteritis and weather. J Clin Microbiol 16(3):478-482.

Bull G. 1980. The weather and deaths from pneumonia. Lancet 1(8183):1405-1408.

Chan PWK, Chew FT, Tan TN, Chua KB, Hooi PS. 2002. Seasonal variation in

respiratory syncytial virus chest infection in the tropics. Pediatr Pulmonol

34(1):47-51.

Page 56: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

38 Chapter 2: Literature Review

Checkley W, Epstein L, Gilman R, Figueroa D, Cama R, Patz J, et al. 2000. Effect of

El Niño and ambient temperature on hospital admissions for diarrhoeal

diseases in Peruvian children. Lancet 355(9202):442-450.

Chou WC, Wu JL, Wang YC, Huang H, Sung FC, Chuang CY. 2010. Modeling the

impact of climate variability on diarrhea-associated diseases in Taiwan

(1996–2007). Sci Total Environ 409(1):43-51.

D'Souza RM, Becker N, Hall G, Moodie K. 2004. Does ambient temperature affect

foodborne disease? Epidemiology 15(1):86-92.

D'Souza, RM, Hall G, Becker NG. 2008. Climatic factors associated with

hospitalizations for rotavirus diarrhoea in children under 5 years of age.

Epidemiol Infect 136(1):56-64.

Djawe K, Levin L, Swartzman A, Fong S, Roth B, Subramanian A, et al. 2013.

Environmental risk factors for pneumocystis pneumonia hospitalizations in

HIV patients. Clin Infect Dis 56(1):74-81.

Drayna P, McLellan S, Simpson P, Li S, Gorelick M. 2010. Association between

rainfall and pediatric emergency department visits for acute gastrointestinal

illness. Environ Health Perspect 118(10):1439-1443.

Ebi K, Exuzides K, Lau E, Kelsh M, Barnston A. 2001. Association of normal

weather periods and El Niño events with hospitalization for viral pneumonia

in females: California, 1983-1998. Am J Public Health 91(8):1200-1208.

Garcia-Vidal C, Labori M, Viasus D, Simonetti A, Garcia-Somoza D, Dorca J, et al.

2013. Rainfall is a risk factor for sporadic cases of legionella pneumophila

pneumonia. PLoS ONE 8(4):e61036.

Page 57: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 39

Green R, Basu R, Malig B, Broadwin R, Kim J, Ostro B. 2010. The effect of

temperature on hospital admissions in nine California counties. Int J Public

Health 55(2):113-121.

Haffejee I. 1995. The epidemiology of rotavirus infections: a global perspective. J

Pediatr Gastroenterol Nutr 20(3):275-286.

Hashizume M, Armstrong B, Hajat S, Wagatsuma Y, Faruque AS, Hayashi T, et al.

2007. Association between climate variability and hospital visits for non-

cholera diarrhoea in Bangladesh: effects and vulnerable groups. Int J

Epidemiol 36(5):1030-1037.

Haynes AK, Manangan AP, Iwane MK, Sturm-Ramirez K, Homaira N, Brooks WA,

et al. 2013. Respiratory syncytial virus circulation in seven countries with

Global Disease Detection Regional Centers. J Infect Dis S3:S246-254.

Herrera-Lara S, Fernández-Fabrellas E, Cervera-Juan Á, Blanquer-Olivas R. 2013.

Do seasonal changes and climate influence the etiology of community

acquired pneumonia? Archivos de Bronconeumología (English Edition),

49(4), 140-145.

IPCC. 2007a. Climate change 2007: the physical science basis. Contribution of

Working Group I to the Fourth Assessment Report of the Intergovernmental

Panel on Climate Change. Cambridge Cambridge University Press.

IPCC. 2007b. Summary for policymakers. In: Climate change 2007: the physical

science basis. Contribution of Working Group I to the Fourth Assessment

Report of the Intergovernmental Panel on Climate Change. Cambridge

University Press, Cambridge. Cambridge: Cambridge University Press.

Jagai JS, Sarkar R, Castronovo D, Kattula D, McEntee J, Ward H, et al. 2012.

Seasonality of rotavirus in South Asia: A meta-analysis approach assessing

Page 58: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

40 Chapter 2: Literature Review

associations with temperature, precipitation, and vegetation Index. PLoS

ONE 7(5):e38168.

Jofre J, Blanch A, Lucena F. 2010. Water-borne infectious disease outbreaks

associated with water scarcity and rainfall events. In S. Sabater & D. Barceló

(Eds.), Water Scarcity in the Mediterranean (pp. 147-159): Springer Berlin

Heidelberg.

Lam LT. 2007. The association between climatic factors and childhood illnesses

presented to hospital emergency among young children. Int J Environ Health

Res 17(1):1-8.

Levy K, Hubbard AE, Eisenberg JN. 2009. Seasonality of rotavirus disease in the

tropics: a systematic review and meta-analysis. Int J Epidemiol 38(6):1487-

1496.

Loh TP, Lai FY, Tan ES, Thoon KC, Tee NWS, Cutter J, et al. 2011. Correlations

between clinical illness, respiratory virus infections and climate factors in a

tropical paediatric population. Epidemiol Infect 139(12):1884-1894.

Murdoch DR, Jennings LC. 2009. Association of respiratory virus activity and

environmental factors with the incidence of invasive pneumococcal disease. J

Infect 58(1):37-46.

Omer SB, Sutanto A, Sarwo H, Linehan M, Djelantik, IGG, Mercer D, et al. 2008.

Climatic, temporal, and geographic characteristics of respiratory syncytial

virus disease in a tropical island population. Epidemiol Infect 136(10):1319-

1327.

Onozuka D, Hashizume M. 2010. Effects of weather variability on infectious

gastroenteritis. Epidemiol Infect 138(2):236-243.

Page 59: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Literature Review 41

Onozuka D, Hashizume M, Hagihara A. 2009. Impact of weather factors on

Mycoplasma pneumoniae pneumonia. Thorax 64(6), 507-511.

Patel MM, Pitzer VE, Alonso WJ, Vera D, Lopman B, Tate J, et al. 2013. Global

seasonality of rotavirus disease. Pediatr Infect Dis J 32(4):e134-147.

Paynter S, Ware RS, Weinstein P, Williams G, Sly PD. 2010. Childhood pneumonia:

a neglected, climate-sensitive disease? Lancet 376(9755), 1804-1805.

Paynter S, Weinstein P, Ware RS, Lucero MG, Tallo V, Nohynek H, et al. 2013.

Sunshine, rainfall, humidity and child pneumonia in the tropics: time-series

analyses. Epidemiol Infect 141(6):1328-1336.

Pica N, Bouvier NM. 2014. Ambient temperature and respiratory virus infection.

Pediatr Infect Dis J 33(3):311-313.

Rudan I, O'Brien K, Nair H, Liu L, Theodoratou E, Qazi S, et al. 2013.

Epidemiology and etiology of childhood pneumonia in 2010: estimates of

incidence, severe morbidity, mortality, underlying risk factors and causative

pathogens for 192 countries. J Glob Health 3(1):10401.

Shaman J, Goldstein E, Lipsitch M. 2011. Absolute humidity and pandemic versus

epidemic influenza. Am J Epidemiol 173(2):127-135.

Shaman J, Kohn M. 2009. Absolute humidity modulates influenza survival,

transmission, and seasonality. Proc Natl Acad Sci U S A 106(9):3243-3248.

Souza AD, Fernandes W, Pavão H, Lastoria G, Edo AA. 2012. Potential impacts of

climate variability on respiratory morbidity in children, infants, and adults. J

Bras Pneumol 38(6):708-715.

Sumi a, Rajendran K, Ramamurthy T, Krishnan T, Nair GB, Harigane K, et al. 2013.

Effect of temperature, relative humidity and rainfall on rotavirus infections in

Kolkata, India. Epidemiol Infect 141(8):1652-1661.

Page 60: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

42 Chapter 2: Literature Review

Tang JW, Lai FYL, Wong F, Hon KLE. 2010. Incidence of common respiratory viral

infections related to climate factors in hospitalized children in Hong Kong.

Epidemiol Infect 138(2):226-235.

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Watson M, Gilmour R, Menzies R, Ferson M, McIntyre P, Network NS. 2006. The

association of respiratory viruses, temperature, and other climatic parameters

with the incidence of invasive pneumococcal disease in Sydney, Australia.

Clin Infect Dis 42(2): 211-215.

Wright DN, Bailey GD, Goldberg LJ. 1969. Effect of temperature on survival of

airborne Mycoplasma pneumoniae. J Bacteriol 99(2):491-495.

Xu YC, Zhu LJ, Xu D, Tao XF, Li SX, Tang LF, et al. 2011. Epidemiological

characteristics and meteorological factors of childhood Mycoplasma

pneumoniae pneumonia in Hangzhou. World J Pediatr 7(3):240-244.

Yusuf S, Piedimonte G, Auais A, Demmler G, Krishnan S, Van Caeseele P, et al.

2007. The relationship of meteorological conditions to the epidemic activity

of respiratory syncytial virus. Epidemiol Infect 135(7):1077-1090.

Zhang K, Oswald EM, Brown DG, Brines SJ, Gronlund CJ, White-Newsome JL, et

al. 2011. Geostatistical exploration of spatial variation of summertime

temperatures in the Detroit metropolitan region. Environ Res 111(8):1046-

1053.

Page 61: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 43

Chapter 3: Results paper one

The spatiotemporal patterns and geographic co-

distribution of childhood pneumonia and diarrhoea in

Queensland, Australia

Xu Z, Hu W, Tong S (2014). The geographic co-distribution and socio-ecological drivers of

childhood pneumonia and diarrhoea in Queensland, Australia. Epidemiology and Infection,

143(5):1096-104.

Page 62: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

44 Chapter 3: Results paper one

Abstract

This study aimed to explore the spatiotemporal patterns and geographic co-distribution of

childhood pneumonia and diarrhoea in Queensland. A seasonal decomposition analysis was

conducted to assess the long-term trend and seasonality of childhood pneumonia and

diarrhoea. A spatial analysis was conducted to explore the spatial patterns of childhood

pneumonia and diarrhoea. A cluster analysis was also used to identify the high-risk clusters

and geographic co-distribution of childhood pneumonia and diarrhoea. The results suggest a

distinct seasonality of childhood pneumonia and diarrhoea. Childhood pneumonia and

diarrhoea mainly distributed in northwest of Queensland. Mount Isa was the high-risk cluster

where childhood pneumonia and diarrhoea co-distributed. Future pneumonia and diarrhoea

prevention and control measures in Queensland should focus more on Mount Isa.

Page 63: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 45

3.1 Introduction

Pneumonia and diarrhoea are the leading causes of child mortality (Liu et al. 2012). In 2011,

two million children died before reaching their fifth birthday because of pneumonia and

diarrhoea worldwide (Walker et al. 2013). While the incidences of mortality due to

pneumonia and diarrhoea have been declining in some industrialized countries, they are still

an important source of morbidity in these regions (Podewils et al. 2004). In Western Pacific

region, the total episodes of pneumonia and diarrhoea in children under five years of age

were 256.3 million and 12.2 million in 2010, respectively (Walker et al. 2013).

Pneumonia and diarrhoea are largely preventable, and hence it is essential to identify the risk

factors and take targeted preventive measures (Bhutta et al. 2013). Existing studies have

confirmed some individual-level biological factors for pneumonia and diarrhoea, such as

underweight, stunting and zinc deficiency (Walker et al. 2013). Some of these poverty-related

risk factors, such as suboptimum breastfeeding, under-nutrition and zinc deficiency, are

shared by pneumonia and diarrhoea, and these overlapping risk factors may result in the

geographic co-distribution of pneumonia and diarrhoea (Walker et al. 2013).

Australia shoulders a considerable burden of childhood pneumonia and diarrhoea (Rudan et

al. 2013; Scallan et al. 2005). It is urgently needed to reveal the spatiotemporal patterns of

childhood pneumonia and diarrhoea in Australia. This study explored the spatiotemporal

patterns and geographic co-distribution of childhood pneumonia and diarrhoea in

Queensland, Australia.

3.2 Methods

Data collection

Queensland is located in the northeast of Australia. Its mean temperature of summer is 25 °C

and mean temperature of winter is 15 °C. There is a significant variation in mean annual

Page 64: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

46 Chapter 3: Results paper one

rainfall across Queensland, varying from less than 150 mm/year in the southwest to 4000

mm/year in the northern coast (Australian Bureau of Statistics 2010). Data on emergency

department visits (EDVs) by postcode from January 1st 2007 through 31st December 2011 in

Queensland were obtained from Queensland Health. These data were collected from

emergency departments of hospitals and rural emergency departments of most Queensland

public facilities through the Emergency Department Information System (EDIS) (Toloo, et

al., 2012).The anonymised EDV data were classified according to the International

Classification of Disease, 10th version (ICD–code10). In this study, we included EDVs with

the principle cause coded as pneumonia (ICD–10 codes: J12–J18) and diarrhoeal disease of

any cause (ICD–10 codes: A00–A03, A04, A05, A06.0–A06.3, A06.9, A07.0–A07.2, A07.9,

A08–A09) among children aged 0–14 years. Ethical approval was obtained from the Human

Research Ethics Committee of Queensland University of Technology (Australia) prior to the

data being collected (number: 1000001168). Patient information was de-identified and thus

no written informed consent was obtained.

Statistical analysis

We plotted the decomposed daily distributions of EDVs for childhood pneumonia and

diarrhoea using a time-series approach. The change of EDVs for childhood pneumonia and

diarrhoea from 2008-2009 to 2010-2011 was calculated using the following equation:

2010 2011 2008 2009( ) /i i iMc EDV EDV population− −= −

Where Mc represents the morbidity change, 2010 2011iEDV − represents the EDVs for childhood

pneumonia (diarrhoea) for postal area i during 2010-2011, 2008 2009iEDV − represents the EDVs

for childhood pneumonia (diarrhoea) for postal area i during 2008-2009, and

ipopulation refers to the population for postal area i.

Page 65: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 47

Time-series analysis was conducted using the R statistical environment, version 2.15.3.

Visual maps were created using ArcGIS version 9.3 (ESRI Inc., Redlands, CA, USA), and

spatial cluster analysis was conducted using SatScan version 9.1.

3.3 Results

Summary statistics

Table 3-1 presents the summary statistics of EDVs for childhood pneumonia and diarrhoea

by postcode in Queensland. The average counts of childhood pneumonia and diarrhoea were

43.7 and 135.8, respectively.

Table 3-1. Summary statistics for EDVs for childhood pneumonia and diarrhoea by postcode

in Queensland, Australia, during 2007-2011

Variables Mean SD Min Max

Pneumonia (cases) 43.7 79.5 0 739

Diarrhoea (cases) 135.8 247.7 0 1750

Page 66: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

48 Chapter 3: Results paper one

Spatial pattern

Figure 3-1 shows the spatial distribution of rates of EDVs for childhood pneumonia and

diarrhoea, illustrating that EDVs for both pneumonia were the highest in central west,

northwest and far north of Queensland, and the EDVs for childhood diarrhoea were the

highest in the northwest of Queensland (Mount Isa). Figure 3-2 and Figure 3-3 show the

spatial distribution of rates of EDVs for childhood pneumonia by age and gender. Figure 3-4

and Figure 3-5 show the spatial distribution of rates of EDVs for childhood diarrhoea by age

and gender. No significant differences between two age groups and genders were observed in

terms of both pneumonia and diarrhoea spatial patterns. Figure 3-6 illustrates the change in

EDVs for childhood pneumonia and diarrhoea from years 2008-2009 to 2010-2011,

indicating that EDVs for pneumonia and diarrhoea changed from northwest or southeast of

Queensland in the past couple of years.

Temporal pattern

Figure 3-7 shows the decomposed daily distributions of EDVs for childhood pneumonia and

diarrhoea, showing a distinct seasonal trend for the two diseases, especially for pneumonia.

This figure indicates that EDVs for childhood pneumonia in Queensland were more likely to

occur in cold season. The particularly great number of EDVs for childhood pneumonia in

2009 is because of the 2009 pandemic H1N1 influenza.

Geographical co-distribution

The cluster results in Figure 3-8 reveal that EDVs for childhood pneumonia and diarrhoea in

Queensland were co-distributed in Mount Isa.

Spatial patterns of climatic factors

As the coming chapters will talk about the effects of climatic factors on childhood pneumonia

and diarrhoea, we presented the spatial patterns of mean temperature and rainfall in

Queensland in Figure 3-9. The results show that mean temperature in northwest of

Page 67: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 49

Queensland was higher than other places. No significant spatial pattern of rainfall was

detected.

Page 68: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

50 Chapter 3: Results paper one

Figure 3-1. The spatial distribution of EDVs for childhood pneumonia and diarrhoea in Queensland, from 2007 to 2011

Pneumonia is on the left side, and diarrhoea is on the right side.

Page 69: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 51

Figure 3-2. The spatial distribution of EDVs for childhood pneumonia by age in Queensland, from 2007 to 2011

Page 70: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

52 Chapter 3: Results paper one

Figure 3-3. The spatial distribution of EDVs for childhood pneumonia by gender in Queensland, from 2007 to 2011

Page 71: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 53

Figure 3-4. The spatial distribution of EDVs for childhood diarrhoea by age in Queensland, from 2007 to 2011

Page 72: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

54 Chapter 3: Results paper one

Figure 3-5. The spatial distribution of EDVs for childhood diarrhoea by gender in Queensland, from 2007 to 2011

Page 73: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 55

Figure 3-6. The change of EDVs for childhood pneumonia and diarrhoea in Queensland, from 2008-2009 to 2010-2011

Pneumonia is on the left side, and diarrhoea is on the right side.

Page 74: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

56 Chapter 3: Results paper one

Figure 3-7. The daily distribution of EDVs for childhood pneumonia and diarrhoea in Queensland, from 2007 to 2011

Pneumonia is on the left side, and diarrhoea is on the right side.

Page 75: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 2: Results paper one 57

Figure 3-8. The spatial clusters of EDVs for childhood pneumonia and diarrhoea in Queensland, from 2007 to 2011

Pneumonia is on the left side, and diarrhoea is on the right side.

Page 76: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

58 Chapter 3: Results paper one

Figure 3-9. The spatial patterns of mean temperature and rainfall in Queensland, from 2007 to 2011

Page 77: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 3: Results paper one 59

3.4 Discussion

This study has yielded several notable findings. There was a strong seasonal trend in

EDVs for childhood pneumonia, with more cases occurring in cold season. Children

suffering pneumonia and diarrhoea who visited emergency departments in

Queensland from 2007 to 2011 were mainly from central west, northwest and far

north of Queensland. According to the cluster analysis results, Mount Isa was the

high risk area for both childhood pneumonia and diarrhoea. Interestingly, in recent

years, Mount Isa has been experiencing a substantial decrease in EDVs for childhood

pneumonia and diarrhoea, and EDVs for childhood pneumonia and diarrhoea were

moving from west to southeast of Queensland.

Several reasons, such as nutritional factors (Vitamin A or Zinc deficiency) (Black et

al.), poverty (Fonseca et al. 1996), and Indigenous status (Janu et al. 2014), may

contribute to the high EDVs for childhood pneumonia and diarrhoea in central west,

northwest and far north of Queensland. Mount Isa city, a major lead, zinc and copper

producer, is the largest emitter of sulphur dioxide, lead and some other metals in

Australia (National Pollutant Inventory 2010). It has been convincingly documented

that the blood lead level of children in Mount Isa, especially for those aged 1–4

years, is way higher than children from other regions of Australia (Queensland

Health, 2009), and the consequent life-long negative health and intellectual impacts

of lead exposure on children have also been extensively reported (Lanphear et al.

2005; Tong et al. 1998). In this study, we found that pneumonia and diarrhoea in

children were co-distributed in Mount Isa, highlighting that there might be some

common risk factors in this area. Exposure to air pollutants (e.g., sulphur dioxide)

emitted by mining could increase hospital admissions for childhood pneumonia

(Barnett et al. 2005). Mining also had a significant adverse effect on semi-arid

Page 78: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

60 Chapter 3: Results paper one

freshwater aquatic system in Mount Isa (Taylor et al. 2009). The densities of

bacterial indicators in remnant pools throughout Leichhardt River have exceeded the

acceptable guidelines, which might expose children to greater risk of diarrhoea. In

this study, we also found the risk areas for childhood pneumonia and diarrhoea

changed from northwest to southeast of Queensland, and the EDVs for childhood

pneumonia and diarrhoea in Mount Isa have been decreasing sharply (though still

high) in recent years, indicating that protective measures may have been taken to

prevent children from continuously adverse impacts of mining. Previous studies have

also proposed that the high morbidity of childhood pneumonia and diarrhoea in

Mount Isa may be attributable to prematurity and intrauterine growth restriction,

under-nutrition, exposure to cigarette smoke, aspiration of infected nasopharyngeal

secretions and social factors of overcrowded living conditions and poor hygiene

(Janu et al. 2014). Many of the children in Mount Isa are members of common

extended families, raising the question of the role of inherited polymorphisms in

genes involved in immunity or inflammation (Thiel et al. 2009).

