BMJ Open Hornshøja, Ane Bærent Fiskera,b, Manuel Fernandesa, Amabelia Rodriguesa, Christine Stabell Benna,b, Peter Aabya,b a) Bandim Health Project, INDEPTH Network, Apartado 861,

For peer review only

Vaccination Coverage and Frequency of Out-of-Sequence Vaccinations in Rural Guinea-Bissau

Journal: BMJ Open

Manuscript ID: bmjopen-2012-001509

Article Type: Research

Date Submitted by the Author: 18-May-2012

Complete List of Authors: Hornshøj, Linda; Bandim Health Project, INDEPTH NETWORK, Fisker, Ane; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut, ; Bandim Health Project Fernandes, Manuel; Bandim Health Project, INDEPTH NETWORK, Rodrigues, Amabelia; Bandim Health Project, INDEPTH NETWORK, Benn, Christine; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut, ; Bandim Health Project, INDEPTH NETWORK,

Aaby, Peter; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut, ; Bandim Health Project, INDEPTH NETWORK,

<b>Primary Subject Heading</b>:

Public health

Secondary Subject Heading: Global health, Infectious diseases, Epidemiology, Health services research

Keywords:

Public health < INFECTIOUS DISEASES, International health services < HEALTH SERVICES ADMINISTRATION & MANAGEMENT, Epidemiology < INFECTIOUS DISEASES, Paediatric infectious disease & immunisation < PAEDIATRICS

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open on 2 F

ebruary 2019 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2012-001509 on 19 Novem

ber 2012. Dow

nloaded from

http://bmjopen.bmj.com/


Vaccination Coverage and Frequency of Out-of-Sequence Vaccinations in Rural

Guinea-Bissau

Linda Hornshøja, Ane Bærent Fisker

a,b, Manuel Fernandes

a, Amabelia Rodrigues

a,

Christine Stabell Benna,b, Peter Aaby

a,b

a) Bandim Health Project, INDEPTH Network, Apartado 861, Bissau, Guinea-Bissau

b) Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project,

Statens Serum Institut, 2300 Copenhagen S, Denmark

Corresponding author:

Linda Hornshøj, Research Center for Vitamins and Vaccines (CVIVA), Bandim

Health Project, Statens Serum Institut, 2300 Copenhagen S, Denmark. E-mail:

[email protected]. Telephone: +45 32683162.

Keywords:

EPI, vaccination, coverage, out-of-sequence, vaccines

Word count:

Abstract 284, text 2552

Page 1 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on 2 February 2019 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2012-001509 on 19 N

ovember 2012. D

ownloaded from



2

ABSTRACT

Objective

The WHO aims for 90% coverage of the Expanded Program on Immunization (EPI),

which in Guinea-Bissau includes bacillus Calmette-Guérin vaccine (BCG) at birth, 3

doses of diphtheria-tetanus-pertussis vaccine (DTP) and oral polio vaccine (OPV) at

6, 10 and 14 weeks and measles vaccine (MV) at 9 months. The WHO assesses

coverage by 12 months of age. The sequence of vaccines may have an effect on child

mortality, but is not considered in official statistics or assessments of programme

performance. We assessed vaccination coverage and frequency of out-of-sequence

vaccinations by 12 and 24 months of age.

Design

Observational cohort study.

Setting and participants

The Bandim Health Project’s rural Health and Demographic Surveillance System in

rural Guinea-Bissau was used, it covers 258 randomly selected villages. Villages are

visited biannually and vaccination cards inspected to ascertain vaccination status.

Between 2003 and 2009 vaccination status by 12 months of age was assessed for 5806

children aged 12-23 months and by 24 months of age for 3792 children aged 24-35

months.

Outcome measures

Coverage of EPI vaccinations and frequency of out-of-sequence vaccinations.

Results

Half of 12-month-old children and 65% of 24-month-old children had completed all

EPI vaccinations. Many children received vaccines out-of-sequence: 54% of BCG

vaccinated children received DTP with or before BCG and 28% of measles vaccinated

children received DTP with or after MV by 12 months of age, by 24 months of age

58% and 35%, respectively.

Conclusions

In rural Guinea-Bissau vaccination coverage by 12 months of age was low, but

continued to increase beyond 12 month of age. The current practice of limiting

vaccinations to children below 12 months of age will lower overall coverage

considerably. More than half of all children received vaccinations out-of-sequence.

Page 2 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



3

ARTICLE SUMMARY

Article focus

- The World Health Organization (WHO) aims for a 90% coverage of the

Expanded Program on Immunization (EPI) by 12 months of age.