To the best of our knowledge, this is the first study to explore the geographic co-

distribution of childhood pneumonia and diarrhoea. The results from this study,

especially the high risk areas of pneumonia and diarrhoea we identified, may have

important implications for future control and prevention for childhood pneumonia

and diarrhoea in Queensland. A major limitation of this study is that the disease data

we collected from emergency departments may underestimate the actual infected

population because only children with severe symptoms would go to emergency

departments for treatment. Data on EDVs by postcode is not ideal in doing spatial

analysis, and we are collecting data on EDVs by Statistical Local Areas (SLA) or

Local Government Areas (LGA) for our future studies.

Page 79: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 3: Results paper one 61

3.5 Conclusions

Childhood pneumonia and diarrhoea were predominantly distributed in northwest of

Queensland, and Mount Isa was the region where these two childhood diseases co-

distributed. In recent years, the high risk areas of these two childhood diseases have

been changing from northwest to southeast of Queensland.

3.6 References

Australian Bureau of Statistics. 2010. Census Data. Available

http://www.abs.gov.au/CDataOnline Retrieved September 24th, 2013

Barnett AG, Williams GM, Schwartz J, Neller AH, Best TL, Petroeschevsky AL, et

al. 2005. Air pollution and child respiratory health. Am J Respir Crit Care

Med 171(11):1272-1278.

Bhutta ZA, Das JK, Walker N, Rizvi A, Campbell H, Rudan I, et al. 2013.

Interventions to address deaths from childhood pneumonia and diarrhoea

equitably: what works and at what cost? Lancet 381(9875):1417-1429.

Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M et al. 2008.

Maternal and child undernutrition: global and regional exposures and health

consequences. Lancet 371(9608):243-260.

Fonseca W, Kirkwood BR, Victora CG, Fuchs SR, Flores JA, Misago C. 1996. Risk

factors for childhood pneumonia among the urban poor in Fortaleza, Brazil: a

case-control study. Bull World Health Organ 74(2): 199-208.

Janu EK, Annabattula BI, Kumariah S, Zajaczkowska M, Whitehall JS, Edwards MJ

et al. 2014. Paediatric hospitalisations for lower respiratory tract infections in

Mount Isa. Med J Aust 200(10): 591-594.

Page 80: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

62 Chapter 3: Results paper one

Lanphear B, Hornung R, Khoury J, Yolton K, Baghurst P, Bellinger D, et al. 2005.

Low-level environmental lead exposure and children's intellectual function:

an international pooled analysis. Environ Health Perspect 113(7):894-899.

Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE, et al. 2012. Global,

regional, and national causes of child mortality: an updated systematic

analysis for 2010 with time trends since 2000. Lancet 379(9832):2151-2161.

Podwils LJ, Mintz ED, Nataro JP, Parashar UD. 2004. Acute, infectious diarrhea

among children in developing countries. Semin Pediatr Infect Dis 15(3):155-

168.

National Pollutant Inventory (2010).

http://www.npi.gov.au/npidata/action/load/individual-facility-

detail/criteria/year/2009/browse-type/Company/regbusiness-

name/MOUNT%2BISA%2BMINES%2BLTD/jurisdiction-

facility/Q020MIM001. Retrieved October 1st 2013

Podewils LJ, Mintz ED, Nataro JP, Parashar UD. 2004. Acute, infectious diarrhea

among children in developing countries. Semin Pediatr Infect Dis 15(3):155-

168.

Queensland Health. 2009. Mount Isa community lead screening program 2006–07: a

report into the results of a blood-lead screening program of 1–4 year old

children in Mount Isa, Queensland.

http://www.health.qld.gov.au/ph/documents/tphn/mtisa_leadrpt.asp

Retrieved October 1st 2013

Rudan I, O'Brien K, Nair H, Liu L, Theodoratou E, Qazi S, et al. 2013.

Epidemiology and etiology of childhood pneumonia in 2010: estimates of

Page 81: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 3: Results paper one 63

incidence, severe morbidity, mortality, underlying risk factors and causative

pathogens for 192 countries. J Glob Health 3(1):10401.

Scallan E, Majowicz SE, Hall G, Banerjee A, Bowman CL, Daly L, et al. 2005.

Prevalence of diarrhoea in the community in Australia, Canada, Ireland, and

the United States. Int J Epidemiol 34(2):454-460.

Taylor MP, Mackay A, Kuypers T, Hudson-Edwards K. 2009. Mining and urban

impacts on semi-arid freshwater aquatic systems: the example of Mount Isa,

Queensland. J Environ Monit 11(5):977-986.

Thiel S, Kolev M, Degn S, Steffensen R, Hansen AG, Ruseva M, et al. 2009.

Polymorphisms in mannan-binding lectin (MBL)-associated serine protease 2

affect stability, binding to MBL, and enzymatic activity. J Immunol

182(5):2939-2947.

Toloo S, Rego J, FitzGerald G, Aitken P, Ting J, Quinn J, et al. 2012. Emergency

Health Services (EHS): Demand and Service Delivery Models. Queensland

University of Technology.

Tong S, Baghurst P, Sawyer MG, Burns J, McMichael AJ. 1998. Declining blood

lead levels and changes in cognitive function during childhood: The port pirie

cohort study. JAMA 280(22):1915-1919.

Walker CLF, Perin J, Katz J, Tielsch J, Black R. 2013. Diarrhea as a risk factor for

acute lower respiratory tract infections among young children in low income

settings. J Glob Health 3(1):10402.

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Page 82: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects
Page 83: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 65

Chapter 4: Results paper two

Impact of temperature on childhood pneumonia estimated

from satellite remote sensing

Xu Z, Liu Y, Ma Z, Li S, Hu W, Tong S (2014). Impact of temperature on childhood

pneumonia estimated from satellite remote sensing. Environmental Research, 132:334-341.

Page 84: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

66 Chapter 4: Results paper two

Abstract

The effect of temperature on childhood pneumonia in subtropical regions is largely unknown

so far. This study examined the impact of temperature on childhood pneumonia in Brisbane,

Australia. A quasi-Poisson generalized linear model combined with a distributed lag non-

linear model was used to quantify the main effect of temperature on emergency department

visits (EDVs) for childhood pneumonia in Brisbane from 2001 to 2010. The model residuals

were checked to identify added effects due to heat waves or cold spells. Both high and low

temperatures were associated with an increase in EDVs for childhood pneumonia. Children

aged 2–5 years (excluding 5 years), and female children were particularly vulnerable to the

impacts of heat and cold. Indigenous children were sensitive to heat. Heat waves and cold

spells had significant added effects on childhood pneumonia, and the magnitude of these

effects increased with intensity and duration. There were changes over time in both the main

and added effects of temperature on childhood pneumonia. Children, especially those female

and Indigenous, should be particularly protected from extreme temperatures. Future

development of early warning systems should take the change over time in the impact of

temperature on children’s health into account.

Page 85: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 67

4.1 Introduction

Climate change has been widely recognized as the biggest health threat in the 21st century

(McMichael 2013), and its possible impact on infectious disease has attracted public health

attention (Altizer et al. 2013). Children, particularly their respiratory system (Sheffield and

Landrigan 2011), are vulnerable to the adverse impact of climate change (McKie 2013).

Pneumonia, the leading killer of children, has been reported responsible for 1.3 million deaths

in children aged under five years in 2011 (Walker et al. 2013), and the global burden of

childhood pneumonia may continue to rise due to the Earth’s increasing average surface

temperature (Walker et al. 2013), though the true scale of the association between

temperature and childhood pneumonia is largely unknown.

Persistent extreme temperatures (i.e., heat waves and cold spells) occur across the globe and

heat waves are projected to become more frequent and intense in the future (Meehl and

Tebaldi 2004), posing a huge challenge to children’s well-being (Xu et al. 2014). Existing

literature indicates that the effects of persistent extreme temperatures on human health can be

attributable to the independent effects of daily ambient temperature (main effect) and of

persistent periods of heat and cold (added effect) (Anderson and Bell 2009; Hajat et al. 2006).

During periods of persistent extreme temperatures, children are more likely to stay indoors,

which may increase crowding and their exposure to biomass fuel smoke from cooking,

possibly resulting in a higher risk of getting pneumonia. However, to our best knowledge,

few data are available on the effects of heat waves or cold spells on childhood pneumonia,

and no study has examined whether heat waves or cold spells have an added effect on

childhood pneumonia.

Epidemiological studies examining the effect of temperature on health tend to use

temperature from one ground-monitoring site or the average from several ground-monitoring

sites, which might result in measurement bias or exposure misclassification, especially for

Page 86: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

68 Chapter 4: Results paper two

those areas without extensive monitoring sites, because temperature across one city is

spatially variable (Zhang et al. 2011), and temperatures in urban areas are normally higher

than those in rural areas because of the urban heat island (Laaidi et al. 2012). Satellite remote

sensing data can substantially supplement ground monitoring networks to quantify the effect

of exposure to environmental hazards on health (Wang et al. 2013). The fundamental bias

satellite remote sensing data reduces is exposure error reduction due to better coverage and

higher spatial resolution. If using weather station data, researchers would probably need to

draw a buffer and assign everybody's temperature exposure to the readings at this central

station. Satellite data, on the other hand, are gridded at a pretty high spatial resolution so that

the exposure estimates can be more accurate. In addition, land surface temperature is

different from air temperature in that it considers the impact of direct solar radiation and the

surface long-wave radiation, so someone will feel hotter under the sun than in the shade even

though the difference in air temperature between under the sun and in the shade is smaller,

and thus it can be strongly related to heat-related morbidity and mortality. Although satellite

remote sensing data have been successfully used to link the relationship between air pollution

and acute health outcomes (Evans et al. 2013; Wang et al. 2013), it has been scarcely applied

to assess the impact of temperature on human health (Estes et al. 2009).

This study used the data on satellite remote sensing temperature and emergency department

visits (EDVs) for childhood pneumonia in Brisbane, Australia, from 2001 to 2010 and aimed

to minimize the measurement bias and answer three research questions: i) What is the

relationship between temperature and EDVs for childhood pneumonia? ii) Is there any added

effect due to heat waves and cold spells? iii) Whether there is any significant change over

time in the effect of temperature on childhood pneumonia across the study period?

4.2 Methods

Page 87: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 69

Data collection

Health data

Brisbane is the capital city of Queensland, Australia. It has a subtropical climate and rarely

experiences very cold temperatures. The daily EDV data from January 1st 2001 to December

31st 2010 classified according to the International Classification of Diseases, 9th version

and10th version (ICD 9 and 10), were obtained from Queensland Health. These data were

originally collected from emergency departments of hospitals and rural emergency

departments of most Queensland public facilities. Specifically, the Brisbane data we used

were from Caboolture Hospital, Ipswich Hospital, Logan Hospital, Mater Children’s Public

Hospital, Princess Alexandra Hospital, Queen Elizabeth II Jubilee Hospital, Redcliffe

Hospital, Redland Hospital, Royal Brisbane and Women’s Hospital, Royal Children’s

Hospital and the Prince Charles Hospital. These hospitals cover the Greater Brisbane region

well. Those coded as pneumonia (ICD 9 codes: 480–486; ICD 10 codes: J12–J18) in children

aged 0–14 years were selected.

Ground-monitoring data

Daily weather data, including rainfall and relative humidity, were supplied by the Australian

Bureau of Meteorology. Data on air pollutants, including daily average particular matter ≤

10µm (PM10) (µg/m3), daily average nitrogen dioxide (NO2) (µg/m3) and daily average ozone

(O3) (ppb), were obtained from the Queensland Department of Environment and Heritage

Protection (former Queensland Environmental Protection Agency).

Satellite remote sensing temperature data

Land surface temperature (LST) is the mean radiative skin temperature of an area of land

resulting from the energy balance between solar heating and land-atmosphere cooling. LST is

more closely related to the physiological activities of leaves, soil moisture, and near-surface

meteorology. Therefore, it has stronger spatial heterogeneity imposed by landscape variations

Page 88: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

70 Chapter 4: Results paper two

than air temperature. The Moderate Resolution Imaging Spectroradiometer (MODIS)

instruments were launched into low Earth polar orbits aboard the Naitonal Aerospace and

Space Administration (NASA)'s Terra and Aqua satellites in 1999 and 2002,

respectively(Anderson et al. 2005; Kaufman et al. 1998). They cross the equator at around

10:30 a.m. and 1:30 p.m. local time, respectively. MODIS LST was retrieved based on a

split-window algorithm that corrects for atmospheric effects based on the differential

absorption in MODIS's two adjacent infrared bands (bands 31 and 32) (Wan and Dozier

1996). Version 5 MODIS LST data have been extensively validated globally, showing that

the accuracy of the MODIS LST product is better than 1 K in most cases (Wan 2008; Wan et

al. 2002). For the current study, Level 3 MODIS Land Surface Temperature data (MOD11B1

for Terra from 2001 to 2010 and MYD11B1 for Aqua from 2002 to 2010) at 6 km spatial

resolution were downloaded from NASA's Level 1 and Atmospheric Archive and

Distribution System (http://ladsweb.nascom.nasa.gov) (Figure 4-1). Each data file contains

both a day time (~10:30 am for Terra, ~1:30 pm for Aqua) and a night time (~10:30 pm for

Terra, and ~1:30 am for Aqua) LST measurement. These LST values retrieved from the two

satellites were averaged to get the daily mean temperature (satellite remote sensing

temperature). Similar approaches have been applied to merge the LST data from Terra and

Aqua in the United States (Crosson et al. 2012).

Page 89: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 71

Figure 4-1. The areas where satellite remote sensing temperature data were collected

Page 90: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

72 Chapter 4: Results paper two

Data analysis

Heat waves and Cold Spells

There is no consistent definition for heat waves or cold spells. We combined temperature

duration and intensity to define heat waves and cold spells: (1) the 1st and 5th percentiles of

daily mean temperature were defined as the cold threshold, and the 95th and 99th percentiles

of the daily mean temperature as the heat threshold; and (2) a minimum of 2–4 consecutive

days with temperatures below the cold threshold or above the heat threshold were required.

Stage I: Estimating the main temperature effects

A quasi-Poisson generalized linear regression model combined with a distributed lag non-

linear model (DLNM) was used to quantify the effect of temperature on EDVs for childhood

pneumonia (Xu et al. 2013b). A natural cubic spline with four degrees of freedom (df) was

used to capture a potentially non-linear temperature effect. A lag of 21 days was used to

quantify the lagged effect of temperature (Xu et al. 2013a). Rainfall and relative humidity

were controlled for by using a natural cubic spline with four df. NO2, PM10 and O3 were

controlled for using a linear function. Seasonal patterns and long-term trends were controlled

by using a natural cubic spline with six df per year of data. Day of week was controlled as a

categorical variable. Influenza epidemics and public holiday were also controlled for in the

model.

Yt ~ Poisson(μt)

Log (μt) = α + β1Tt,l + ns(RHt, 4) + ns(Rainfallt, 4) + β2PM10t + β3O3t + β4NO2t

+ns(Timet,6)+ β5Day of Weekt + β6Influenzat + β7Holidayt

Where t is the day of the observation; Yt is the observed daily EDVs for childhood pneumonia

on day t; α is the model intercept; Tt,l is a matrix obtained by applying the DLNM to

temperature; β1 is vector of coefficients for Tt,l, and l is the lag days; ns(RHt, 4) is a natural

cubic spline with four degree of freedom for relative humidity; ns(Rainfallt, 4) is a natural

Page 91: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 73

cubic spline with four degree of freedom for rainfall; PM10t, O3t, and NO2t are the

concentrations of PM10, O3, and NO2 on day t; ns(Timet,6) is a natural cubic spline with six

degrees of freedom for seasonality and long-term trend; Day of Weekt is the categorical

variable as EDVs varied with week days and weekends; Influenzat is the number of lab-

confirmed influenza cases on day t; Holidayt is a binary variable which is “1” if day t was a

holiday.

We checked the temperature–pneumonia plot and chose the temperature corresponding to the

lowest risk as the reference temperature. We quantified the relative risk of EDVs for

childhood pneumonia associated with high temperature (29.6°C, 99th centile of mean

temperature) relative to the reference temperature (chosen to be 23.0°C). Similarly, we

calculated the relative risk of EDVs for childhood pneumonia associated with low

temperature (9.8°C, 1st centile of mean temperature) relative to the reference temperature

(23.0°C). We specifically examined the association between temperature and EDVs for

childhood pneumonia for every five years (2001–2005, 2002–2006, 2003–2007, 2004–2008,

2005–2009 and 2006–2010) to test whether there was any change over time in this

association.

Stage II Examining the added effects of heat waves and cold spells

We used the residuals of stage I as the dependent variable of stage II model to quantify the

possible added effect of heat waves and cold spells, meaning that the main effect of

temperature has been removed in stage II (Xu et al. 2013a). We assumed a maximum lag of

21 days for examining the lagged effects of heat waves and cold spells. EDVs for childhood

pneumonia on days of heat waves and cold spells were compared with those non-extreme

temperature days.

Log (μt) = Log (μt1) +β1Ct,l + β2Ht,l t=1,2,….,n

Page 92: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

74 Chapter 4: Results paper two

Where Log (μt1) is the estimated EDVs for childhood pneumonia counts on day t from the

stage-I model; Ct,l is a matrix applying DLNM to cold spells; and Ht,l is a matrix applying

DLNM to heat waves.

All data analysis was conducted using R (V 2.15), and “dlnm” package was used to fit the

regression model. Sensitivity analysis was conducted by adjusting the dfs for temperature and

time to assess the robustness of model choices.

4.3 Results

Summary statistics

Table 4-1 shows the summary statistics of daily climatic variables, air pollutants, influenza

and EDVs for childhood pneumonia. The mean value of satellite remote sensing temperature

was 19.8 °C. There were 17,238 EDVs for childhood pneumonia, with a daily mean of 4.7

cases. Figure 4-2 plots the EDVs for childhood pneumonia (decomposed), weather variables,

and air pollutants, showing a strong seasonal pattern of EDVs for childhood pneumonia,

satellite remote sensing temperature, O3 and NO2.

Page 93: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 75

Table 4-1. Summary statistics for climatic variables, air pollutants and paediatric pneumonia in Brisbane, Australia, 2001–2010

Variables Mean SD Min

Percentile

Max

25 50 75

RS mean temperature (°C)* 19.8 4.6 6.5 16.2 19.9 23.3 32.6

Relative humidity (%) 65.0 15.0 13.0 56.0 65.0 75.0 100.0

Rainfall (mm) 2.2 8.3 0 0 0 0.4 149.0

O3 (ppb) 13.4 4.6 1.7 10.2 12.8 16.0 34.2

PM10 (µg/m3) 16.1 18.4 3.9 11.5 14.3 17.8 960.0

NO2 (µg/m3) 7.2 4.3 0 4.0 6.3 9.8 25.3

Influenza 1.6 2.1 0 0.6 1.3 2.0 28

Pneumonia 4.7 4.1 0 2 4 6 53

*RS: remote sensing

Page 94: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

76 Chapter 4: Results paper two

010

2030

4050

data

-20

24

6

seas

onal

34

56

78

trend

-10

010

2030

402002 2004 2006 2008 2010

rem

aind

er

time

Figure 4-2a. The decomposed distribution of EDVs for paediatric pneumonia in Brisbane,

from 2001 to 2010

Page 95: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 77

1015

2025

3035

rsm

eant

emp

050

100

150

rain

fall

2040

6080

100

2002 2004 2006 2008 2010

hum

idity

Time

Figure 4-2b. The daily distributions of climate variables in Brisbane, from 2001 to 2010

Page 96: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

78 Chapter 4: Results paper two

020

040

060

080

0

pm10

510

1520

2530

35

o30

510

1520

25

2002 2004 2006 2008 2010

no2

Time

Figure 4-2c. The daily distributions of air pollutants in Brisbane, from 2001 to 2010

Page 97: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 79

Table 4-2 indicates the Spearman correlation between weather variables, air pollutants and

EDVs for childhood pneumonia. Childhood pneumonia was negatively correlated with

temperature and relative humidity, but positively correlated with air pollutants. Figure 4-3

reveals the scatter plots of pneumonia and weather variables.

Table 4-2. Spearman’s correlation between daily weather variables, air pollutants and paediatric pneumonia in Brisbane, Australia, from 2001–2010

RS mean temperature†

Relative humidity Rainfall PM10 O3 NO2 Pneumonia

RS mean temperature† 1.00

Relative humidity -0.25* 1.00

Rainfall -0.03 0.38* 1.00

PM10 0.30* -0.29* -0.31* 1.00

O3 0.13* -0.28* -0.07* 0.31* 1.00

NO2 -0.55* 0.13* -0.14* 0.01 -0.10* 1.00

Pneumonia -0.32* 0.03 0.06* 0.07* 0.10* 0.28* 1.00

* P<0.01; † RS, remote sensing

Page 98: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

80 Chapter 4: Results paper two

pneumo

10 15 20 25 30 35 20 40 60 80 100

010

2030

4050

1015

2025

3035

rsmean

rainfall

050

100

150

0 10 20 30 40 50

2040

6080

100

0 50 100 150

humidity

Figure 4-3. The pairwise plot of paediatric pneumonia, mean temperature, rainfall and

relative humidity in Brisbane, from 2001 to 2010

Page 99: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 81

Effect of temperature on EDVs for childhood pneumonia

The overall effect of temperature on EDVs for childhood pneumonia is presented in Figure 4-

4. EDVs for childhood pneumonia increased in both low and high temperatures. The impacts

of temperature on age-, gender- and ethnicity-specific EDVs for childhood pneumonia are in

Table 4-3. Children aged 2–5 years and female children appeared particularly vulnerable to

the temperature effect. Interestingly, Indigenous children were more sensitive to the heat

effect, and non-Indigenous children were more vulnerable to the cold effect.

10 15 20 25 30 35

12

34

56

78

Remote sensing temperatu

RR

(ped

iatri

c pn

eum

onia

)

Figure 4-4. The overall effect of mean temperature on paediatric pneumonia in Brisbane,

from 2001 to 2010

Page 100: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

82 Chapter 4: Results paper two

Table 4-3. The cumulative effect of high and low temperatures on EDVs for paediatric pneumonia, with 99th percentile (29.6 °C) and 1st percentile (10.4 °C) of temperature relative to reference temperature (23°C)

Diseases Heat effect (Relative risk (95% CI)) Cold effect (Relative risk (95% CI))

Lag 0–1 Lag 0–13 Lag 0–21 Lag 0–1 Lag 0–13 Lag 0–21

All ages 1.07(0.92,1.26) 1.33(0.93,1.91) 1.72(1.07,2.76)* 0.89(0.77,1.03) 1.30(0.91,1.86) 2.76(1.71,4.47)*

(0,1) 0.95(0.62,1.44) 1.02(0.42,2.49) 0.99(0.31,3.18) 1.10(0.77,1.56) 1.08(0.43,2.69) 1.87(0.55,6.32)

[1,2) 0.94(0.69,1.29) 1.17(0.57,2.40) 1.47(0.58,3.72) 0.92(0.71,1.21) 1.33(0.72,2.81) 1.91(0.76,4.76)

[2,5) 1.08(0.85,1.37) 1.27(0.74,2.18) 2.04(1.01,4.11)* 0.78(0.63,1.01) 1.37(0.80,2.36) 3.03(1.47,6.26)

[5,14] 1.18(0.89,1.57) 1.42(0.72,2.78) 1.14(0.46,2.81) 1.00(0.76,1.31) 1.46(0.73,2.88) 1.76(0.83,3.71)

Male 0.96(0.76,1.21) 1.11(0.66,1.86) 1.35(0.69,2.64) 0.88(0.72,1.07) 1.11(0.67,1.82) 2.18(1.11,4.25)*

Female 1.17(0.96,1.42) 1.50(0.95,2.36) 2.02(1.11,3.66)* 0.90(0.75,1.08) 1.48(0.94,2.35) 3.36(1.81,6.24)*

Indigenous 1.13(0.58,2.21) 1.92(0.74,4.98) 4.20(1.03,8.16)* 0.80(0.45,1.41) 0.79(0.20,3.55) 1.46(0.23,9.25)

Non-indigenous 1.07(0.91,1.26) 1.25(0.87,1.80) 1.57(0.97,2.54) 0.90(0.78,1.04) 1.36(0.94,1.96) 2.89(1.77,4.71)*

*P-value<0.05

Page 101: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 83

The added effects of heat waves and cold spells

The daily excess EDVs for childhood pneumonia on heat wave days and cold spell days are

in Table 4-4. Using the heat wave definitions of two, three or four days with the temperature

over the 95th centile, we did not find any significant added effect of heat waves on EDVs for

childhood pneumonia. While, using the temperature over the 99th centile as the temperature

cut-off, we found there were significant added effects of heat waves on EDVs for childhood

pneumonia, and the EDVs due to added effect increased from three to seven when the heat

wave duration increased from two to three consecutive days. A significant increase in EDVs

for childhood pneumonia during cold spells was found while using the definition of four days

with the temperature over the 95th centile.

Page 102: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

84 Chapter 4: Results paper two

Table 4-4. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2001 to 2010

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 58 0 (-1,1) ≤5th 52 1 (-1,2)

≥99th 6 3 (1,5) ≤1st 2 -1 (-6,4)

≥3

≥95th 31 0 (-1,2) ≤5th 22 1 (-1,3)

≥99th 2 7 (1,13) ≤1st - - -

≥4

≥95th 15 0 (-1,2) ≤5th 12 3 (1,5)

≥99th - - - ≤1st - - -

Page 103: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 85

Change over time in the effect of temperature on childhood pneumonia

The change of temperature effect on EDVs for childhood pneumonia over time can be seen in

Figure 4-5. The effect of high temperature on EDVs for childhood pneumonia experienced a

decreasing trend, while low temperature impact on EDVs for childhood pneumonia

experienced an increasing trend.

Figure 4-5. The change over time in the temperature effect on childhood pneumonia

Left hand side: hot effect; right hand side: cold effect; p1= 2001-2005, p2=2002-2006, p3=2003-2007, p4=2004-2008, p5=2005-2009, p6=2006-2010.

Page 104: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

86 Chapter 4: Results paper two

Table 4-5 presents the added effect of heat waves and cold spells on EDVs for childhood

pneumonia during the six periods (2001–2005, 2002–2006, 2003–2007, 2004–2008, 2005–

2009 and 2006–2010), revealing that the added effect of heat waves and cold spells varied

greatly over time. The statistically significant added effect of cold spells on EDVs for

childhood pneumonia occurred in the last two periods (2005–2009 and 2006–2010).

Page 105: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 87

Table 4-5a. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2001 to 2005

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 26 1 (-1,2) ≤5th 22 0 (-1,1)

≥99th 4 2 (-3,6) ≤1st 1 -2 (-12,8)

≥3

≥95th 11 1 (-2,3) ≤5th 12 0 (-1,1)

≥99th 1 0 (-10,10) ≤1st - - -

≥4

≥95th 4 2 (-4,8) ≤5th 9 0 (-2,1)

≥99th - - - ≤1st - - -

Page 106: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

88 Chapter 4: Results paper two

Table 4-5b. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2002 to 2006

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 22 1 (-1,2) ≤5th 23 0 (-1,1)

≥99th 2 3* (1,5)* ≤1st 1 -4 (-14,6)

≥3

≥95th 9 1 (-1,3) ≤5th 12 0 (-2,1)

≥99th - - - ≤1st - - -

≥4

≥95th 4 2 (-2,5) ≤5th 9 -1 (-2,1)

≥99th - - - ≤1st - - -

*P<0.05

Page 107: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 89

Table 4-5c. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2003 to 2007

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 22 1 (-1,2) ≤5th 25 0 (-1,1)

≥99th 2 3* (1,5)* ≤1st 1 -3 (-12,7)

≥3

≥95th 9 1 (-1,3) ≤5th 10 0 (-2,1)

≥99th - - - ≤1st - - -

≥4

≥95th 4 1 (-3,5) ≤5th 6 0 (-2,1)

≥99th - - - ≤1st - - -

*P<0.05

Page 108: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

90 Chapter 4: Results paper two

Table 4-5d. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2004 to 2008

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 24 1 (-1,2) ≤5th 34 0 (-1,1)

≥99th 3 -1 (-6,4) ≤1st 1 -2 (-12,8)

≥3

≥95th 10 1 (-1,3) ≤5th 15 0 (-1,2)

≥99th 1 5* (1,10)* ≤1st - - -

≥4

≥95th 4 2 (-1,5) ≤5th 8 0 (-2,2)

≥99th - - - ≤1st - - -

*P<0.05

Page 109: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 91

Table 4-5e. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2005 to 2009

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 29 0 (-1,1) ≤5th 34 1 (0,2)

≥99th - - - ≤1st 1 0 (-12,11)

≥3

≥95th 14 0 (-1,1) ≤5th 14 2 (0,4)

≥99th - - - ≤1st - - -

≥4

≥95th 9 0 (-2,2) ≤5th 7 3* (1,5)*

≥99th - - - ≤1st - - -

*P<0.05

Page 110: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

92 Chapter 4: Results paper two

Table 4-5f. Paediatric pneumonia due to the added effect of heat waves and cold spells in Brisbane, Australia, from 2006 to 2010

Heat Waves Cold Spells

No. of Consecutive

Days Percentile Days Pneumonia 95% CI Percentile Days Pneumonia 95% CI

≥2

≥95th 33 0 (-1,1) ≤5th 27 2 (0,3)

≥99th 2 -4 (-12,4) ≤1st 1 3 (-9,14)

≥3

≥95th 19 0 (-1,2) ≤5th 7 0 (-2,2)

≥99th 1 -2 (-10,4) ≤1st - - -

≥4

≥95th 12 0 (-2,2) ≤5th 2 15* (9,21)*

≥99th - - - ≤1st - - -

*P<0.05

Page 111: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 4: Results paper two 93

4.4 Discussion

This is the first study using satellite remote-sensing data to quantify the temperature-

pneumonia relationship and it has yielded several novel findings: i) Both low and

high temperatures were associated with an increase in childhood pneumonia; ii)

Children aged 2–5 years and female children were more vulnerable to temperature

effects on pneumonia, compared with children in other age groups and male children,

respectively. Indigenous children were more sensitive to the heat effect, compared

with non-Indigenous children; iii) Both heat waves and cold spells had added effects

on childhood pneumonia, and the magnitude of the added effects increased with

intensity and duration; iv) There was a decreasing trend in the high temperature

effect on childhood pneumonia, while the low temperature effect on childhood

pneumonia experienced an increasing trend. Meanwhile, the impact of heat waves

and cold spells on childhood pneumonia varied over time.

Previous studies looking at the impact of temperature on either mortality or

morbidity mainly rely on the data obtained from ground monitors (Basu and Samet

2002; Basu 2009; Ye et al. 2012), which may be not representative of the whole

population exposure (Kloog et al. 2013). Satellite remote sensing technology has

provided an unprecedented chance to increase the accuracy and precision of

environmental variable measurements (Goetz et al. 2000). As climate change

progresses, the global surface average temperature will increase, and cold-related

adverse impact on human well being may decrease accordingly (Xu et al. 2012). It is

pivotal to explore whether the decreasing cold-related impact can offset the

increasing heat-related impact, as climate change continues. Using the satellite

remote sensing data, we found the magnitude of the main effects of heat and cold

temperatures on childhood pneumonia was similar, suggesting that the increase in

Page 112: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

94 Chapter 4: Results paper two

heat-related pneumonia would be compensated by a reduction in cold-related

pneumonia, and hence EDVs for childhood pneumonia in Brisbane attributable to the

main effect of temperature may not increase sharply in the near future. However, as

globe warms, the high risk season may differ, and the pattern may change.

Surface temperatures in urban areas are usually higher than rural regions, which may

have an exacerbating effect during heat waves (Johnson et al. 2009). Using the

temperature in the urban areas to examine the effect of heat waves on morbidity or

mortality in populations living in both urban and rural locations may also cause

measurement bias (Zeger et al. 2000). We used satellite remote sensing temperature

to avoid this problem and found that there were significant added effects of heat

waves and cold spells on childhood pneumonia, which increased with intensity and

duration. In the future, more frequent, intense, and longer-lasting persistent extreme

temperatures will occur as climate change continues (Meehl and Tebaldi 2004),

especially in Australia (IPCC 2013), and therefore the burden of childhood

pneumonia due to heat waves and cold spells might increase accordingly, which

requires the government to develop effective strategies incorporating other child

protective health measures to mitigate and adapt to adverse impact of heat waves and

cold spells (Xu et al. 2014).

The significant effect of temperature on childhood pneumonia we observed in this

study is not in accord with some previous studies. For example, Paynter et al. have

looked at the relationship between temperature and clinical pneumonia cases in

children <3 years in Bohol Province, Philippines, but did not find significant

association between temperature and childhood pneumonia (Paynter et al. 2013).

Temperature effect on childhood pneumonia can largely be due to its impact on the

aetiological pathogens. Existing science suggests that low temperature is associated

Page 113: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

95

Chapter 4: Results paper two 95

with peaks of respiratory syncytial virus (RSV) (Yusuf et al., 2007), and

Streptococcus pneumoniae (Herrera-Lara et al. 2013; Watson et al. 2006), and high

temperature may increase the replication and survival of Mycoplasma pneumoniae

(Onozuka et al. 2009; Xu et al. 2011), Pneumocystis (Djawe et al. 2013) and

Legionella pneumophila (Herrera-Lara et al. 2013). The data we collected did not

include the information of lab-confirmed pneumonia pathogens, and thus we could

not separately analyse the relations between temperature and different aetiological

pathogens of childhood pneumonia.

Indigenous children have been found particularly vulnerable to high temperature in

this study, echoing to the findings that high temperatures substantially increased the

hospitalization risk for Indigenous Australians (Guo et al. 2013). Indigenous children

do not have adequate access to heat adaptation infrastructures, and they experience

more poverty than non-Indigenous children, which may render their great

vulnerability to heat (Ford 2012). The poor household infrastructure also adds their

risk of being exposed to extreme heat and cold (Bailie et al. 2010). Pneumonia in

children aged 2–5 years and female children were sensitive to both heat and cold,

which may be due to their anthropometry, body composition and social behaviour

(e.g., daily activity).

In this study, we also examined the changes in both main and added effects of

temperature on childhood pneumonia over time, and found that heat appeared to have

a decreasing impact on childhood pneumonia across a ten year study period, but cold

impact experienced an increasing trend, implying that children in Brisbane may have

gradually adapted to the heat effect while are still quite sensitive to cold effect. The

increasing use of air conditioning in Brisbane may contribute to the fact that heat

impact on childhood pneumonia declined in the past decade (Ostro et al. 2010), and

Page 114: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

96 Chapter 4: Results paper two

the increasing cold effect on childhood pneumonia across the study period may be

due to the fact that Brisbane children rarely experience cold days, and they may not

take precautionary initiatives before cold days come. The change over time in main

effect of temperature on childhood pneumonia has an important implication for early

warning systems for extreme temperatures, because the existing heat alert systems or

early warning systems (Díaz et al. 2006; Nicholls et al. 2008) typically are based on

the average risks of temperature over multiple years but have not taken the temporal

variation of temperature impact into account (Xu et al. 2014). The impacts of heat

waves and cold spells on childhood pneumonia also experienced great changes over

time in this study. Effect of heat waves occurred in the first couple of periods (2002–

2006, 2003–2007 and 2004–2008), and effect of cold spells happened in the last two

periods (2005–2009 and 2006–2010), indicating that parents and caregivers of

children, especially those with the history of pneumonia, should take precautionary

measures, particularly during cold spells in the future.

This study has several strengths. As the first study using the satellite remote sensing

technology to measure children’s temperature exposure, it greatly minimizes

measurement error. We assessed both the main and added effects of temperature on

childhood pneumonia, and found the change over time in the main and added effects,

which gives important implications for future childhood pneumonia prevention. The

similar magnitude of cold and heat main effects on childhood pneumonia we

observed in this study indicates that temperature-related burden of childhood

pneumonia in Brisbane may not change dramatically due to increasing temperature in

the future. Several limitations of this study should also be acknowledged. First, the

spatial resolution of 6km*6km may be too crude to reflect the exposure at individual

level. However, we will be trying to get higher spatial resolution (e.g., 3km*3km or

Page 115: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

97

Chapter 4: Results paper two 97

1km*1km) data and will use it in the future. Second, only one city is included in this

study, and thus it should be cautious to generalize our findings to other regions with

different climates. Third, people are mostly exposed to air temperature rather than

land surface temperature. However, we found that the correlation between air

temperature and surface temperature in Brisbane was very high (r>0.98), and thus we

think it is appropriate to use surface temperature as a temperature indicator to assess

its effects on childhood pneumonia.

4.5 Conclusions

Both high and low temperatures increased the risk of childhood pneumonia. As

climate change continues, persistent extreme temperatures increase, and children

with pneumonia history, especially those who are 2–5 years, female and Indigenous,

are at particular risk. Parents and caregivers should take precautionary measures to

protect children from being attacked by future frequent, intense and long-lasting

extreme temperatures. Policy makers should be aware of the temporal change in

temperature effect on children’s health while developing early warning systems.

4.6 References

Altizer S, Ostfeld RS, Johnson PTJ, Kutz S, Harvell CD. 2013. Climate change and

infectious diseases: From evidence to a predictive framework. Science

341(6145): 514-519.

Anderson B, Bell M. 2009. Weather-related mortality: how heat, cold, and heat

waves affect mortality in the United States. Epidemiology 20(2):205-213.

Page 116: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

98 Chapter 4: Results paper two

Anderson TL, Charlson RJ, Bellouin N, Boucher O, Chin M, Christopher SA, et al.

2005. An “A-Train” strategy for quantifying direct climate forcing by

anthropogenic aerosols. Bull Amer Meteor Soc 86(12):1795-1809.

Bailie R, Stevens M, McDonald E, Brewster D, Guthridge S. 2010. Exploring cross-

sectional associations between common childhood illness, housing and social

conditions in remote Australian Aboriginal communities. BMC Public Health

10(1):147.

Basu R. 2009. High ambient temperature and mortality: a review of epidemiologic

studies from 2001 to 2008. Environ Health 8(1):40.

Basu R, Samet JM. 2002. Relation between elevated ambient temperature and

mortality: A review of the epidemiologic evidence. Epidemiol Rev

24(2):190-202.

Crosson WL, Al-Hamdan MZ, Hemmings SNJ, Wade GM. 2012. A daily merged

MODIS Aqua–Terra land surface temperature data set for the conterminous

United States. Remote Sens Environ 119:315-324.

Díaz J, Linares C, Tobías A. 2006. A critical comment on heat wave response plans.

Eur J Public Health 16(6):600.

Djawe K, Levin L, Swartzman A, Fong S, Roth B, Subramanian A, et al. 2013.

Environmental risk factors for pneumocystis pneumonia hospitalizations in

HIV patients. Clin Infect Dis 56(1):74-81.

Estes M, Al-Hamdan M, Crosson W, Estes S, Quattrochi D, Kent S, et al. 2009. Use

of remotely sensed data to evaluate the relationship between living

environment and blood pressure. Environ Health Perspect 117(12):1832-

1838.

Page 117: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

99

Chapter 4: Results paper two 99

Evans J, van Donkelaar A, Martin RV, Burnett R, Rainham DG, Birkett NJ, et al.

2013. Estimates of global mortality attributable to particulate air pollution

using satellite imagery. Environ Res 120:33-42.

Ford JD. 2012. Indigenous health and climate change. Am J Public Health

102(7):1260-1266.

Goetz S, Prince S, Small J. 2000. Advances in satellite remote sensing of

environmental variables for epidemiological applications. Adv Parasitol

47:289-307.

Guo Y, Wang Z, Li S, Tong S, Barnett A. 2013. Temperature sensitivity in

indigenous Australians. Epidemiology 24(3):471-472.

Hajat S, Armstrong B, Baccini M, Biggeri A, Bisanti L, Russo A, et al. 2006. Impact

of high temperatures on mortality: Is there an added heat wave effect?

Epidemiology 17(6): 632-638.

Herrera-Lara S, Fernández-Fabrellas E, Cervera-Juan Á, Blanquer-Olivas R. 2013.

Do seasonal changes and climate influence the etiology of community

acquired pneumonia? Arch Bronconeumol 49(4):140-145.

IPCC. 2013. Summary for policymakers. In: Climate change 2013: the physical

science basis. Contribution of Working Group I to the Fifth Assessment

Report of the Intergovernmental Panel on Climate Change. Cambridge

University Press, Cambridge.