- The third dose of diphtheria-tetanus-pertussis vaccine (DTP) is used as a

proxy for vaccination coverage and performance.

- Several studies indicate that the sequence of vaccinations affects child

mortality, but current polices do not take out-of-sequence vaccinations into

account.

Key messages

- Only half of all children were fully vaccinated by 12 months of age, but the

coverage continued to increase beyond 12 months of age.

- Limiting vaccinations to children below 12 months of age will lower overall

vaccination coverage considerably.

- More than half of the children received vaccinations out-of sequence.

Strengths and limitations of this study

- The study population was sizeable and the continued demographic

surveillance has many advantages compared with other household surveys.

- The study was limited to rural Guinea-Bissau and may not necessarily be

representative of other low-income settings.

Page 3 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



4

INTRODUCTION

In 1974 the World Health Organization (WHO) launched the Expanded Program on

Immunization (EPI), to provide routine vaccinations against tuberculosis, polio,

diphtheria, tetanus, pertussis and measles to all children. At the time less than 5% of

children worldwide were vaccinated against these diseases.[1] In most low-income

countries the standard schedule for EPI vaccines includes bacillus Calmette-Guérin

vaccine (BCG) and oral polio vaccine (OPV) at birth, diphtheria- tetanus-pertussis

vaccine (DTP) and OPV at 6, 10 and 14 weeks and measles vaccine (MV) at 9 months

of age.[2]

The Global Immunization Vision and Strategy 2006-2015 (GIVS) developed by the

WHO/UNICEF focused on ensuring equity in access to vaccination and aimed for a

90% national vaccination coverage and a 80% vaccination coverage in every district

by 2010.[3]

WHO and UNICEF estimates of vaccination coverage are based on reports of

vaccinations performed by service providers (administrative data) and surveys of

vaccination history (coverage surveys). The estimates refer to vaccinations given

during routine immunisation services to children below 12 months of age, not

including supplemental activities such as OPV and MV campaigns. The

administrative data are based on the number of vaccine doses administered to the

estimated target population, most commonly the estimated number of children

surviving their first year of life.[4] The WHO EPI cluster methodology is commonly

used for coverage surveys. Survey results typically report on annual cohorts of

children so that all children included have had 12 months to receive all vaccinations

(being between 12–23 months of age at the time of the survey as shown in figure

1).[4,5]

Studies from low-income countries suggest that the routine vaccinations have effects

on mortality beyond that of protecting against the target disease. These effects are

referred to as non-specific effects (NSE). BCG and MV have beneficial effects in

lowering overall child morbidity and mortality,[6-10] whereas DTP may be associated

with increased morbidity and mortality.[6,9-17] The NSEs depend on the most recent

vaccine and giving vaccines out-of-sequence may therefore affect child mortality.

Page 4 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



5

Worryingly several studies have shown that DTP given together with MV or after MV

is associated with increased child morbidity and mortality.[12,18-21]

The purpose of this study was to determine the overall vaccination coverage of EPI

routine vaccinations and to determine the prevalence of out-of-sequence vaccinations

in rural Guinea-Bissau. Both outcomes were determined among 12-month-old and 24-

month-old children.

METHODS

Setting

In 1978, the Bandim Health Project (BHP) was established as a Health and

Demographic Surveillance System (HDSS) in Guinea-Bissau. Outside the capital,

Guinea-Bissau is divided into ten health regions. Since 1990, five of these regions

have been visited biannually by mobile teams from BHP. At each visit the teams

collect information about health, socio-economic and vaccination status. In 2006, the

remaining regions were included in the surveillance and the mobile teams now cover

258 villages in all ten regions. The study population in the HDSS is based on 182

randomly selected clusters of 100 women of fertile age and their children below five

years of age. In the cluster areas all new households are registered and in all

households new women and children taking up residence are added. The HDSS

cohort is therefore continuously growing.

Study population

The study population for the present study was extracted from the HDSS cohort. The

population base for the 12-month-old children was children born from the 1st January

2004 until 31st of December 2006, registered before 12 months of age and resident in

the study area when completing 12 months of age. For the 24-month-old children,

children born from the 1st January 2004 until 31st of December 2005, registered before

24 months of age and resident in the study area when completing 24 months of age

were included.

Vaccinations

Guinea-Bissau follows the standard EPI and children in the present study were to be

vaccinated in accordance with the EPI schedule shown in figure 1.