Johnson D, Wilson J, Luber G. 2009. Socioeconomic indicators of heat-related health

risk supplemented with remotely sensed data. Int J Health Geogr 8(1):57.

Kaufman YJ, Herring DD, Ranson KJ, Collatz GJ. 1998. Earth Observing System

AM1 mission to Earth. Geoscience and Remote Sensing, IEEE Transactions

on 36(4): 1045-1055.

Page 118: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

100 Chapter 4: Results paper two

Kloog I, Ridgway B, Koutrakis P, Coull B, Schwartz J. 2013. Long- and short-term

exposure to PM2.5 and mortality: using novel exposure models.

Epidemiology 24(4):555-561.

Laaidi K, Zeghnoun A, Dousset B, Bretin P, Vandentorren S, Giraudet E, et al. 2012.

The impact of heat islands on mortality in Paris during the August 2003 heat

wave. Environ Health Perspect 120(2):254-259.

McKie R. 2013. Children will suffer most as climate change increases in coming

decades, say scientists. BMJ 347:f5799.

McMichael AJ. 2013. Globalization, climate change, and human health. N Engl J

Med 368(14):1335-1343.

Meehl GA, Tebaldi C. 2004. More intense, more frequent, and longer lasting heat

waves in the 21st century. Science, 305(5686):994-997.

Nicholls N, Skinner C, Loughnan M, Tapper N. 2008. A simple heat alert system for

Melbourne, Australia. Int J Biometeorol 52(5):375-384.

Onozuka D, Hashizume M, Hagihara A. 2009. Impact of weather factors on

Mycoplasma pneumoniae pneumonia. Thorax 64(6):507-511.

Ostro B, Rauch S, Green R, Malig B, Basu R. 2010. The effects of temperature and

use of air conditioning on hospitalizations. Am J Epidemiol 172(9):1053-

1061.

Paynter S, Weinstein P, Ware RS, Lucero MG, Tallo V, Nohynek H, et al. 2013.

Sunshine, rainfall, humidity and child pneumonia in the tropics: time-series

analyses. Epidemiol Infect 141(06):1328-1336.

Sheffield P, Landrigan P. 2011. Global climate change and children's health: threats

and strategies for prevention. Environ Health Perspect 119(3):291-298.

Page 119: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

101

Chapter 4: Results paper two 101

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Wan Z. 2008. New refinements and validation of the MODIS Land-Surface

Temperature/Emissivity products. Remote Sens Environ 112(1):59-74.

Wan Z, Zhang Y, Zhang Q, Li Z. 2002. Validation of the land-surface temperature

products retrieved from Terra Moderate Resolution Imaging

Spectroradiometer data. Remote Sens Environ 83(1–2):163-180.

Wang Z, Liu Y, Hu M, Pan X, Shi J, Chen F, et al. 2013. Acute health impacts of

airborne particles estimated from satellite remote sensing. Environ Int

51:150-159.

Watson M, Gilmour R, Menzies R, Ferson M, McIntyre P, Network NS. 2006. The

association of respiratory viruses, temperature, and other climatic parameters

with the incidence of invasive pneumococcal disease in Sydney, Australia.

Clin Infect Dis 42(2): 211-215.

Xu YC, Zhu LJ, Xu D, Tao XF, Li SX, Tang LF, et al. (2011). Epidemiological

characteristics and meteorological factors of childhood Mycoplasma

pneumoniae pneumonia in Hangzhou. World J Pediatr 7(3):240-244.

Xu Z, Huang C, Hu W, Turner LR, Su H, Tong S. 2013a. Extreme temperatures and

emergency department admissions for childhood asthma in Brisbane,

Australia. Occup Environ Med 70(10):730-735.

Xu Z, Huang C, Su H, Turner L, Qiao Z, Tong S. 2013b. Diurnal temperature range

and childhood asthma: a time-series study. Environ Health 12(1):12.

Xu Z, Sheffield Pe, Su H, Wang X, Bi Y, Tong, S. 2014. The impact of heat waves

on children’s health: a systematic review. Int J Biometeorol 58(2):239-47

Page 120: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

102 Chapter 4: Results paper two

Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, et al. 2012. Climate change and

children’s health—A call for research on what works to protect children. Int J

Environ Res Public Health 9(9):3298-3316.

Ye X, Wolff R, Yu W, Vaneckova P, Pan X, Tong S. 2012. Ambient temperature

and morbidity: a review of epidemiological evidence. Environ Health

Perspect 120(1):19-28.

Yusuf S, Piedimonte G, Auais A, Demmler G, Krishnan S, Van Caeseele P, et al.

2007. The relationship of meteorological conditions to the epidemic activity

of respiratory syncytial virus. Epidemiol Infect 135(07):1077-1090.

Zeger S, Thomas D, Dominici F, Samet J, Schwartz J, Dockery D, et al. 2000.

Exposure measurement error in time-series studies of air pollution: concepts

and consequences. Environ Health Perspect 108(5):419-426.

Zhang K, Oswald EM, Brown DG, Brines SJ, Gronlund CJ, White-Newsome JL, et

al. 2011. Geostatistical exploration of spatial variation of summertime

temperatures in the Detroit metropolitan region. Environ Res 111(8):1046-

1053.

Wan Z, Dozier J. 1996. A generalized split-window algorithm for retrieving land-

surface temperature from space. Geoscience and Remote Sensing, IEEE

Transactions on 34(4), 892-905.

Page 121: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 5: Results paper three 103

Chapter 5: Results paper three

Assessment of the temperature effect on childhood

diarrhoea using satellite imagery

Xu Z, Liu Y, Ma Z, Toloo GS, Hu W, Tong S (2014). Assessment of the temperature effect

on childhood diarrhoea using satellite imagery. Scientific Reports, 4:5389.

Page 122: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

104 Chapter 5: Results paper three

Abstract

A quasi-Poisson generalized linear model combined with a distributed lag non-linear model

was used to quantify the main effect of temperature on emergency department visits (EDVs)

for childhood diarrhoea in Brisbane from 2001 to 2010. Residual of the model was checked

to examine whether there was an added effect due to heat waves. The change over time in

temperature-diarrhoea relation was also assessed. Both low and high temperatures had

significant impact on childhood diarrhoea. Heat waves had an added effect on childhood

diarrhoea, and this effect increased with intensity and duration of heat waves. There was a

decreasing trend in the main effect of heat on childhood diarrhoea in Brisbane across the

study period. Brisbane children appeared to have gradually adapted to mild heat, but they are

still very sensitive to persistent extreme heat. Development of future heat alert systems

should take the change in temperature-diarrhoea relation over time into account.

Page 123: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

105

Chapter 5: Results paper three 105

5.1 Introduction

Climate change has impacted and will increasingly influence human health, especially in the

context of rapid globalization (McMichael 2013). Children are particularly vulnerable to

climate change impact (Sheffield and Landrigan, 2011). They may experience greater risk of

infectious diseases (e.g., diarrhoea) as global surface average temperature increases

(Checkley et al. 2000).

Prior studies have well documented that heat waves may increase morbidity and mortality

(Gasparrini and Armstrong 2011; Ma et al. 2013). Some researchers have claimed that the

impact of heat waves on human health may be due to both the main effect of daily high

temperature and the added effect of persistent periods of heat (Anderson and Bell 2009;

Gasparrini and Armstrong 2011; Hajat et al. 2006). As climate change continues, there will

be more frequent, more intense and longer-lasting heat waves (Meehl and Tebaldi 2004).

Food chain, from food preparation stage to production process, may be affected by persistent

high temperatures, possibly resulting in more food-borne diseases (D'Souza et al. 2004).

Some studies have reported that food poisoning (Graham et al. 1995; Bentham and Langford

2001) and electrolyte imbalance (Knowlton et al. 2009) are more likely to occur during

periods of persistent hot temperatures. However, studies on the effect of heat waves on

childhood diarrhoea are scarce.

Prior studies looking at the impact of temperature on diarrhoeal diseases mainly used time-

series approach and obtained temperature data averaged from a network of sites (Checkley et

al. 2000; Hashizume et al. 2007), and the several monitoring sites are normally in or nearby

the urban areas (Xu et al. 2013b). This may render measurement bias because temperature

usually varies spatially across one city (Zhang et al. 2011) due to urban heat island (Laaidi et

al. 2012). Satellite-based monitoring data can largely solve this problem, given its broad

spatial coverage. Estes et al. have applied the remote sensing technology to examining the

Page 124: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

106 Chapter 5: Results paper three

effect of temperature on blood pressure (Estes et al. 2009). However, to date, no study has

used satellite remote sensing data to examine the relationship between temperature and

childhood diarrhoea.

This study used the data on satellite remote sensing temperature and attempted to address

three research issues: i) What is the relationship between temperature and emergency

department visits (EDVs) for childhood diarrhoea in Brisbane, Australia? ii) Is there any

added effect attributable to heat waves? iii) Is there any change over time in the effect of

temperature on childhood diarrhoea during the study period?

5.2 Methods

Data collection

Public hospital emergency departments are a significant and high-profile component of

Australia’s health care system (FitzGerald et al. 2012). EDVs data, which were classified

according to International Classification of Diseases, 9th and 10th versions (ICD-9 and ICD-

10), were supplied by Queensland Health. The details of the Brisbane data (selected hospitals

and covered regions, etc.) have been clarified in Chapter 4. We selected the following codes

for diarrhoea in children aged 0–14 years: ICD-9 codes: 001–003, 004, 005, 006.0–006.2,

007.0–007.5, 008–009; ICD–10 codes: A00–A03, A04, A05, A06.0–A06.3, A06.9, A07.0–

A07.2, A07.9, A08–A09. Existing evidence suggests that there is significant difference in the

seasonal variations of infection caused by various pathogens (Chui et al. 2011; Naumova et

al., 2007). Viral (008–009, and A08–A09), bacterial (001–003, 004, 005, and A00–A03, A04,

A05) and parasitic infections (006.0–006.2, 007.0–007.5 and A06.0–A06.3, A06.9, A07.0–

A07.2, A07.9) were separately analysed. Ethical approval was obtained from the Human

Research Ethics Committee of Queensland University of Technology (Australia) prior to the

data being collected. Patient information was de-identified and thus no written informed

Page 125: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

107

Chapter 5: Results paper three 107

consent was obtained. Data on rainfall and relative humidity were obtained from the

Australian Bureau of Meteorology. The data were collected from eight monitor stations

throughout Brisbane, and then averaged. Details on the collection of land surface temperature

have been discussed in the Methods section of Chapter four of this thesis.

Data analysis

To assess the main effect of temperature on EDVs for childhood diarrhoea, we used a quasi-

Poisson generalized linear model combined with a distributed lag non-linear model (DLNM)

(Xu et al. 2013). We used a “natural cubic spline–natural cubic spline” DLNM to examine

the temperature effect using four degrees of freedom (df) for both temperature and lag

dimensions, and selected a lag of 10 days to capture the possible lagged effect (Xu et al.

2013). Rainfall, relative humidity, long-term trend, seasonality, day of week and public

holiday were controlled for using the same approaches in Chapter four.

After all the other parameters were confirmed, we checked the temperature–diarrhoea plot

and chose the reference temperature by visual inspection. We calculated the relative risk of

EDVs for childhood diarrhoea associated with high temperature (29.6°C, 99th percentile of

mean temperature) and low temperature (10.4°C, 1st percentile of mean temperature) relative

to the reference temperature (chosen to be 16.0°C). To detect the change over time in the

association between temperature and diarrhoea, we specifically quantified the effect of

temperature on EDVs for childhood diarrhoea for a sliding window of five years (2001–2005,

2002–2006, 2003–2007, 2004–2008, 2005–2009 and 2006–2010). Further, the same

approach in Chapter four was used to detect whether there was any added effect of heat

waves on childhood diarrhoea. All data analysis was conducted using R environment

(Version 2.15). The sensitivity analysis was conducted by adjusting df for temperature and

time.

Page 126: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

108 Chapter 5: Results paper three

5.3 Results

Summary statistics

There were a total of 58166 EDVs for childhood diarrhoea during the study period. Table 5-1

presents the summary statistics of daily weather variables and EDVs for childhood diarrhoea

in the total children population and each subgroup. The mean value of satellite remote

sensing temperature was 19.8°C. The average values of relative humidity and rainfall were

65.0% and 2.2 mm, respectively. The mean value of daily EDVs for childhood diarrhoea was

15.9 (range=10–91), with the predominant pathogen being virus (mean=15.6). There were

very few EDVs for bacterial (mean=0.3) and parasitic (mean=0.04) diarrhoea every day.

Figure 5-1 shows the daily distributions of decomposed EDVs for childhood diarrhoea and

weather variables, illustrating a strong seasonal trend for diarrhoea and satellite remote

sensing temperature. The daily distributions of EDVs for viral, bacterial and parasitic

diarrhoea were presented in Figure 5-2.

Page 127: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

109

Chapter 5: Results paper three 109

Table 5-1. Summary statistics for climatic variables and paediatric diarrhoea in Brisbane, Australia, 2001–2010

Variables Mean SD Min Percentile

Max 25 50 75

RS mean temperature (°C)* 19.8 4.6 6.5 16.2 19.9 23.3 32.6

Relative humidity (%) 65.0 15.0 13.0 56.0 65.0 75.0 100.0

Rainfall (mm) 2.2 8.3 0 0 0 0.4 149.0

Diarrhoea 15.9 8.8 0 10 14 20 91

Viral diarrhoea 15.6 8.8 0 10 14 20 91

Bacterial diarrhoea 0.3 0.7 0 0 0 0 7

Parasitic diarrhoea 0.04 0.2 0 0 0 0 3

(0-1) 4.0 2.7 0 2 4 6 18

[1-2) 4.0 3.3 0 2 3 5 25

[2,5) 5.0 3.5 0 2 4 6 37

[5,14] 2.9 1.9 0 1 2 4 15

Male 7.3 4.6 0 4 7 9 47

Female 8.6 5.2 0 5 8 11 44

Indigenous 0.7 1.0 0 0 1 2 15

Non-indigenous 15.2 8.8 0 9 14 20 91

* RS: remote sensing

Page 128: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

110 Chapter 5: Results paper three

Figure 5-1. The daily distributions of EDVs for paediatric diarrhoea and climatic factors in Brisbane, from 2001 to 2010

The left side is the temporal distribution of diarrhoea, and the right side is the temporal distributions of remote sensing temperature, rainfall and

relative humidity.

Page 129: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

111

Chapter 5: Results paper three 111

Figure 5-2. The daily distribution of diarrhoea caused by different pathogens

From the top to the bottom: total, virus, bacteria and parasite.

Page 130: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

112 Chapter 5: Results paper three

The Spearman correlations between climate variables and EDVs for childhood diarrhoea are

presented in Table 5-2. EDVs for childhood diarrhoea were positively correlated with

satellite remote sensing temperature (r=0.04, P<0.01), and negatively correlated with relative

humidity (r=-0.11, P<0.01).

Table 5-2. Spearman’s correlation between daily weather conditions, air pollutants and paediatric diarrhoea in Brisbane, Australia, from 2001–2010

RS mean temperature†

Relative humidity Rainfall Diarrhoea Viral

diarrhoea Bacterial diarrhoea

Parasitic diarrhoea

RS mean temperature† 1.00

Relative humidity -0.25* 1.00

Rainfall -0.03 0.38* 1.00

Diarrhoea 0.04* -0.11* 0.01 1.00

Viral diarrhoea 0.02 -0.13* -0.01 0.99* 1.00

Bacterial diarrhoea 0.02 -0.04* -0.01 -0.03 -0.11* 1.00

Parasitic diarrhoea 0.01 -0.01 -0.02 0.05* 0.03 -0.03 1.00

* P<0.01; †RS, remote sensing

Page 131: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

113

Chapter 5: Results paper three 113

The effect of temperature on EDVs for childhood diarrhoea

Figure 5-3 reveals that both low and high temperatures were associated with increase in

EDVs for childhood diarrhoea. Table 5-3 quantitatively depicts the effects of temperature on

EDVs for childhood diarrhoea by pathogen, age, gender and Indigenous status. Due to the

very limited number of parasitic diarrhoea, only the results for viral and bacterial diarrhoea

were presented. No significant relationship between temperature and bacterial diarrhoea was

found. The relative risk (RR) of diarrhoea during hot days in children aged 1–2 years (not

including 2 years) was (RR: 1.17; 95% Confidence interval (CI): 1.10–1.25) greater than

children of other age groups, and the RR of diarrhoea during cold days in children aged 2–5

years (RR: 1.10; 95% CI: 1.03–1.18) was greater than other age groups. The effects of

extreme temperatures on male children and Indigenous children appeared to be higher than

female children and non-Indigenous children, respectively. Heat effect on EDVs for

childhood diarrhoea was acute, mainly occurring on the current day of exposure, and cold

effect happened after several days of exposure.

Page 132: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

114 Chapter 5: Results paper three

Figure 5-3. The overall effect of mean temperature on paediatric diarrhoea in Brisbane, from 2001 to 2010

Page 133: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

115

Chapter 5: Results paper three 115

Table 5-3. The cumulative effect of high and low temperatures on EDVs for paediatric diarrhoea in Brisbane, with 99th percentile (29.6 °C) and 1st (10.4°C) of temperature relative to reference temperature (16 °C)

Diseases Heat effect (Relative risk (95% CI)) Cold effect (Relative risk (95% CI))

Lag 0–1 Lag 0–7 Lag 0–10 Lag 0–1 Lag 0–7 Lag 0–10

Total 1.08(1.04,1.13)* 0.99(0.96,1.02) 1.01(0.96,1.06) 1.04(0.99,1.09) 1.02(0.98.1.05) 1.05(1.01,1.10)*

Viral diarrhoea 1.10(1.06,1.13)* 1.01(0.97,1.05) 0.99(0.93,1.05) 1.02(0.99,1.06) 1.01(0.98,1.05) 1.06(1.01,1.12)*

Bacterial diarrhoea 1.01(0.83,1.23) 0.85(0.68,1.07) 1.34(0.95,1.91) 1.12(0.92,1.37) 1.05(0.84,1.28) 0.87(0.61,1.24)

(0-1) 1.06(1.01,1.13)* 0.99(0.95,1.03) 0.98(00.90,1.06) 1.06(0.99,1.13) 1.01(0.97,1.05) 0.99(0.92,1.07)

[1-2) 1.17(1.10,1.25)* 0.98(0.94,1.03) 1.07(0.98,1.17) 1.03(0.97,1.09) 1.01(0.97.1.05) 1.06(0.98,1.04)

[2,5) 1.07(1.01,1.13)* 1.00(0.96,1.04) 1.01(0.93,1.10) 1.03(0.97,1.09) 1.02(0.98,1.06) 1.10(1.03,1.18)*

[5,14] 1.03(0.96,1.11) 0.99(0.94,1.04) 0.97(0.88,1.06) 1.07(0.98,1.17) 1.02(0.97,1.07) 1.09(1.01,1.18)*

Male 1.10(1.05,1.15)* 0.98(0.95,1.02) 1.01(0.95,1.08) 1.04(0.99,1.09) 1.02(0.98,1.05) 1.06(1.01,1.12)*

Female 1.07(1.02,1.12)* 0.99(0.96,1.03) 1.00(0.94,1.07) 1.03(0.99,1.08) 1.02(0.99,1.05) 1.05(0.99,1.11)

Indigenous 1.10(1.01,1.28)* 1.05(0.95,1.16) 1.12(0.92,1.36) 1.15(0.98,1.33) 1.01(0.92,1.11) 1.18(1.01,1.39)*

Non-indigenous 1.08(1.04,1.12)* 0.99(0.96,1.01) 1.00(0.95,1.06) 1.03(0.98,1.09) 1.02(0.99,1.04) 1.03(1.01,1.05)* *P-value<0.05

Page 134: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

116 Chapter 5: Results paper three

The added effect of heat waves

Table 5-4 shows the daily excess EDVs for childhood pneumonia on heat wave days

compared with non-heat wave days. We found no apparent added effect of heat waves on

EDVs for childhood diarrhoea while using the heat wave definitions of two or more

consecutive days with the temperature over the 95th percentile. However, we found

significant added effects of heat waves on EDVs for childhood diarrhoea at the temperature

threshold over the 99th percentile. Further, with heat wave days increasing from two to three

consecutive days, the number of EDVs for childhood diarrhoea due to added effect of heat

waves rose from three to seven.

Page 135: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

117

Chapter 5: Results paper three 117

Table 5-4. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2001 to 2010

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 58 0 (-1,1)

≥99th 6 3* (1,5)

≥3

≥95th 31 0 (-1,1)

≥99th 2 7* (2,13)

≥4

≥95th 15 0 (-1,2)

≥99th - - -

*P<0.05

Page 136: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

118 Chapter 5: Results paper three

Change over time in the effect of temperature on childhood diarrhoea

Figure 5-4 illustrates the change in the temperature effect of EDVs for childhood diarrhoea

over time. The heat effects increased slightly from Period 1 (2001-2005) to Period 2 (2002-

2006) and showed a decreasing trend thenceforward. No significant change over time in the

cold effect on EDVs for childhood diarrhoea was found.

Table 5-5 shows the added effects of heat waves on EDVs for childhood diarrhoea in the six

periods (2001–2005, 2002–2006, 2003–2007, 2004–2008, 2005–2009 and 2006–2010).

Statistically significant added effect of heat waves on EDVs for childhood diarrhoea was

observed only in the last period (2006–2010).

Figure 5-4. The change over time in the temperature effect on childhood diarrhoea

Left hand side: hot effect; right hand side: cold effect; p1= 2001-2005, p2=2002-2006,

p3=2003-2007, p4=2004-2008, p5=2005-2009, p6=2006-2010.

Page 137: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

119

Chapter 5: Results paper three 119

Table 5-5a. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2001 to 2005

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 26 1 (0,1)

≥99th 4 1 (-2,5)

≥3

≥95th 11 1 (-1,2)

≥99th 1 0 (-7,7)

≥4

≥95th 4 0 (-1,2)

≥99th - - -

Page 138: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

120 Chapter 5: Results paper three

Table 5-5b. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2002 to 2006

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 22 0 (-1,1)

≥99th 2 1 (-2,4)

≥3

≥95th 9 0 (-1,2)

≥99th - - -

≥4

≥95th 4 0 (-3,2)

≥99th - - -

Page 139: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

121

Chapter 5: Results paper three 121

Table 5-5c. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2003 to 2007

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 22 0 (-1,1)

≥99th 2 0 (-3,3)

≥3

≥95th 9 0 (-1,1)

≥99th - - -

≥4

≥95th 4 -1 (-2,1)

≥99th - - -

Page 140: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

122 Chapter 5: Results paper three

Table 5-5d. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2004 to 2008

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 24 0 (-1,1)

≥99th 3 0 (-4,4)

≥3

≥95th 10 -1 (-2,1)

≥99th 1 2 (-9,13)

≥4

≥95th 4 -1 (-3,1)

≥99th - - -

Page 141: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

123

Chapter 5: Results paper three 123

Table 5-5e. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2005 to 2009

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 29 0 (-1,1)

≥99th - - -

≥3

≥95th 14 5 (-8,18)

≥99th - - -

≥4

≥95th 9 -2 (-4,1)

≥99th - - -

Page 142: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

124 Chapter 5: Results paper three

Table 5-5f. Paediatric diarrhoea due to the added effect of heat waves in Brisbane, Australia, from 2006 to 2010

Heat Waves

No. of Consecutive Days Percentile Days Diarrhoea 95% CI

≥2

≥95th 33 0 (-1,1)

≥99th 2 7* (1,14)*

≥3

≥95th 19 0 (-1,1)

≥99th 1 13* (1,25)*

≥4

≥95th 12 0 (-2,2)

≥99th - - -

*P<0.05

Page 143: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 5: Results paper three 125

5.4 Discussion

The spatial variability of temperature across a city has been well documented in the

literature (Kestens et al. 2011). Existing studies quantifying the impact of

temperature on childhood diarrhoea many used data collected from ground monitors

in a city (Checkley et al. 2000). Due to the limited number of ground monitoring

sites, the temperature collected may not well represent the exposure of whole

population, possibly resulting in measurement bias in the effect estimates. In this

study, we used satellite remote sensing data to minimise this problem. This study

examined the effects of both high and low temperatures as well as heat waves on

childhood diarrhoea, while our previous work only examined the effects of

temperature variation on childhood diarrhoea (Xu et al. 2013). Both heat and cold

were associated with increase in EDVs for childhood diarrhoea in Brisbane. An

added effect of heat waves on childhood diarrhoea was found, though this effect

varied greatly across the study period. The effect of high temperature on childhood

diarrhoea showed a decreasing trend over time.

Both heat and cold have been found to be associated with increases in EDVs for

childhood diarrhoea in Brisbane, which may be partially explained by three reasons.

First, high temperature promotes the growth of bacteria (Hashizume et al. 2007),

while low temperature increases the replication and survival of virus, e.g., rotavirus

(D'Souza et al. 2008). Second, high temperature may impact the food chain, from

food preparation stage to production process (D'Souza et al. 2004), and expose

children more to contaminated food. Third, extremely low and high temperatures

may alter children’s hygiene behaviours (e.g., water drinking behaviour).

As climate change continues, global surface average temperature will increase, and

heat-related diarrhoea burden may increase accordingly, but cold related diarrhoea

Page 144: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

126 Chapter 5: Results paper three

burden may decrease, especially for viral diarrhoea which favours cold temperature

(Xu et al. 2012). Hence, it is essential to explore the balance between cold and heat

effects on childhood diarrhoea. In this study, we found cold effect on childhood

diarrhoea was greater than heat effect, which might be explained by the fact that in

industrialized countries, interventions to improve hygiene and sanitation may

decrease the occurrence of diarrhoea caused by bacteria and parasites, but for

rotavirus-related diarrhoea which is spread from person-to person, these

interventions may be less effective, and virus may be the dominant aetiological

pathogen in these regions (Olesen et al. 2005; Malek et al. 2006; Parashar et al.

2009). This finding implies that EDVs for childhood diarrhoea in Brisbane related to

the main effect of temperature may not increase greatly as climate change progresses.

In this study, we found Indigenous children were particularly vulnerable to the

impact of temperature on diarrhoea, which corresponds to the findings from a cohort

study reporting that Indigenous Australians were very sensitive to high and low

temperatures (Guo et al. 2013). Indigenous children require more public health

attention in Australia. They have restricted access to medical service and climate

change adaptation infrastructures, and high or cold temperature may trigger or

exacerbate their existing health problems. The poor housing conditions may also

render their greater vulnerability to heat or cold impact (Bailie et al. 2010). The RR

of diarrhoea in male children during extreme temperatures was greater than female

children, which might be partially due to their body composition (Maeda et al. 2005)

and behaviours (White-Newsome et al. 2011). Basu et al. argued that differences in

the effect of temperature on males and females varied among different locations and

populations (Basu and Samet 2002), and we believe that the differences of

temperature sensitivity between boys and girls may even vary with disease types.

Page 145: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

127

Chapter 5: Results paper three 127

An added effect of heat waves on childhood diarrhoea has been observed in this

study, and this effect increased with the intensity and duration of heat waves,

suggesting that the burden of childhood diarrhoea associated with heat waves may

increase as more frequent, intense and longer-lasting heat waves are projected to

occur in the future (Meehl and Tebaldi 2004; IPCC 2013). Parents and caregivers

should be educated and made aware of this risk and take precautionary measures to

protect their children during heat waves, and the government may also consider the

development of an heat early warning system as it will substantially decrease

children’s disease burden in heat waves (Xu et al. 2013). Interestingly, we found that

the effect of high temperature on childhood diarrhoea had a declining trend across

the study period, but the added effect of heat waves appeared to increase in recent

years. The decreasing trend in the main effect of heat on childhood diarrhoea may be

partially explained by the decreasing mean temperatures in the last four years (Figure

1). The finding also imply that children in Brisbane might be experiencing better

hygiene standards and/or have increasingly adapted to mild heat in recent years, but

persistent extremely hot days still pose a huge challenge to the health of their

intestinal system.

There are several strengths of this study. This is the first study to apply the satellite

remote sensing technology to quantifying the temperature-diarrhoea association,

which minimized the measurement bias. Our study examined the balance between

heat and cold effects on childhood diarrhoea and firstly reported the added effect of

heat waves on childhood diarrhoea. Two main weaknesses should also be

acknowledged. First, the ecological design restricts us to explore the possible

confounders (people’s drinking behaviour, etc.) and may cause ecological fallacy.

Page 146: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

128 Chapter 5: Results paper three

Second, we did not have the pathogen data and thus could not specifically analyse

the relation between temperature and diarrhoea caused by different pathogens.

In conclusion, both hot and cold temperatures were associated with childhood

diarrhoea, and male children and Indigenous children appeared to be at higher risk.

Heat waves had an added effect on childhood diarrhoea, which increased with

intensity and duration of heat waves. Parents, caregivers, schools and the government

should take action to enhance the children’s intestinal health particularly during

extreme temperatures and promote protective measures in advance.

5.5 References

Anderson B, Bell M. 2009. Weather-related mortality: how heat, cold, and heat

waves affect mortality in the United States. Epidemiology 20(2):205-213.

Bailie R, Stevens M, McDonald E, Brewster D, Guthridge S. 2010. Exploring cross-

sectional associations between common childhood illness, housing and social

conditions in remote Australian Aboriginal communities. BMC Public Health

10(1):147.

Basu R, Samet JM. 2002. Relation between elevated ambient temperature and

mortality: A review of the epidemiologic evidence. Epidemiol Rev

24(2):190-202.

Bentham G, Langford I. 1995. Climate change and the incidence of food poisoning

in England and Wales. Int J Biometeorol 39(2):81-86.

Bentham G, Langford I. 2001. Environmental temperatures and the incidence of food

poisoning in England and Wales. Int J Biometeorol 45(1):22-26.

Page 147: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

129

Chapter 5: Results paper three 129

Checkley W, Epstein L, Gilman R, Figueroa D, Cama R, Patz J, et al. 2000. Effect of

El Niño and ambient temperature on hospital admissions for diarrhoeal

diseases in Peruvian children. Lancet 355(9202):442-450.

Chui KKH, Jagai JS, Griffiths JK, Naumova EN. 2011. Hospitalization of the elderly

in the United States for nonspecific gastrointestinal diseases: A search for

etiological clues. Am J Public Health 101(11):2082-2086.

D'Souza RM, Becker N, Hall G, Moodie K. 2004. Does ambient temperature affect

foodborne disease? Epidemiology 15(1):86-92.

D'Souza RM, Hall G, Becker NG. 2008. Climatic factors associated with

hospitalizations for rotavirus diarrhoea in children under 5 years of age.

Epidemiol Infect 136(01):56-64.

Estes M, Al-Hamdan M, Crosson W, Estes S, Quattrochi D, Kent S, et al. 2009. Use

of remotely sensed data to evaluate the relationship between living

environment and blood pressure. Environ Health Perspect 117(12):1832-

1838.

FitzGerald G, Toloo S, Rego J, Ting J, Aitken P, Tippett V. 2012. Demand for public

hospital emergency department services in Australia: 2000–2001 to 2009–

2010. Emerg Med Australas 24(1):72-78.

Gasparrini A, Armstrong B. 2011. The impact of heat waves on mortality.

Epidemiology 22(1):68-73.

Guo Y, Wang Z, Li S, Tong S, Barnett A. 2013. Temperature sensitivity in

indigenous Australians. Epidemiology 24(3):471-472.

Hajat S, Armstrong B, Baccini M, Biggeri A, Bisanti L, Russo A, et al. 2006. Impact

of high temperatures on mortality: Is there an added heat wave effect?

Epidemiology 17(6): 632-638.

Page 148: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

130 Chapter 5: Results paper three

Hashizume M, Armstrong B, Hajat S, Wagatsuma Y, Faruque AS, Hayashi T, et al.

2007. Association between climate variability and hospital visits for non-

cholera diarrhoea in Bangladesh: effects and vulnerable groups. Int J

Epidemiol 36(5):1030-1037.

IPCC. 2013. Summary for policymakers. In: Climate change 2013: the physical

science basis. Contribution of Working Group I to the Fifth Assessment

Report of the Intergovernmental Panel on Climate Change. Cambridge

University Press, Cambridge.

Kestens Y, Brand A, Fournier M, Goudreau S, Kosatsky T, Maloley M, et al. 2011.

Modelling the variation of land surface temperature as determinant of risk of

heat-related health events. Int J Health Geogr 10(1):7.

Knowlton K, Rotkin-Ellman M, King G, Margolis H, Smith D, Solomon G, et al.

2009. The 2006 California heat wave: impacts on hospitalizations and

emergency department visits. Environ Health Perspect 117(1):61-67.

Laaidi K, Zeghnoun A, Dousset B, Bretin P, Vandentorren S, Giraudet E, et al. 2012.

The impact of heat islands on mortality in Paris during the August 2003 heat

wave. Environ Health Perspect 120(2):254-259.

Ma W, Yang C, Chu C, Li T, Tan J, Kan H. 2013. The impact of the 2008 cold spell

on mortality in Shanghai, China. Int J Biometeorol 57(1):179-184.

Maeda T, Sugawara A, Fukushima T, Higuchi S, Ishibashi K. 2005. Effects of

lifestyle, body composition, and physical fitness on cold tolerance in humans.

J Physiol Anthropol Appl Human Sci 24(4):439-443.

Malek MA, Curns AT, Holman RC, Fischer TK, Bresee JS, Glass RI, et al. 2006.

Diarrhea- and rotavirus-associated hospitalizations among children less than

5 years of age: United States, 1997 and 2000. Pediatrics 117(6):1887-1892.

Page 149: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

131

Chapter 5: Results paper three 131

McMichael AJ. 2013. Globalization, climate change, and human health. N Engl J

Med 368(14): 1335-1343.

Meehl GA, Tebaldi C. 2004. More intense, more frequent, and longer lasting heat

waves in the 21st century. Science 305(5686):994-997.

Naumova EN, Jagai JS, Matyas B, Demaria A, Macneill IB, Griffiths JK. 2007.

Seasonality in six enterically transmitted diseases and ambient temperature.

Epidemiol Infect 135(02):281-292.

Olesen B, Neimann J, Böttiger B, Ethelberg S, Schiellerup P, Jensen C, et al. 2005.

Etiology of diarrhea in young children in Denmark: a case-control study. J

Clin Microbiol 43(8):3636-3641.

Parashar UD, Burton A, Lanata C, Boschi-Pinto C, Shibuya K, Steele D, et al. 2009.

Global mortality associated with rotavirus disease among children in 2004. J

Infect Dis 200(S1):S9-S15.

Sheffield P, Landrigan P. 2011. Global climate change and children's health: threats

and strategies for prevention. Environ Health Perspect 119(3):291-298.

Tong S, Wang XY, Barnett AG. 2010. Assessment of heat-related health impacts in

Brisbane, Australia: Comparison of different heatwave definitions. PLoS

ONE 5(8):e12155.

White-Newsome JL, Sánchez BN, Parker EA, Dvonch JT, Zhang Z, O’Neill MS.

2011. Assessing heat-adaptive behaviors among older, urban-dwelling adults.

Maturitas 70(1):85-91.

Xu Z, Huang C, Turner LR, Su H, Qiao Z, Tong S. 2013b. Is diurnal temperature

range a risk factor for childhood diarrhea? PLoS ONE 8(5):e64713.

Xu Z, Sheffield P, Su H, Wang X, Bi Y, Tong S. 2014. The impact of heat waves on

children’s health: a systematic review. Int J Biometeorol 58(2):239-247.

Page 150: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

132 Chapter 5: Results paper three

Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, et al. 2012. Climate change and

children’s health—A call for research on what works to protect children. Int J

Environ Res Public Health 10;9(9):3298-316.

Zhang K, Oswald EM, Brown DG, Brines SJ, Gronlund CJ, White-Newsome JL, et

al. 2011. Geostatistical exploration of spatial variation of summertime

temperatures in the Detroit metropolitan region. Environ Res 111(8):1046-

1053.

Page 151: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 6: Results paper four 133

Chapter 6: Results paper four

Temperature variability and childhood pneumonia: an

ecological study

Zhiwei Xu, Wenbiao Hu, Shilu Tong

Citation: Xu Z, Hu W, Tong S (2014). Temperature variability and childhood

pneumonia: an ecological study. Environmental Health, 13(1):51.

Page 152: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

134 Chapter 6:Results paper four

Abstract

Background: Few data on the relationship between temperature variability and

childhood pneumonia are available. This study attempted to fill this knowledge gap.

Methods: A quasi-Poisson generalized linear regression model combined with a

distributed lag non-linear model was used to quantify the impacts of diurnal

temperature range (DTR) and temperature change between two neighbouring days

(TCN) on emergency department visits(EDVs) for childhood pneumonia in Brisbane,

from 2001 to 2010, after controlling for possible confounders.

Results: An adverse impact of TCN on EDVs for childhood pneumonia was

observed, especially in winter, and the magnitude of this impact increased greatly

from the first five years (2001–2005) to the second five years (2006–2010). Children

aged 5–14 years, female children and Indigenous children were particularly

vulnerable to TCN impact. However, there was no significant association between

DTR and EDVs for childhood pneumonia.

Conclusions: As climate change progresses, unstable weather pattern may occur,

and parents and caregivers of children should be aware of the high risk of pneumonia

posed by big TCN and take precautionary measures to protect children, especially

those with a history of respiratory diseases.

Page 153: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

135

Chapter 6:Results paper four 135

6.1 Introduction

Pneumonia is the top cause of mortality in children under five years (Walker et al.

2013). It is estimated that in 2010, worldwide, there were 120 million episodes of

pneumonia in children younger than five (Walker et al. 2013). Pneumonia is highly

preventable, and hence it is particularly important to explore the risk factors which

drive the incidence of pneumonia and further to prevent children from being exposed

to these risk factors.

Many nutritional, socioeconomic and environmental factors are involved in the

occurrence of pneumonia (Black et al. 2008; Fonseca et al. 1996; Paynter et al.

2010). The possible impact of climatic factors on pneumonia transmission has

attracted increasing research attention (Paynter et al. 2010; Paynter et al., 2013).

Both high and low temperatures have been reported to be associated with increased

pneumonia incidence (Ebi et al. 2001; Green et al. 2010). However, the potential

impact of temperature variability on childhood pneumonia has not been researched

yet, even though big temperature changes may stress on the respiratory system (Song

et al. 2008).

There are several ways to define temperature variability (Karl et al. 1995). For

example, the difference in daily maximum and minimum temperatures (i.e., diurnal

temperature range (DTR)) (Kan et al. 2007), and the mean temperature difference

from one day to the next (i.e., temperature change between two neighbouring days

(TCN)) (Guo et al. 2011; Lin et al. 2013). Previous studies have highlighted that big

DTR or TCN may pose a big threat to the respiratory system of human (Guo et al.

2011; Kan et al. 2007; Lin et al. 2013), especially for children (Xu et al. 2013c). We

hypothesized that great DTR or TCN might be associated with increase in childhood

Page 154: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

136 Chapter 6:Results paper four

pneumonia cases, and we used the data on emergency department visits (EDVs) for

childhood pneumonia in Brisbane from 2001 to 2010 to test our hypothesis.

6.2 Methods

Data collection

Data on EDVs from 1st January 2001 to 31st December 2010 classified according to

the International Classification of Diseases, 9th version and10th version (ICD 9 and

10) were supplied by Queensland Health. The details of the Brisbane data (selected

hospitals and covered regions, etc.) have been clarified in Chapter 4. We extracted

those cases coded as pneumonia (ICD 9 codes: 480–486; ICD 10 codes: J12–J18) in

children aged 0–14 years. Data on climate variables, including maximum and

minimum temperatures, rainfall and relative humidity, were obtained from Australian

Bureau of Meteorology. DTR was calculated as daily maximum temperature minus

daily minimum temperature (Kan et al. 2007). Daily mean temperature was the

average of daily maximum and minimum temperatures, and TCN was calculated as

mean temperature of the current day minus mean temperature of the previous day

(Guo, et al., 2011). Data on air pollutants, including daily average particular matter ≤

10 µm (PM10) (µg/m3), daily average nitrogen dioxide (NO2) (µg/m3) and daily

average ozone (O3) (ppb), were retrieved from the Queensland Department of

Environment and Heritage Protection.

Data analysis

Distributed lag non-linear model (DLNM) was developed to incorporate both lagged

and the non-linear effects of temperature on mortality or morbidity (Gasparrini et al.

2010). The detailed mechanism underlying DLNM has been introduced in (Xu et al.

2013d). A quasi-Poisson generalized linear regression combined with DLNM was

Page 155: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

137

Chapter 6:Results paper four 137

used to quantify the association between DTR (or TCN) and EDVs for childhood

pneumonia.