Page 5 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



6

A child in Guinea-Bissau can be vaccinated at hospitals, most health centres and

during outreach visits to the villages. All national vaccines are provided through

UNICEF and are free of charge for the parents. However, in order for a child to

receive its vaccinations it must have a vaccination card. The card is used to register

the child’s personal information, vaccinations and the weight of the child. Although

the card should also be free of charge, it is most often sold by the health workers for

the equivalent of $ 1(unpublished data). Since September 2007, the BHP teams have

been accompanied by nurses who vaccinate children missing doses of OPV, DTP or

MV and children receive a vaccination card free of charge. In selection of the two

cohorts for this study, children were only included if they had reached 12 or 24

months of age before the BHP nurses began administering vaccines.

Assessment of vaccination coverage

To emulate the EPI survey methodology used by WHO, we determined the

vaccination coverage by 12 months of age in children aged 12-23 months at home

visits. Vaccination status was determined based on the first visit among children aged

12-23 months in 2005-2009, at which vaccination status could be assessed by

inspecting the vaccination card. If the child did not have a vaccination card at the time

of our visits, it was defined as unvaccinated if the mother confirmed that the child had

never received any vaccines. If we were not able to verify a child’s vaccinations at

any visit in the interval of 12-23 months of age either by 1) by seeing the vaccination

card or 2) ascertaining that the child had never received vaccinations, the child was

not included in the analysis. The vaccination status by 24 months of age was

determined using the same method for children visited at 24-35 months of age.

Definitions

We defined fully vaccinated as completion of the EPI vaccinations not including OPV

at birth (OPV0), because very few children received OPV0 in the rural areas.

Out-of-sequence vaccinations were 1) BCG administered together with any dose of

DTP, 2) any dose of DTP with or before BCG, 3) MV with any dose of DTP and 4)

any dose of DTP with or after MV.

Statistical analysis

Page 6 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



7

Relative risks and 95% confidence intervals (CI) were calculated using Poisson

regression with robust variance estimates.[22] Coverage for different regions was

compared using chi-2-tests. Data was analysed using STATA 11.0.

RESULTS

We determined vaccination status for 5806 children by 12 months of age and for 3792

children by 24 months of age (figure 2). A larger proportion of children were assessed

by 12 months of age (68%) than by 24 months of age (59%), RR=1.15 (95% CI=1.12-

1.18).

Vaccination coverage

By 12 months of age only 50% of children were fully vaccinated if whether or not a

child had received OPV0 was disregarded. If OPV0 was included in the definition of

“fully vaccinated” only 9% of 12-month-old children were fully vaccinated. There

were large regional differences: only 26% of children were fully vaccinated in

Tombali compared with 65% in Gabu and Bafata, the regions with highest coverage

(table 1). By 24 months of age the percentage of fully vaccinated children was 65%, a

29% (95%CI=25-34%) increase. The increase ranged from 20% to 87% in the

different regions (table 1, figure 3).

Vaccine-specific coverages by 12 months of age were: BCG 89%, OPV0 13%, OPV1

91%, OPV2 82%, OPV3 66%, DTP1 92%, DTP2 83%, DTP3 68% and MV 61%,

with large regional differences (table 2). There was only a slight increase in the

vaccines the children receive early in life when comparing vaccine coverage at 12 and

24 months of age, but the increase for vaccines given later in life was much higher.

BCG, OPV1 and DTP1 increased by 3%, whereas OPV3 increased by 12%, DTP3 by

11% and MV increased by 28% (figure 4). The DTP3 coverage was higher than the

MV coverage by 12 months of age in all regions, except for Tombali. By 24 months

of age the overall MV coverage was higher than DTP3, but in four out of ten regions

(Cacheu, Quinara, Bijagós and Bolama) the MV coverage was still lower than the

DTP3 coverage (data not shown).

Out-of-sequence vaccinations

Page 7 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



8

Among children who had received BCG by 12 months of age, 49% received BCG

together with a dose of DTP and 54% DTP with or before BCG. By 24 months of age

these proportions were 52% and 58% (table 3). The region of Bijagós managed to stay

well below 30%. In comparison in Tombali more than 75% of children in both age

groups received DTP with or before BCG (table 3).

Among children who had received MV by 12 months of age, 26% received it together

with a dose of DTP; this figure was 29% by 24 months of age. Among the children

who had received MV, 28% had received DTP with or after MV by 12 months of age,

and this increased to 35% by 24 months of age, a 25% (95%CI=16-34%) increase

(table 3). Again there were large regional differences (table 3).