Yt ~ quasiPoisson(μt)

Log (μt) = α + βDTRt,l (βTCNt,l)+ns(Tt,l, 3)+ns(RHt, 3) + ns(PM10t, 3) + ns(O3t, 3)

+ ns(NO2t, 3) +ns(Timet,8)+ η1Holiday + η2Day of Weekt

Where t is the day of the observation; Yt is the observed daily childhood pneumonia

on day t; α is the model intercept; DTR t,l is a matrix obtained by applying DLNM to

DTR, TCN t,l is a matrix obtained by applying DLNM to TCN, Tt,l is a matrix

obtained by applying the DLNM to temperature; β is vector of coefficients for Tt,l,

and l is the lag days; ns(RHt, 3) is a natural cubic spline with three degrees of

freedom for relative humidity; ns(PM10t, 3) is a natural cubic spline with three

degrees of freedom for PM10; ns(O3t, 3) is a natural cubic spline with three degrees of

freedom for O3; ns(NO2t, 3) is a natural cubic spline with three degrees of freedom

for NO2; ns(Timet,8) is a natural cubic spline with eight degrees of freedom per year

for long-term trend and seasonality; Holiday is the public holiday, and Day of Weekt

is the categorical day of the week with a reference day of Sunday.

Previous studies have revealed that there might be a lagged effect of temperature

variability on human health, and the relationship between temperature variability and

respiratory diseases appears to be non-linear (Guo et al. 2011; Lin et al. 2013; Xu et

al. 2013c; Xu et al. 2013d). Thus, we used DLNM to incorporate the non-linear and

lagged effect (Gasparrini, et al., 2010). Specifically, DTR (or TCN) and lag were

incorporated using a “natural cubic spline–natural cubic spline” approach. The model

included lags up to 21 days for DTR (or TCN) and mean temperature (Xu et al.

2013a). For all other confounders (i.e., relative humidity, PM10, O3, and NO2), we

used lags up to 10 days.

Page 156: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

138 Chapter 6:Results paper four

All data analysis was conducted using the R statistical environment (v 2.15), and

“dlnm” package was used to fit the regression. In the sensitivity analysis, we changed

the df for DTR, TCN and time. We also excluded 2009 data as there was a big

pneumonia spike in 2009.

6.3 Results

Figure 6-1 shows the daily distributions of childhood pneumonia, mean temperature,

DTR and TCN, revealing that there was a seasonal trend in childhood pneumonia,

mean temperature, and DTR. The 2009 pneumonia peak (due to H1N1 flu pandemic)

is also revealed in this figure. Table 6-1 depicts the Spearman correlations between

climate variables, air pollutants and childhood pneumonia. There was a negative

correlation between DTR and mean temperature (r= -0.44, P<0.01). TCN was

positively correlated with mean temperature (r= 0.14, P<0.01). Further, DTR was

positively correlated with childhood pneumonia (r=0.21, P<0.01), while no

significant correlation between TCN and childhood pneumonia was observed.

Page 157: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

139

Chapter 6:Results paper four 139

010

2030

4050

pneu

mon

ia10

2030

tem

pera

ture

05

1020

30

DTR

-10

-50

5

2002 2004 2006 2008 2010

TCN

Time

Figure 6-1. The daily distributions of EDVs for paediatric pneumonia, mean

temperature, DTR and TCN in Brisbane, from 2001 to 2010

Page 158: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

140 Chapter 6:Results paper four

Table 6-1. Spearman’s correlation between daily weather conditions, air pollutants and paediatric pneumonia in Brisbane, Australia, from 2001–2010

Mean temperature DTR† TCN† Relative

humidity Rainfall PM10 O3 NO2 Pneumonia

Mean temperature 1.00

DTR† -0.44* 1.00

TCN† 0.14* 0.03 1.00

Relative humidity -0.06* -

0.33* 0.09* 1.00

Rainfall 0.16* -0.49* -0.03 0.38* 1.00

PM10 0.16* 0.31* 0.08* -0.29* -0.31* 1.00

O3 0.03 0.20* 0.03 -0.28* -0.07* 0.31* 1.00

NO2 -0.67* 0.42* -0.02 0.13* -0.14* 0.01 -0.10* 1.00

Pneumonia -0.37* 0.21* -0.01 0.03 0.06* 0.07* 0.10* 0.28* 1.00

* P<0.01; DTR, Diurnal temperature range; TCN, temperature change between two neighbouring days

Page 159: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

141

Chapter 6:Results paper four 141

Figure 6-2 shows the exposure-response relationship between temperature variability

and childhood pneumonia. No significant association between DTR and childhood

pneumonia was observed. While, negative TCN (< -2 °C), meaning a big

temperature decrease from one day to the next, increased the risk of childhood

pneumonia. There were more than 50 days every year with TCN below -2 °C

(|TCN|>2), with most days with temperature drop > 2 °C occurring in the second half

of each year (June to December) (Figure 6-3). Figure 6-4 shows the pattern of lagged

effect of TCN impact on childhood pneumonia, revealing that TCN effect lasted for

nearly three weeks. It can be seen from Figure 6-5 that children aged 5–14 years,

female children and Indigenous children appeared to be more vulnerable to the TCN

impact compared with children of other groups.

As there was a distinct seasonality in childhood pneumonia, with peak in winter, we

specifically analysed the TCN impact on childhood pneumonia in summer

(December, January, and February) and winter (June, July and August), and found

that this impact mainly occurred in winter (Figure 6-6).

To test whether there was a change over time in the effect of TCN on childhood

pneumonia, we splitted the ten years into two periods (2001–2005 and 2006–2010).

It can be seen from Figure 6-7 that the effect of TCN on childhood pneumonia during

period two was much greater than it was during period one.

Figure 6-8 shows the results excluding the 2009 data, indicating that the magnitude

of TCN effect on childhood pneumonia in Brisbane declined greatly without 2009

pneumonia spike, although the shape of the TCN-pneumonia relationship was

similar.

Page 160: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

142 Chapter 6:Results paper four

0 5 10 15 20 25 30

0.5

1.5

2.5

3.5

DTR (°C)

RR

(ped

iatri

c pn

eum

onia

)

-10 -5 0 5

05

1015

20

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

Figure 6-2. The overall effects of DTR and TCN on paediatric pneumonia in

Brisbane, from 2001 to 2010

Page 161: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

143

Chapter 6:Results paper four 143

Figure 6-3. Monthly average number of days with TCN < -2 °C (|TCN|>2)

Page 162: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

144 Chapter 6:Results paper four

0 5 10 15 20

1.00

1.05

1.10

1.15

1.20

TCN effect

Lag

RR

(ped

iatri

c pn

eum

onia

)

Figure 6-4. The lagged effect of TCN on childhood pneumonia

Page 163: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

145

Chapter 6:Results paper four 145

Figure 6-5. The effect of TCN on the total-, age-, gender- and ethnic-specific

childhood pneumonia in Brisbane, from 2001 to 2010 (Relative risk: The risk of EDVs for childhood pneumonia on days with temperature drop =5.7 °C relative to

days with temperature drop= 2.0 °C)

Page 164: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

146 Chapter 6:Results paper four

-10 -5 0 5

02

46

810

Summer

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

-10 -5 0 5

05

1015

20

Winter

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

Figure 6-6. The effects of TCN on childhood pneumonia in summer and winter

Page 165: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

147

Chapter 6:Results paper four 147

-5 0 5

02

46

810

2001-2005

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

-5 0 5

02

46

810

2006-2010

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

Figure 6-7. The overall effects of TCN on childhood pneumonia during two periods

Page 166: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

148 Chapter 6:Results paper four

-10 -5 0 5

02

46

810

TCN (°C)

RR

(ped

iatri

c pn

eum

onia

)

Figure 6-8. The overall effect of TCN on childhood pneumonia in Brisbane, from

2001 to 2010 (excluding 2009)

Page 167: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

149

Chapter 6:Results paper four 149

6.4 Discussion

This study has quantified the impacts of both DTR and TCN on childhood

pneumonia and yielded several notable findings. A big temperature decrease from

one day to the next (|TCN|>2) may increase the EDVs for childhood pneumonia, and

this effect lasted for around three weeks. Every year, children were at the risk of big

TCNs for more than 50 days, and big TCNs mainly occurred in the second half of

each year, especially in winter. Children aged 5–14 years, female children and

Indigenous children were at particular risk. Further, there was a change in the effect

of TCN on childhood pneumonia over time, and big TCN may contribute to the 2009

pneumonia spike. No significant relationship between DTR and childhood

pneumonia was observed.

As climate change progresses, not only the global surface average temperature, but

also the frequency of unstable weather patterns (e.g., sharp increase/decrease in

temperature) will increase, as projected (Hansen et al. 2012), which poses a

significant challenge to public health sectors. Children are particularly vulnerable to

both extreme temperatures (Xu et al. 2013b) and temperature variation (Xu et al.

2013c), due partially to their relatively less-developed thermoregulation capability

(Xu et al. 2012). In this study, we found a sharp temperature drop was followed by

significantly increased EDVs for childhood pneumonia, and the TCN impact lasted

for roughly three weeks. When big temperature change happens, children’s

temperature regulation system will firstly cope with the adverse impact. While,

temperature change which exceeds certain limits may trigger the symptoms of

existing respiratory diseases in a few days, and it may take another a few days from

the occurrence of symptoms to seeking for healthcare. The greater effect of TCN on

Page 168: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

150 Chapter 6:Results paper four

childhood pneumonia in winter than summer was also detected in this study, which

may be due to the greater number of extreme TCNs in winter.

With regards to prior studies looking at the impact of TCN on human health, some

studies have observed significantly increased respiratory-related mortality associated

with big TCN (Lin et al. 2013) while some others did not find significant results

(Guo et al. 2011). Our study stands out of previous studies by specifically focusing

on childhood pneumonia and controlling for a range of possible confounders. Most

previous studies assessing the impact of DTR on mortality or morbidity due to

respiratory diseases found significant effects of DTR (Cheng et al. 2014), while our

study did not find a significant relation between DTR and childhood pneumonia. The

disparity between findings of prior studies and our study may be explained by

different socioeconomic factors which may modify TCN effects on health (Malik et

al. 2012), different thermoregulatory capacity between children and adults (Xu et al.

2012), and different health seeking patterns between people of different regions.

Immunity plays an important role in respiratory infections (Curriero et al. 2002), and

infectious respiratory diseases are mainly caused by the infections of pathogens.

Weather change may affect humoral and cellular immunity (Bull 1980) as well as the

survival and replication of pathogens. From 2001 to 2010, every year, there were

more than 50 days with a great negative TCN, especially in winter, meaning that

parents and caregivers should take precautionary measures to protect their children,

especially those children with pre-existing respiratory conditions, if weather

forecasting reports a big temperature decrease in the coming couple of days.

In this study, we also found age, gender and Indigenous-status modified the

relationship between TCN and pneumonia. The school-aged children (5–14 years)

were more sensitive to TCN compared with children in other age groups, which

Page 169: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

151

Chapter 6:Results paper four 151

might be because they play outdoors more often and thus exposed more to the

outdoor temperature change. Another reason of school-aged children’s greater

vulnerability to big temperature change may be because school-aged children,

especially girls, are more likely to wear trendy clothes, and sometimes focus more on

the style than the thickness of those clothes. The difference in vulnerability to TCN

between two genders may be due to their body composition (Frascarolo et al. 1990),

though some researchers argued that differences in the effect of temperature on

males and females is variable among different locations and populations (Basu

2009). Our results also suggest that Indigenous children were more sensitive to TCN

effect compared with non-Indigenous children. Previous studies have reported that

the burden of pneumonia in Indigenous children is 10 to 20 fold higher than non-

Indigenous children, and they have longer hospital admissions and are more likely to

have multiple admissions with pneumonia (Burgner and Richmond 2005).

Indigenous children have limited access to infrastructures, and experience more

poverty than non-Indigenous children, possibly resulting in their greater vulnerability

to both extreme temperatures and temperature variation (Ford 2012).

We found that the effect of TCN on childhood pneumonia during 2006–2010 was

much greater than it was during 2001–2005, which may partially be attributable to

the TCN impact on 2009 pneumonia spike. This finding indicates that children might

be impacted more by sharp temperature decrease in the future if unstable weather

patterns occur as projected. Elucidating the impact of temperature variability on

children’s health is essential for the improvement of public health. The findings of

our study not only remind parents and children’s caregivers to take good care of

children in the days of big TCN, but also imply that government should take

temperature variability into account while developing the future early warning

Page 170: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

152 Chapter 6:Results paper four

systems because TCN may substantially impact children’s health, independent of

temperature.

This study has two major strengths. First, this is, to our knowledge, the first study to

look at the impact of DTR and TCN on childhood pneumonia. Second, the change

over time in the TCN effect on childhood pneumonia which we observed in this

study may encourage future studies to explore the temporal variability of TCN

impact on children’s health in a longer time period. Two weaknesses should be

acknowledged. First, this is a one city study, which means further generalization of

our findings to regions of different climates should be cautious. Second, the

aggregated data on temperature and EDVs for childhood pneumonia we used may

result in some biases in exposure and/or outcome measures.

6.5 Conclusions

No relationship between DTR and childhood pneumonia was observed. A sharp

temperature decrease from one day to the next had an adverse impact on childhood

pneumonia, especially in winter, and the magnitude of this impact increased in recent

years. Big TCNs may have contributed to the 2009 pneumonia peak in Brisbane.

Children aged 5–14 years, female children and Indigenous children are at particular

risk. As climate change continues, unstable weather patterns may be more frequent,

and the findings of this study have important implications for public health policy to

protect children from the impact of pneumonia.

6.6 References

Basu R. 2009. High ambient temperature and mortality: a review of epidemiologic

studies from 2001 to 2008. Environ Health 8(1):40.

Page 171: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

153

Chapter 6:Results paper four 153

Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. 2008.

Maternal and child undernutrition: global and regional exposures and health

consequences. Lancet 371(9608):243-260.

Bull G. 1980. The weather and deaths from pneumonia. Lancet 1(8183):1405-1408.

Burgner D, Richmond P. 2005. The burden of pneumonia in children: an Australian

perspective. Paediatr Respir Rev 6(2):94-100.

Cheng J, Xu Z, Zhu R, Wang X, Jin L, Song J, et al. 2014. Impact of diurnal

temperature range on human health: a systematic review. Int J Biometeorol,

doi: 10.1007/s00484-014-0797-5

Curriero FC, Heiner KS, Samet JM, Zeger SL, Strug L, Patz JA. 2002. Temperature

and mortality in 11 cities of the eastern United States. Am J Epidemiol

155(1):80-87.

Ebi K, Exuzides K, Lau E, Kelsh M, Barnston A. 2001. Association of normal

weather periods and El Niño events with hospitalization for viral pneumonia

in females: California, 1983-1998. Am J Public Health 91(8):1200-1208.

Fonseca W, Kirkwood B, Victora C, Fuchs S, Flores J, Misago C. 1996. Risk factors

for childhood pneumonia among the urban poor in Fortaleza, Brazil: a case--

control study. Bull World Health Organ 74(2):199-208.

Ford JD. 2012. Indigenous health and climate change. Am J Public Health

102(7):1260-1266.

Frascarolo P, Schutz Y, Jequier E. 1990. Decreased thermal conductance during the

luteal phase of the menstrual cycle in women. J Appl Physiol 69(6):2029-

2033.

Gasparrini A, Armstrong B, Kenward MG. 2010. Distributed lag non-linear models.

Stat Med 29(21):2224-2234.

Page 172: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

154 Chapter 6:Results paper four

Green R, Basu R, Malig B, Broadwin R, Kim J, Ostro B. 2010. The effect of

temperature on hospital admissions in nine California counties. Int J Public

Health 55(2):113-121.

Guo Y, Barnett AG, Yu W, Pan X, Ye X, Huang C, et al. 2011. A large change in

temperature between neighbouring days increases the risk of mortality. PLoS

ONE 6(2):e16511.

Hansen J, Sato M, Ruedy R. 2012. Perception of climate change. Proc Natl Acad Sci

U S A 109(37):E2415-23

Kan H, London SJ, Chen H, Song G, Chen G, Jiang L, et al. 2007. Diurnal

temperature range and daily mortality in Shanghai, China. Environ Res

103(3):424-431.

Karl TR, Knight RW, Plummer N. 1995. Trends in high-frequency climate

variability in the twentieth century. Nature 377(6546):217-220.

Lin H, Zhang Y, Xu Y, Xu X, Liu T, Luo Y, et al. 2013. Temperature changes

between neighboring days and mortality in summer: A distributed lag non-

linear time series analysis. PLoS ONE 8(6):e66403.

Malik S, Awan H, Khan N. 2012. Mapping vulnerability to climate change and its

repercussions on human health in Pakistan. Global Health 8(1):31.

Paynter S, Ware RS, Weinstein P, Williams G, Sly PD. 2010. Childhood pneumonia:

a neglected, climate-sensitive disease? Lancet 376(9755):1804-1805.

Paynter S, Weinstein P, Ware RS, Lucero MG, Tallo V, Nohynek H, et al. 2013.

Sunshine, rainfall, humidity and child pneumonia in the tropics: time-series

analyses. Epidemiol Infect 141(06):1328-1336.

Page 173: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

155

Chapter 6:Results paper four 155

Song G, Chen G, Jiang L, Zhang Y, Zhao N, Chen B, et al. 2008. Diurnal

temperature range as a novel risk factor for COPD death. Respirology

13(7):1066-1069.

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Xu Z, Etzel RA, Su H, Huang C, Guo Y, Tong S. 2012. Impact of ambient

temperature on children's health: A systematic review. Environ Res 117:120-

131.

Xu Z, Hu W, Su H, Turner LR, Ye X, Wang J, et al. 2013a. Extreme temperatures

and paediatric emergency department admissions. J Epidemiol Community

Health 68(4):304-311.

Xu Z, Huang C, Hu W, Turner LR, Su H, Tong S. 2013b. Extreme temperatures and

emergency department admissions for childhood asthma in Brisbane,

Australia. Occup Environ Med 70(10):730-735.

Xu Z, Huang C, Su H, Turner L, Qiao Z, Tong S. 2013c. Diurnal temperature range

and childhood asthma: a time-series study. Environ Health 12(1):12.

Xu Z, Huang C, Turner LR, Su H, Qiao Z, Tong S. 2013d. Is diurnal temperature

range a risk factor for childhood diarrhea? PLoS ONE 8(5):e64713.

Page 174: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects
Page 175: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

157

Chapter 7:Results paper five 157

Chapter 7: Results paper five

The impact of temperature variability on childhood

diarrhoea

Zhiwei Xu, Wenbiao Hu, Shilu Tong

Page 176: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

158 Chapter 7:Results paper five

Abstract

Background: The association between temperature and diarrhoea has been well

researched. However, the relationship between temperature variability and childhood

diarrhoea has not been unveiled.

Methods: A quasi-Poisson generalized linear model combined with a distributed lag

non-linear model (DLNM) was used to examine the effect of temperature variability

(Diurnal temperature range (DTR), and temperature change between two

neighbouring days (TCN)) on EDVs for childhood diarrhoea, controlling for

potential confounders.

Results: High DTR and TCN were significantly associated with an increase in EDVs

for childhood diarrhoea in Brisbane. Every year, from May to September, especially

July, children were at high risk posed by high DTR and low TCN. Male children

were particularly vulnerable to the adverse impacts of DTR and TCN on diarrhoea.

Conclusions: As climate change continues, unstable weather patterns may occur

more often, and it would be necessary to develop effective strategies to protect

children from being harmed by climate impacts.

Page 177: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

159

Chapter 7:Results paper five 159

7.1 Introduction

There is a wide spread consensus that climate is changing due to anthropogenic

activities, and climate change poses a huge threat to human well-being (McMichael

2013). Children are particularly vulnerable to the adverse impact of climate change

(Sheffield and Landrigan 2011). Paediatric infectious diseases (e.g., diarrhoea) may

increase due to the increasing global surface average temperature (Xu et al. 2012b).

The effect of high temperature on childhood diarrhoea has been well-documented

(Checkley et al. 2000; Green et al. 2010). However, the relationship between

temperature variability and childhood diarrhoea has not been unveiled yet, even

though the frequency of unstable weather patterns will increase as climate change

progresses (Epstein 2005), and big temperature variability may stress on immune

system (Bull 1980) and jeopardize children’s resistance to intestinal aetiological

agents.

Temperature variability can be defined in several ways. Diurnal temperature range

(DTR), temperature change from one day to the next (TCN), and the standard

deviation of temperature in a certain period of time, are the most frequently used

definitions (Karl et al. 1995). Herein, we adopted the first two temperature indexes

(i.e., DTR and TCN) as the indicators of temperature variability. We hypothesized

that big temperature variability may increase the cases of childhood diarrhoea, and

used data on emergency department visits (EDVs) in Brisbane, Australia, from 2001

to 2010 to testify our hypothesis.

7.2 Methods

Data collection

Page 178: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

160 Chapter 7:Results paper five

Data on EDVs were supplied by Queensland Health. The details of the data (selected

hospitals and covered regions, etc.) have been clarified in Chapter 4. We selected as

the following codes for diarrhoea in children aged 0–14 years: ICD-9 codes: 001–

003, 004, 005, 006.0–006.2, 007.0–007.5, 008–009; ICD–10 codes: A00–A03, A04,

A05, A06.0–A06.3, A06.9, A07.0–A07.2, A07.9, A08–A09. Data on daily maximum

temperature, minimum temperature, relative humidity, and rainfall were obtained

from Australian Bureau of Meteorology. DTR was calculated as daily maximum

temperature minus daily minimum temperature (Kan et al. 2007). We averaged the

daily maximum temperature and minimum temperature to get the daily mean

temperature, and TCN was calculated as mean temperature of the current day minus

mean temperature of the previous day (Guo et al. 2011).

Data analysis

A quasi-Poisson generalized linear model combined with a distributed lag non-linear

model (DLNM) was used to examine the effect of temperature variability

(DTR/TCN) on EDVs for childhood diarrhoea. A natural cubic spline with four

degrees of freedom (df) was used to capture a potentially non-linear effect of

DTR/TCN on childhood diarrhoea. Mean temperature was controlled for using a

“natural cubic spline–natural cubic spline” approach with lags up to 21 days (Xu

2013a). Rainfall and relative humidity were controlled for by using a natural cubic

spline with four df. Seasonal patterns and long-term trends were controlled by using

a natural cubic spline with seven df per year of data. Day of week was controlled for

as a categorical variable. Public holiday was also controlled for in the model.

There might be a lagged effect of temperature variability on human health, and we

used DLNM to incorporate the non-linear and lagged effect (Gasparrini et al. 2010).

Page 179: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

161

Chapter 7:Results paper five 161

Specifically, temperature variability (DTR/TCN) and lag were incorporated using a

“natural cubic spline–natural cubic spline” approach with lags up to 10 days.

All data analysis was conducted using the R statistical environment (v 2.15), and

“dlnm” package was used to fit the regression. Sensitivity analysis was conducted by

changing the df for DTR, TCN and time.

7.3 Results Table 7-1 presents the summary statistics of climate variables and EDVs for

childhood diarrhoea. The mean value of DTR was 14.7 °C (Range: 0.7–30.2 °C), and

the mean value of TCN was 0 °C (-9.5–6.8 °C). The daily average value of EDVs for

childhood diarrhoea was 15.9, with a range of 0 to 91. Figure 7-1 shows the daily

distribution of mean temperature, DTR, TCN and EDVs for childhood diarrhoea.

There was a distinct seasonal trend for childhood diarrhoea, mean temperature and

DTR.

Page 180: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 7: Results paper five 162

Table 7-1. Summary statistics for climatic variables, air pollutants and paediatric diarrhoea in Brisbane, Australia, 2001–2010

Variables Mean SD Min

Percentile

Max

25 50 75

Mean temperature

(°C) 20.1 4.9 6.9 16.2 20.4 23.8 34.9

DTR (°C)† 14.7 5.0 0.7 11.2 14.7 18.4 30.2

TCN (°C)† 0 2.0 -9.5 -1.0 0.2 1.4 6.8

Relative humidity

(%) 65.0 15.0 13.0 56.0 65.0 75.0 100.0

Rainfall (mm) 2.2 8.3 0 0 0 0.4 149.0

Diarrhoea 15.9 8.8 0 10.0 14.0 20.0 91.0

†DTR, Diurnal temperature range; TCN, temperature change between two neighbouring days

Page 181: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 7: Results paper five 163

020

4060

80

diar

rhea

1015

2025

3035

mea

ntem

p0

510

1520

2530

dtr

-10

-50

5

2002 2004 2006 2008 2010

tcn

Time

Figure 7-1. The daily distributions of EDVs for paediatric diarrhoea, mean temperature, DTR and TCN in Brisbane, from 2001 to 2010

Page 182: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

164 Chapter 7:Results paper five

Table 7-2 shows the Spearman correlations between climate variables and EDVs for

childhood diarrhoea, revealing that childhood diarrhoea was positively correlated

with both DTR and TCN. No correlation value was over 0.5, meaning that multi-

collinearity was not likely an issue in the subsequent modelling.

Table 7-2. Spearman’s correlation between daily weather conditions, air pollutants and paediatric diarrhoea in Brisbane, Australia, from 2001–2010

Mean temperature DTR TCN Relative

humidity Rainfall Diarrhoea

Mean temperature

1.00

DTR† -0.44* 1.00

TCN† 0.14* 0.03 1.00

Relative humidity

-0.06* -0.33* 0.09* 1.00

Rainfall 0.16* -0.49* -0.03 0.38* 1.00

Diarrhoea -0.02 0.09* 0.05* -0.11* 0.01 1.00

* P<0.01; DTR, Diurnal temperature range; TCN, temperature change between two neighbouring days

Page 183: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

165

Chapter 7:Results paper five 165

The cumulative effects of DTR/TCN on EDVs for childhood diarrhoea were shown

in Figure 7-2. DTR over 17 °C or TCN lower than -2 °C, was associated with a

significant increase in EDVs for childhood diarrhoea. Every year, there were more

than 50 days with TCN below -2 °C, and more than 120 days with DTR over 17 °C

during 2001 and 2010. Figure 7-3 reveals that days with both high DTR and TCN

occurred mainly from May to September (especially July).

The effects of DTR/TCN on EDVs for childhood diarrhoea by age, gender and

Indigenous status was presented in Figure 7-4. Male children appeared to be more

vulnerable to both DTR and TCN compared with female children.

Page 184: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

166 Chapter 7:Results paper five

0 5 10 15 20 25 30

0.5

1.0

1.5

2.0

DTR (°C)

RR

(ped

iatri

c di

arrh

ea)

-5 0 5

1.0

2.0

3.0

4.0

TCN (°C)

RR

(ped

iatri

c di

arrh

ea)

Figure 7-2. The overall effects of DTR and TCN on paediatric diarrhoea in Brisbane, from 2001 to 2010

Page 185: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

167

Chapter 7:Results paper five 167

1 2 3 4 5 6 7 8 9 10 11 12

DTR>17°C & TCN<-2°C

Month

Num

ber o

f day

s

05

1015

2025

30

Figure 7-3. The monthly distribution of days when DTR > 17 °C and TCN < -2 °C

Page 186: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

168 Chapter 7:Results paper five

Figure 7-4a. The effect of DTR on the total-, age-, gender- and ethnic-specific

childhood diarrhoea in Brisbane, from 2001 to 2010

Figure 7-4b. The effect of TCN on the total-, age-, gender- and ethnic-specific

childhood diarrhoea in Brisbane, from 2001 to 2010

Page 187: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

169

Chapter 7:Results paper five 169

7.4 Discussion As climate change progresses, not only the global surface average temperature, but

also the frequency of unstable weather patterns (e.g., sharp increase/decrease in

temperature) will increase, as projected (Epstein 2005), which poses a big challenge

to public health sectors. Children are particularly vulnerable to both extreme

temperatures (Xu et al. 2013b) and temperature variation (Xu et al. 2013c), due

partially to their relatively less-developed thermoregulation system (Xu et al. 2012a).

This study found that high DTR and TCN were significantly associated with an

increase in EDVs for childhood diarrhoea in Brisbane. Every year, from May to

September, especially July, children were at high risk posed by big DTR and TCN.

Male children were particularly vulnerable to the adverse impact of DTR and TCN

on diarrhoea.

The results of this study support our hypothesis that there was a significant

association between DTR and childhood diarrhoea. Currently, the exact mechanism

by which exposure to a large DTR can increase the risk of childhood morbidity

remains largely unknown. Bull argued that sudden changes in weather conditions

may affect either humoral or cellular immunity (Bull 1980). Very young children

have a relatively immature immune system (Gerba et al. 1996) and low self-care

capacity (Xu 2012a), which might result in a greater vulnerability to temperature

change. Studies have found that sudden changes in the temperature of inhaled air are

associated with the release of inflammatory mediators by mast cells (Graudenz et al.

2006; Togias et al. 1985), which could also be related to higher diarrhoea prevalence

(Feng et al. 2007; Ramsay et al. 2010). Further study is necessary to determine which

biomarkers are affected by DTR.

Page 188: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

170 Chapter 7:Results paper five

In this study, we also found a sharp temperature drop was followed by significantly

increased EDVs for childhood diarrhoea. With regards to prior studies looking at the

impact of TCN on human health, some studies have observed significantly increased

mortality associated with big TCN (Guo et al. 2011; Lin et al. 2013). In this study,

the TCN effect increased rapidly below -2 °C, highlighting that both parents and

medical staff should be made aware of the particularly high risk posed by large

temperature drop between two days and diarrhoea-related morbidity in children. In

Brisbane during the study period, children were exposed to the risk of both large

DTR and TCN from May to September, indicating that caregivers of children should

take precautionary measures in winter to protect children from being attacked by

large DTR and TCN.

This study is the first study to examine the relation between childhood diarrhoea and

TCN. Additionally, we used advanced methods to assess this association, which was

able to incorporate not only non-linear effects of temperature and DTR but also

lagged effects in the one model. Several limitations of this study should be

acknowledged. This is an ecological study, and therefore some bias due to exposure

misclassification may be inevitable. Only one city was examined, which might limit

the generalizability of our results. However, the findings of this research may

encourage a large scale, multi-centre study in the future.

7.5 Conclusions This study demonstrates a significant relationship between DTR/TCN and childhood

diarrhoea. Large DTR and TCN may trigger and increase childhood diarrhoea among

children. As climate change continues, unstable weather patterns may occur more

Page 189: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

171

Chapter 7:Results paper five 171

often, and it would be necessary to develop effective strategies to protect children

from being harmed by climate impacts.

7.6 References Bull G. 1980. The weather and deaths from pneumonia. Lancet 1(8183):1405-1408.

Checkley W, Epstein L, Gilman R, Figueroa D, Cama R, Patz J, et al. 2000. Effect of

El Niño and ambient temperature on hospital admissions for diarrhoeal

diseases in Peruvian children. Lancet 355(9202):442-450.

Epstein PR. 2005. Climate change and human health. N Engl J Med 353(14):1433-

1436.

Feng, BS, He SH, Zheng PY, Wu L, Yang PC. 2007. Mast cells play a crucial role in

staphylococcus aureus peptidoglycan-induced diarrhea. Am J Pathol

171(2):537-547.

Gasparrini A, Armstrong B, Kenward MG. 2010. Distributed lag non-linear models.

Stat Med 29(21):2224-2234.

Gerba C, Rose J, Haas C. 1996. Sensitive populations: who is at the greatest risk? Int

J Food Microbiol 30(1-2):113-123.

Graudenz GS, Landgraf RG, Jancar S, Tribess A, Fonseca SG, Faé KC, et al. 2006.

The role of allergic rhinitis in nasal responses to sudden temperature changes.

J Allergy Clin Immunol 118(5):1126-1132.

Green R, Basu R, Malig B, Broadwin R, Kim J, Ostro B. 2010. The effect of

temperature on hospital admissions in nine California counties. Int J Public

Health 55(2):113-121.

Page 190: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

172 Chapter 7:Results paper five

Guo Y, Barnett AG, Yu W, Pan X, Ye X, Huang C, et al. 2011. A large change in

temperature between neighbouring days increases the risk of mortality. PLoS

ONE 6(2):e16511.

Kan H, London SJ, Chen H, Song G, Chen G, Jiang L, et al. 2007. Diurnal

temperature range and daily mortality in Shanghai, China. Environ Res

103(3):424-431.

Karl TR, Knight RW, Plummer N. 1995. Trends in high-frequency climate

variability in the twentieth century. Nature 377(6546):217-220.

Lin H, Zhang Y, Xu Y, Xu X, Liu T, Luo Y, et al. 2013. Temperature changes

between neighboring days and mortality in summer: A distributed lag non-

linear time series analysis. PLoS ONE 8(6):e66403.

McMichael AJ. 2013. Globalization, climate change, and human health. N Engl J

Med 368(14): 1335-1343.

Ramsay DB, Stephen S, Borum M, Voltaggio L, Doman, DB. 2010. Mast cells in

gastrointestinal disease. Gastroenterol Hepatol 6(12):772-777.

Sheffield P, Landrigan P. 2011. Global climate change and children's health: threats

and strategies for prevention. Environ Health Perspect 119(3):291-298.

Togias AG, Naclerio RM, Proud D, Fish JE, Adkinson NF, Kagey-Sobotka A, et al.

1985. Nasal challenge with cold, dry air results in release of inflammatory

mediators. Possible mast cell involvement. J Clin Invest 76(4):1375-1381.

Xu Z, Etzel RA, Su H, Huang C, Guo Y, Tong S. 2012a. Impact of ambient

temperature on children's health: A systematic review. Environ Res

117(0):120-131.

Page 191: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

173

Chapter 7:Results paper five 173

Xu Z, Hu W, Su H, Turner LR, Ye X, Wang J, Tong S. 2013a. Extreme temperatures

and paediatric emergency department admissions. J Epidemiol Community

Health 68(4):304-311.

Xu Z, Huang C, Hu W, Turner LR, Su H, Tong S. 2013b. Extreme temperatures and

emergency department admissions for childhood asthma in Brisbane,

Australia. Occu Environ Med 70(10):730-735.

Xu Z, Huang C, Su H, Turner L, Qiao Z, Tong S. 2013c. Diurnal temperature range

and childhood asthma: a time-series study. Environ Health 12(1):12.

Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, et al. 2012b. Climate change and

children’s health—A call for research on what works to protect children. Int J

Environ Res Public Health 9(9):3298-3316.

Page 192: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Chapter 7: Results paper five 174

Page 193: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

175

Chapter 8: Discussion and conclusions 175

Chapter 8: Discussion and conclusions

Previous chapters (Chapters three to seven) have presented the main results of each

study, as well as discussions and conclusions. In this chapter, I discuss the key

findings of this thesis as an integrated body of work, and explore the mechanisms

and implications of the findings and make recommendations for future research

direction.

8.1 Key findings

In the spatiotemporal analysis (Chapter four), I found children suffering pneumonia

and diarrhoea who were admitted to emergency departments in Queensland from

2007 to 2011 were mainly from central west, northwest and far north of Queensland.

Mount Isa was the region at highest risk for both childhood pneumonia and

diarrhoea.

The subsequent time-series studies (Chapters four and five) using satellite remote

sensing temperature data revealed that both high and low temperatures were

associated with an increase in emergency department visits (EDVs) for childhood

pneumonia and diarrhoea in Brisbane from 2001 to 2010. Heat waves and cold spells

had added effects on childhood pneumonia, and the magnitude of the effects

increased with the intensity and duration of heat waves and cold spells. Heat waves

also had an added effect on childhood diarrhoea. The high temperature effects on

both childhood pneumonia and diarrhoea experienced a decreasing trend from 2001

to 2010, while the cold effects on pneumonia and diarrhoea changed little over this

Page 194: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

176 Chapter 8: Discussion and conclusions

period. Indigenous children appeared to be more vulnerable to the impacts of high

and low temperatures on pneumonia and diarrhoea than non-Indigenous children.

Several findings from the two studies examining the effects of temperature

variability (diurnal temperature range (DTR) and temperature change between two

neighbouring days (TCN)) on childhood pneumonia and diarrhoea (Chapters six and

seven) are noteworthy. Big TCNs (<-2 °C) were associated with increases in EDVs

for childhood pneumonia and diarrhoea, and big DTR may increase EDVs for

childhood diarrhoea. Every year, from May to September, especially July, children

were at high risk posed by big DTRs and TCNs.

8.2 Mechanisms and implications

Mining industry is one of the mainstay industries of Australia. Mount Isa, the city

located adjacent to Mount Isa Mines, is the largest emitter of sulphur dioxide, lead

and other metals in Australia (National Pollution Inventory, 2010). The adverse

impacts of exposure to environmental hazards on human health in Mount Isa,

especially increased blood lead level in children because of high lead exposure, have

attracted public health attention since early 1990s (Munksgaard et al. 2010).

Considerable evidence has shown that some of the negative health, intellectual and

socio-behavioural effects due to high blood lead level can be lifelong (Lanphear et al.

2005; Mark et al. 2011).

The results of our study indicated that there was a highest risk for childhood

pneumonia and diarrhoea in Mount Isa, suggesting that there might be some risk

factors associated with both pneumonia and diarrhoea in this area. The semi-arid

freshwater aquatic system contaminated by mining may cause more diarrhoea

Page 195: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

177

Chapter 8: Discussion and conclusions 177

episodes in children (Mark et al. 2009), and air pollutants emitted by mining may

increase childhood pneumonia cases (Barnett et al. 2005). How to minimize the

health impacts of environmental hazard exposure among Mount Isa children has been

discussed by not only public health sector but also industry and government

authorities (Mark et al. 2010). Beyond environmental risk factors, social

disadvantages (Walker et al. 2013), as well as cultural barriers to health care

(McBain-Rigg and Veitch 2011), may also affect the health of children in Mount Isa.

Difference in access to health care matter, as do differences in lifestyle, but the key

determinants of social inequalities in health lie in the circumstances where children

are born, grow, live and work.

In the context of climate change, two pivotal steps of public health surveillance are to

monitor temperature-health exposure-response function and to monitor the

vulnerabilities (social vulnerabilities, environmental vulnerabilities and underlying

health conditions) (Pascal et al. 2012). It has been documented that climate change

may impact respiratory infections and diarrhoeal diseases in Australia (Harley et al.

2011). Thus, I conducted the time-series studies to quantify the effects of both

extreme temperatures and prolonged extreme temperatures on childhood pneumonia

and diarrhoea and to identify the populations vulnerable to these effects. In these

studies, I used the satellite remote sensing data, which have been proven to be more

accurate than monitor-based data especially for those areas without extensive

monitoring networks (Lee et al. 2012). Prior research mainly focused on quantifying

the impact of hot temperature on children’s health (Green et al. 2010). However, it is

essential to examine the balance of hot and cold temperature-related morbidity

because future increasing temperature may also reduce the occurrence of cold-related

diseases (Xu et al. 2012). In this study, I found the magnitude of the main effects of

Page 196: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

178 Chapter 8: Discussion and conclusions

heat and cold on childhood pneumonia was similar, and the main effect of cold on

childhood diarrhoea was even greater than heat effect, indicating that the burden of

childhood pneumonia and diarrhoea in Brisbane due to the main effects of

temperature may not substantially increase as climate change progresses, even

though there are still a lot of uncertainties.

The findings of our study also suggested that the effects of heat on childhood

pneumonia and diarrhoea appeared to decline in the past decade, even though the

relatively short study period restricts us to fully unveil the long-term trend (e.g., over

100 years) in the relationship between heat and childhood pneumonia and diarrhoea.

The decreasing trend in the heat effects on childhood pneumonia and diarrhoea may

partially be explained by two reasons: First, children in Brisbane may have gradually

adapted to heat effect as Brisbane normally has a very long summer every year.

Second, the increasing use of air conditioning in recent years may also reduce

children’s exposure to heat (Ostro et al. 2010). In our study, heat waves and cold

spells were found to have added effects on childhood pneumonia, and heat waves

may also increase EDVs for childhood diarrhoea in Brisbane. It is projected that, as

climate change progresses, there will be more frequent, more intense and longer-

lasting heat waves (Meehl and Tebaldi 2004), and therefore, it is critical to develop a

comprehensive heat early warning system in Brisbane targeting to protect children

from the adverse impact of heat (Xu et al. 2014).

Indigenous children were found more vulnerable to the impacts of extreme

temperatures on pneumonia and diarrhoea compared with non-Indigenous children in

this thesis. Indigenous Australians are generally the least healthy population of all

Indigenous populations in the world (Australian Institute of Health and Welfare

2012), and they face enormous disadvantages compared with general Australians. An

Page 197: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

179

Chapter 8: Discussion and conclusions 179

Indigenous child is 2.5 times more likely to be born into the lowest income family,

and has a one in two chance of living in a one-parent family when compared with the

general population (National Mental Health Commission. Australian Government.