DISCUSSION

Main results

Vaccinations are important in lowering overall child mortality. In our study, only half

of all children were fully vaccinated by 12 months of age and 65% by 24 months of

age (table 1, figure 2). Only estimating vaccination coverage among 12-month-old

children therefore underestimates final coverage considerably. Many children had

initiated the vaccination program, but not received all the EPI vaccines by 12 months

of age. This was reflected in high coverage of vaccines scheduled early in life, with

little increased from 12 to 24 months of age, whereas the increase in vaccines

scheduled later was much higher: MV increased by 28% (25-31%) (figure 4).

Regional differences in vaccination coverage and frequency of out-of-sequence

vaccinations were large (tables 1,3).

A substantial number of children received out-of-sequence vaccinations. More than

half of all BCG vaccinated children received DTP with or before BCG and a third of

MV vaccinated children received DTP with or after MV (table 3). As vaccination

coverage increased among 24-month-old children, so did the frequency of out-of-

sequence vaccinations (tables 1,3).

Strengths and weaknesses

The HDSS clusters are randomly selected and the sample size substantial. The

villages have been visited biannually for years, thus offering the opportunity to verify

Page 8 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



9

and correct ambiguous data. We believe that the continued surveillance contributes to

more reliable data than other types of surveys. For example, in this study, children

with previously registered vaccinations could not be misclassified as unvaccinated at

a later visit if the vaccinations card had been lost. A study from Malawi showed a

BCG scar rate of 70% among children without documents, making it clear how

vaccination coverage can be underestimated if children without vaccination cards are

simply classified as unvaccinated.[23] We also registered pregnancies and births and

could thereby determine date of birth fairly precisely if there was no written record.

Comparison to other studies

We found that estimating vaccination coverage among 12-month-old children

underestimated final coverage. Similarly, Clark and Sanderson found that DTP3

coverage increased from 67% to 74% and MV from 59% to 77%, when comparing

the coverage at 12 months of age to the coverage at 36 months of age in survey data

from 45 low- and middle-income countries.[24]

In our study we emulated the methodology used by the WHO to make the results

comparable. A WHO EPI cluster survey was conducted in the capital Bissau in 2005

using vaccination cards or history to confirm vaccination status. It found vaccine

coverage to be: BCG 87%, OPV3 60%, DTP3 59% and MV 71%.[25] When we

compared their results to ours BCG coverage did not differ, WHO OPV3 and DTP3

coverage was 10-13% (5-18%) lower and WHO MV coverage was 16% (11-21%)

higher than our estimates (table 2,[25]). The higher MV coverage may in part be

explained by our study not including data from the urban Bissau. In Bissau the

coverage is higher as vaccinations are more accessible. However, surprisingly higher

coverage of all antigens was not found in the WHO survey. If the WHO survey

underestimated the true coverage it could in part be due to the use of history, i.e.

maternal recall, to confirm vaccination status. The validity of recall is correlated to

the number of doses received and the age of the child. Mothers tend to recall

vaccinations more accurately for younger children with fewer doses and the use of

recall seems to underestimate vaccination coverage.[26,27] In the WHO survey, visits

are made to randomly selected houses according to the survey method.[5] Unlike in

our study, there are no follow-up visits to households in which the vaccination status

for a child can not be determined at the time of the visit. The WHO EPI survey

Page 9 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



10

methodology has previously been criticised for lacking methodological rigour and for

introducing bias.[28,29]

Implications

At present vaccination coverage estimates used by WHO and United Nations

Children’s Fund (UNICEF) emphasises the coverage of DTP3 by 12 months of age,

not taking into account the timing of the vaccines received or vaccines received at a

later age.[30,31] This has two consequences, which may have large implications.

First, the emphasis on DTP3 leads to more focus on DTP3 and less focus on measles

vaccine. It is often used by funding agencies such as the Global Alliance for Vaccines

and Immunizations (GAVI) to evaluate programme performance[32] and

improvement of vaccination coverage is often linked to additional funding.[33] As a

result many African countries now have a lower coverage of MV than DTP3.[2] This

is unfortunate. Several studies show that MV lowers overall child morbidity and

mortality[6-10,13-15] and several studies suggest that DTP is associated with

increased female morbidity and mortality.[6,9-15]

Second, our data was from before 2007, and it showed that MV coverage increased by

28% during the second year of life. However, the current donor focus on vaccination

status by 12 months of age leads to many countries, including Guinea-Bissau, have

now stopped vaccinating children above 12 months of age, because they do not count

in the statistics. The current policy not to vaccinate children above 12 months of age

gives children only a 3-month-time-window to be measles vaccinated. This will

undoubtedly lead to a lower MV coverage and an increasing number of children being

susceptible to measles infection.