2012), and all these factors make them vulnerable to environmental hazards. Climate

change may exacerbate current health disparities between Indigenous and non-

Indigenous children (Fritze et al. 2008), and hence future climate change adaptation

and mitigation plans aiming to prevent Australians from adverse impacts of extreme

temperatures should focus more on Indigenous children. The health status of

Australia’s Aboriginal and Torres Strait Islander children is a significant public

health concern. The overall hospital admission rate for acute lower respiratory

infections in Indigenous children of northwest Queensland was approximately four

times higher than the rate for non-Indigenous children (Janu et al. 2014). Climate

change may exacerbate current health disparities between Indigenous and non-

Indigenous children (Fritze et al. 2008), and hence future climate change adaptation

and mitigation plans aiming to protect Australian children from adverse impacts of

extreme temperatures should focus more on Indigenous communities, especially for

those who are living in the distant regions (e.g., Mount Isa).

Climate change will not only increase global surface mean temperature but also

impact temperature variability (Schar et al. 2004). It has been found that increasing

temperature variability largely contributed to the record-breaking heat wave affecting

the whole Europe in 2003 (Schar et al. 2004). There are three widely accepted ways

to define temperature variability: DTR (Xu et al. 2013), TCN (Lin et al. 2013) and

standard deviation of mean temperature within a certain period of time (one week,

one month or one year) (Xu et al. 2014a). Our studies looking at the impacts of DTR

and TCN on childhood pneumonia and diarrhoea found that when mean temperature

Page 198: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

180 Chapter 8: Discussion and conclusions

decreased for more than 2 °C from one day to the next, EDVs for childhood

pneumonia and diarrhoea increased significantly, suggesting that parents and

caretakers of children in Brisbane should be more aware of the temperature change in

two neighbouring days, especially in July. DTR over 17 °C was also associated with

an increase in EDVs for childhood diarrhoea, highlighting that more research and

targeted health policies and programs are needed to minimize the risk posed by great

DTRs.

The association between infectious diseases and weather conditions has long been

appreciated. Back to early 20th century, a study in Netherlands has demonstrated the

relationship between the increase in upper respiratory tract infections and outdoor

cold temperature (van Loghem1928). “Winter gastroenteritis” was a recognized

illness of early childhood before rotavirus was identified. In a review of 34 studies

conducted prior to 1990, Cook et al. have summarized that rotavirus infection

typically occur in winter, which puts forth the hypothesis that cold temperature may

increase rotavirus diseases (Cook et al. 1990). While subtle changes in local

temperature may play a role in explaining the seasonal cycling of childhood

pneumonia and diarrhoea in some settings, some other climatic factors, including

relative humidity (Moe and Shirley 1982), rainfall (Ansari et al. 1991) and sunshine

(Paynter et al. 2013), may also contribute to the occurrence of these infectious

diseases. Further, the transmission of these infectious diseases involves a lot of

factors, such as host behaviour and susceptibility, as well as spread and survivability

of pathogens etc., and climatic factors alone cannot fully explain its complexity. It is

still far from satisfactory to unfold the drivers of the transmission of childhood

infectious diseases.

Page 199: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

181

Chapter 8: Discussion and conclusions 181

8.3 Strengths and Limitations

The impact of climate change on children’s health is scarcely researched so far. This

thesis has several strengths. First, it utilized the satellite remote sensing data to

quantify the effects of temperature on childhood pneumonia and diarrhoea, which

greatly minimized the measurement bias and therefore drew more accurate

conclusions. Second, it examined the balance between high and low temperature

effects on children’s health, which provided pivotal information on “how

temperature-related burden of childhood pneumonia and diarrhoea may change as

climate change proceeds”. Third, to the best of my knowledge, it examined the

effects of heat waves and cold spells on childhood pneumonia and diarrhoea for the

first time, and identified the relative importance of extremely high (low)

temperatures and sustained high (low) temperatures in the occurrence of childhood

pneumonia and diarrhoea. Finally, it assessed the impacts of temperature variation on

childhood pneumonia and diarrhoea, which suggests that the public health sector in

Brisbane take precautionary measures ahead of not only extreme temperatures and

but also big temperature variations.

This thesis also has several weaknesses. First, the ecological design which used the

aggregated data on temperature and EDVs for childhood pneumonia and diarrhoea in

Brisbane may result in some biases in exposure and/or outcome measures. The

ecological design also largely restricted us to examine the causation between

temperature and childhood pneumonia and diarrhoea. Second, due to the data

availability issue, we used the patients’ postcodes to do the spatial analysis (Chapter

3), which is not ideal compared with data by Statistical Local Areas (SLA) and Local

Government Areas (LGA). We are collecting data on EDVs for childhood diseases

Page 200: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

182 Chapter 8: Discussion and conclusions

by SLA for implementing further studies. Third, there were some missing values in

the dataset, especially for the Indigenous status of the patients, which might to some

extent impact the conclusions we drew in the thesis. Forth, the time-series studies

(Chapters 4 to 7) only focused on one subtropical city, and thus it needs to be

cautious to generalize our findings to regions of other climates. Fifth, the range of

climatic variables we used is limited compared with a previous study looking at the

impacts of temperature, sun hours, rainfall and relative humidity on childhood

pneumonia (Paynter et al. 2013). Sixth, we did not have the pathogen data, which

restricted us to quantify the effects of temperature on specific pathogens.

8.4 Future research directions

Existing studies looking at the impacts of temperature on childhood pneumonia and

diarrhoea adopted various statistical approaches and yielded different types of

outputs, which makes it hard to quantitatively combine the findings together. Further,

majority of previous studies are one-city-only study, whereas children’s vulnerability

to the adverse impacts of extreme temperatures and large temperature variability

depends largely on their access to the climate change adaptation infrastructures of the

city they live, and thus it is necessary to conduct a large scale study using a

consistent statistical method to assess the heterogeneity in the relationship between

temperature and childhood pneumonia and diarrhoea across different cities. Even for

the children living in the same city but different suburbs, they may have different

sensitivities to temperature impact. Therefore, it is also strongly needed to evaluate

the with-in-city heterogeneity in children’s sensitivity to the adverse impacts of

Page 201: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

183

Chapter 8: Discussion and conclusions 183

temperature on childhood pneumonia and diarrhoea, as well as to identify the drivers

behind, facilitating the future resource allocation to the priority subpopulations.

Different pathogens of childhood pneumonia and diarrhoea have different seasonal

variations, and the incubation period varies for viral and bacterial infections as well,

which requires future studies using lab-confirmed pathogen data to more accurately

quantify the effects of temperature on childhood pneumonia and diarrhoea caused by

different pathogens. Further, under reporting issue exists in surveillance data. Some

children with mild symptoms may do not want to seek medical care. Milinovich et al.

(Milinovich et al. 2014) highlighted using Internet resource to monitor emerging

infectious diseases, which can also be applied to the traditional children’s diseases

area. Using data collected from Internet-based surveillance system to quantify the

relationship between climate change and childhood pneumonia and diarrhoea may

more adequately capture the complexity of this relationship.

As climate change continues, the temperature-related burden of childhood

pneumonia and diarrhoea may vary accordingly (Walker et al. 2013). It is important

to project the burden of childhood pneumonia and diarrhoea under different climate

change scenarios, especially for those regions with a great risk for these diseases

(Huang et al. 2011).

8.5 Conclusions

In conclusion, this thesis contributes to the scientific knowledge in three ways: I) it

identified the risk areas of EDVs for childhood pneumonia and diarrhoea in

Queensland, Australia; II) it quantified the impacts of extreme temperatures (ie., heat

and cold) and prolonged extreme temperatures (ie., heat waves and cold spells) on

Page 202: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

184 Chapter 8: Discussion and conclusions

EDVs for childhood pneumonia and diarrhoea; III) it examined the effects of

temperature fluctuation on EDVs for childhood pneumonia and diarrhoea.

Australia shoulders a big burden of childhood pneumonia and diarrhoea. Mount Isa

was observed as the high risk area where EDVs for both childhood pneumonia and

diarrhoea, even though EDVs for childhood pneumonia and diarrhoea experienced a

big decrease from 2007 to 2011. In light of the fact that children in Mount Isa are

exposed to enormous environmental hazards (polluted air and contaminated water,

etc.) due to mining activity, precautionary initiatives should be taken to protect

children in this area from illness.

Climate change will increase global average temperature, and cause more frequent,

intense, and longer-lasting heat waves, which may result in increases in the EDVs for

both childhood pneumonia and diarrhoea in Brisbane. Although, as climate change

continues, childhood pneumonia and diarrhoea cases due to mild high and low

temperatures may not increase substantially, burden of childhood pneumonia and

diarrhoea associated with prolonged extreme temperatures may increase in the future.

Other than absolute temperature, temperature variability, especially a temperature

decrease over 2 °C from one day to the next which most frequently occurred in July

every year, may increase EDVs for childhood pneumonia and diarrhoea.

The fundamental motivation of this thesis is to assist with the development of climate

change adaptation strategies aiming to protect children in Queensland. In particular,

the main and added effects of heat on childhood pneumonia and diarrhoea in the past

decade, the change over time in the impact of heat waves on childhood pneumonia

and diarrhoea, the temperature variation effects on childhood pneumonia and

diarrhoea, and the high vulnerability of Indigenous children to all these effects,

Page 203: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

185

Chapter 8: Discussion and conclusions 185

should be taken into account while stakeholders are designing and implementing

future climate change adaptation plans for Queensland.

8.6 References

Australian Institute of Health and Welfare. 2012. Australia’s Health 2012. Canberra.

Ansari SA, Springthorpe VS, Sattar SA. 1991. Survival and vehicular spread of

human rotaviruses: possible relation to seasonality of outbreaks. Rev Infect

Dis 13(3):448-461.

Barnett AG, Williams GM, Schwartz J, Neller AH, Best TL, Petroeschevsky AL, et

al. 2005. Air pollution and child respiratory health. Am J Respir Crit Care

Med 171(11):1272-1278.

Cook SM, Glass RI, Lebaron CW, Ho MS. Global seasonality of rotavirus infections.

Bull World Health Organ 68(2):171-177.

Fritze J, Blashki G, Burke S, Wiseman J. 2008. Hope, despair and transformation:

Climate change and the promotion of mental health and wellbeing. Int J Ment

Health Syst 2(1):13.

National Mental Health Commission. Australian Government. 2012. A contributing

life: the 2012 national report card on mental health and suicide prevention.

Green R, Basu R, Malig B, Broadwin R, Kim J, Ostro B. 2010. The effect of

temperature on hospital admissions in nine California counties. Int J Public

Health 55(2):113-121.

Harley D, Bi P, Hall G, Swaminathan A, Tong S, Williams C. 2011. Climate change

and infectious diseases in Australia: Future prospects, adaptation options, and

research priorities. Asia Pac J Public Health 23(2s):54S-66S.

Page 204: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

186 Chapter 8: Discussion and conclusions

Huang C, Barnett A, Wang X, Vaneckova P, FitzGerald G, Tong S. 2011. Projecting

future heat-related mortality under climate change scenarios: a systematic

review. Environ Health Perspect 119(12):1681-1690.

Janu EK, Annabattula BI, Kumariah S, Zajaczkowska M, Whitehall JS, Edwards MJ,

et al. Paediatric hospitalisations for lower respiratory tract infections in

Mount Isa. Med J Aust 200(10):591-594.

van Loghem JJ. 1928. An epidemiological contribution to the knowledge of the

respiratory diseases. J Hyg (Lond) 28(1):33-54.

Lanphear B, Hornung R, Khoury J, Yolton K, Baghurst P, Bellinger D, et al. 2005.

Low-level environmental lead exposure and children's intellectual function:

an international pooled analysis. Environ Health Perspect 113(7):894-899.

Lee S, Serre M, Donkelaar DV, Martin R, Burnett R, Jerrett M. 2012. Comparison of

geostatistical interpolation and remote sensing techniques for estimating

long-term exposure to ambient PM2.5 concentrations across the continental

United States. Environ Health Perspect 120(12):1727-1732.

Lin H, Zhang Y, Xu Y, Xu X, Liu T, Luo Y, et al. 2013. Temperature changes

between neighboring days and mortality in summer: A distributed lag non-

linear time series analysis. PLoS ONE 8(6):e66403.

Mcbain-Rigg KE, Veitch C. Cultural barriers to health care for Aboriginal and

Torres Strait Islanders in Mount Isa. Aust J Rural Health 19(2): 70-74.

Meehl GA, Tebaldi C. 2004. More intense, more frequent, and longer lasting heat

waves in the 21st century. Science 305(5686):994-997.

Milinovich GJ, Williams GM, Clements ACA, Hu W. 2014. Internet-based

surveillance systems for monitoring emerging infectious diseases. Lancet

Infect Dis 14(2):160-168.

Page 205: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

187

Chapter 8: Discussion and conclusions 187

Moe K, Shirley J. 1982. The effects of relative humidity and temperature on the

survival of human rotavirus in faeces. Arch Virol 72(3):179-186.

Munksgaard NC, Taylor MP, Mackay A. 2010. Recognising and responding to the

obvious: the source of lead pollution at Mount Isa and the likely health

impacts. Med J Aust 193(3):131-132.

National Pollution Inventory (2010).

http://www.npi.gov.au/npidata/action/load/individual-facility-

detail/criteria/year/2009/browse-type/Company/regbusiness-

name/MOUNT%2BISA%2BMINES%2BLTD/jurisdiction-

facility/Q020MIM001. Retrieved October 1st 2013

Ostro B, Rauch S, Green R, Malig B, Basu R. 2010. The effects of temperature and

use of air conditioning on hospitalizations. Am J Epidemiol 172(9):1053-

1061.

Pascal M, Viso AC, Medina S, Delmas MC, Beaudeau P. 2012. How can a climate

change perspective be integrated into public health surveillance? Public

Health 126(8):660-667.

Paynter S, Weinstein P, Ware RS, Lucero MG, Tallo V, Nohynek H, et al. 2013.

Sunshine, rainfall, humidity and child pneumonia in tropics: time-series

analyses. Epidemiol Infect 141(6):1328-1336.

Schar C, Vidale PL, Luthi D, Frei C, Haberli C, Liniger MA, Appenzeller C. 2004.

The role of increasing temperature variability in European summer

heatwaves. Nature 427(6972):332-336.

Taylor MP, Mackay A, Kuypers T, Hudson-Edwards K. 2009. Mining and urban

impacts on semi-arid freshwater aquatic systems: the example of Mount Isa,

Queensland. J Environ Monit 11(5):977-986.

Page 206: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

188 Chapter 8: Discussion and conclusions

Taylor MP, Schniering CA. 2010. The public minimization of the risks associated

with environmental lead exposure and elevated blood lead levels in children,

Mount Isa, Queensland, Australia. Arch Environ Occup Health 65(1):45-48.

Taylor MP, Schniering CA, Lanphear BP, Jones AL. 2011. Lessons learned on lead

poisoning in children: One-hundred years on from Turner's declaration. J

Paediatr Child Health 47(12):849-856.

Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, et al. 2013. Global

burden of childhood pneumonia and diarrhoea. Lancet 381(9875):1405-1416.

Xu Z, Etzel RA, Su H, Huang C, Guo Y, Tong S. 2012. Impact of ambient

temperature on children's health: A systematic review. Environ Res 117:120-

131.

Xu Z, Hu W, Wang X, Huang C, Tong S. 2014a. The impact of temperature

variability on years of life lost. Epidemiology 25(2):313-314.

Xu Z, Huang C, Su H, Turner L, Qiao Z, Tong S. 2013. Diurnal temperature range

and childhood asthma: a time-series study. Environmental Health 12(1):12.

Xu Z, Sheffield P, Su H, Wang X, Bi Y, Tong S. 2014b. The impact of heat waves

on children’s health: a systematic review. Int J Biometeorol 58(2):239-247.

Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, et al. 2012b. Climate change and

children’s health—A call for research on what works to protect children. Int J

Environ Res Public Health 9(9):3298-3316.

Page 207: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

Appendices 189

Appendices

Conference:

“2013 Conference of International Society for Environmental Epidemiology”, 20-23

August 2013, Basel, Switzerland (Poster presentation: Extreme temperatures and

paediatric emergency department admissions)

Journal Articles:

1. Xu Z, Hu W, Zhang Y, Wang X, Zhou M, Su H, Huang C, Tong S,

Guo Q (2015). Exploration of diarrhoea seasonality and its drivers in China.

Scientific Reports, 5:8241.

2. Xu Z, Hu W, Tong S (2015). The geographic co-distribution and

socio-ecological drivers of childhood pneumonia and diarrhea in Queensland,

Australia. Epidemiology and Infection, 143(5):1096-104.

3. Cheng J, Wu J, Xu Z, Zhu R, Wang X, Li K, Wen L, Yang H, Su H

(2014). Associations between extreme precipitation and childhood hand, foot

and mouth disease in urban and rural areas in Hefei, China. Science of the

Total Environment, 497-498: 484-90.

4. Cheng J, Zhu R, Xu Z, Wang X, Li K, Su H. Temperature variation

between neighboring days and mortality: a distributed lag non-linear analysis

(2014). International Journal of Public Health, 59(6):923-31.

5. Xu Z, Hu W, Zhang Y, Wang X, Tong S, Zhou M (2014).

Spatiotemporal pattern of bacillary dysentery in China from 1990 to 2009:

What is the driver behind? PLoS ONE, 9(8):e104329.

Page 208: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

190 Appendices

6. Xu Z, Zhang W, Hu W, Tong S (2014). Seasonal amplitude of

hemorrhagic fever with renal syndrome in China: a call for attention to

neglected regions. Clinical Infectious Diseases, 59(7):1040-2.

7. Xu Z, Hu W, Tong S (2014). Temperature variability and childhood

pneumonia: an ecological study. Environmental Health, 13(1):51.

8. Xu Z, Liu Y, Ma Z, Toloo GS, Hu W, Tong S (2014). Assessment of

the temperature effect on childhood diarrhoea using satellite imagery.

Scientific Reports, 4:5389.

9. Xu Z, Liu Y, Ma Z, Li S, Hu W, Tong S (2014). Impact of

temperature on childhood pneumonia estimated from satellite remote sensing.

Environmental Research, 132:334-341.

10. Cheng J, Xu Z, Zhu R, Wang X, Jin L, Song J, Su H (2014). Impact

of diurnal temperature range on human health: a systematic review.

International Journal of Biometeorology, doi: 10.1007/s00484-014-0797-5

11. Xu Z, Hu W, Wang X, Huang C, Tong S. (2014). The impact of

temperature variability on years of life lost. Epidemiology, 25(2): 313-314.

12. Xu Z, Sheffield PE, Su H, Wang X, Bi Y, Tong S. (2014). The impact

of heat waves on children’s health: a systematic review. International Journal

of Biometeorology, 58(2):239-247.

13. Xu Z, Hu W, Su H, Turner LR, Ye X, Wang J, Tong S. (2013).

Extreme temperatures and paediatric emergency department admissions.

Journal of Epidemiology and Community Health, 68(4):304-311.

Page 209: ASSESSMENT OF TEMPERATURE EFFECTS ON CHILDHOOD … · 2015. 5. 6. · This thesis identified the risk areas of childhood pneumonia and diarrhoea in Queensland and quantified the effects

191

Appendices 191

14. Xu Z, Hu W, Williams G, Clements AG, Kan H, Tong S. (2013). Air

pollution, temperature and pediatric influenza in Brisbane, Australia.

Environmental International, 59:384-388.

15. Xu Z, Huang C, Hu W, Turner LR, Su H, Tong S. (2013). Extreme

temperatures and emergency department admissions for childhood asthma in

Brisbane, Australia. Occupational and Environmental Medicine, 70(10): 730-

735.

16. Xu Z, Huang C, Hu W, Turner LR, Su H, Qiao Z, Tong S. (2013). Is

diurnal temperature range a risk factor for childhood diarrhoea? PLoS ONE

8(5):e64713.

17. Huang C, Barnett AG, Xu Z, Chu C, Wang X, Turner LR, Tong S.

(2013). Managing the health effects of temperature in response to climate

change: challenges ahead. Environmental Health Perspectives, 121(4): 415-

419.

18. Xu Z, Huang C, Hu W, Turner LR, Su H, Qiao Z, Tong S. (2013).

Diurnal temperature range and childhood asthma: a time-series study.

Environmental Health, 12:12.

19. Xu Z, Sheffield PE, Hu W, Su H, Yu W, Qi X, Tong S. (2012).

Climate change and children’s health: a call for research on what works to

protect children. International Journal of Environmental Research and Public

Health, 9(9): 3298-3316.

20. Xu Z, Etzel RA, Su H, Huang C, Guo Y, Tong S. (2012). Impact of

temperature on children’s health: a systematic review. Environmental

Research, 117: 120-131.