CONCLUSION

Our study showed that the vaccination coverage in Guinea-Bissau is far from the

WHO goal of 90%. Only half of all 12-month-old children were fully vaccinated, with

great regional differences in vaccination coverage. Coverage based on vaccinations

received during the first 12 months of life significantly underestimated the final

coverage. We found that a considerable number of children received out-of-sequence

vaccinations. As several studies suggest that out-of-sequence vaccinations have an

impact on overall survival[18-21] such deviations from the schedule need to be taken

Page 10 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



11

into account and should be reported as performance indicators for vaccination

programmes in likening with specific antigen coverage.

The current drive to limit vaccinations to children below 12 months of age[34] will

prevent many children from completing their vaccination schedule and this negative

effect will be strongest for measles vaccination. This will lower heard immunity to

measles infections and increase the risk of epidemics. Since all studies show that

measles vaccination administered as the most recent vaccine has a beneficial effect on

child survival[6-8,10,14] it will have unfortunate consequences that an increasing

proportion of children will have DTP as their most recent vaccination.

[10,12,15,17,35]

Funding

This study was supported by the Danish Medical Research Council, the Aase and

Ejnar Danielsen's Foundation and DANIDA. The Bandim Health Project received

support from DANIDA to conduct their rural HDSS. CVIVA receives support from

the Danish National Research Foundation. CSB is funded by the European Research

Council (ERC-2009-StG, grant agreement no. 243149). PA holds a research

professorship grant from the Novo Nordic Foundation. Funding agencies had no hand

in study design, execution, data analysis, data interpretation or the writing of this

paper.

Competing interests

None declared.

Contributors

LH, ABF, CSB and PA designed and planned the study. LH, ABF, MF and AR

supervised data collection and data entry. LH carried out the statistical analysis with

help from ABF, CSB and PA. LH wrote the first draft of the paper and holds the

primary responsibility for its final content. All authors contributed to and approved

the final manuscript.

Data sharing statement

No additional data available.

Page 11 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Table 1. Vaccination coverage for 12-month-old children and 24-month-old children in rural Guinea-Bissau.

Region

Vaccination Coverage

12-month-old children 24-month-old children

n

Fully vaccinated

excl. OPV0 (n(%))

n

Fully vaccinated

excl. OPV0 n(%))

All 5806 2903 (50) 2446 3792 (65)

Oio 908 261 (29) 212 546 (39)

Biombo 898 406 (45) 261 482 (54)

Gabu 1071 695 (65) 510 614 (83)

Cacheu 457 265 (58) 322 224 (70)

Bafata 855 552 (65) 492 422 (86)

Quinara 493 219 (44) 431 252 (58)

Tombali 440 115 (26) 365 178 (49)

Bijagós 142 88 (62) 113 96 (85)

Bolama 124 75 (60) 94 75 (79)

São Domingos 418 227 (54) 333 217 (65)

p-value for same

coverage in all

regions <0.0001 <0.0001

OPV, oral polio vaccine.

Page 12 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

on 2 February 2019 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2012-001509 on 19 November 2012. Downloaded from



13

Table 2. Vaccine coverage among 12-month-old children in rural Guinea-Bissau.

Region n

Vaccine (n(%))

BCG OPV0 OPV1 OPV2 OPV3 DTP1 DTP2 DTP3 MV

All 5806 5146 (89) 781 (13) 5311 (91) 4756 (82) 3836 (66) 5338 (92) 4806 (83) 3968 (68) 3567 (61)

Oio 908 740 (82) 39 (4) 787 (87) 623 (69) 393 (43) 798 (88) 669 (74) 483 (53) 445 (49)

Biombo 898 765 (85) 202 (22) 793 (88) 696 (78) 554 (62) 801 (89) 701 (78) 558 (62) 494 (55)

Gabu 1071 999 (93) 136 (13) 1020 (95) 984 (92) 870 (81) 1023 (96) 987 (92) 882 (82) 792 (74)

Cacheu 457 425 (93) 92 (20) 440 (96) 412 (90) 345 (75) 442 (97) 415 (91) 359 (79) 314 (69)

Bafata 855 803 (94) 97 (11) 816 (95) 783 (92) 689 (81) 818 (96) 785 (92) 704 (82) 612 (72)

Quinara 493 422 (86) 30 (6) 447 (91) 387 (79) 307 (62) 447 (91) 385 (78) 301 (61) 261 (53)

Tombali 440 348 (79) 56 (13) 357 (81) 275 (63) 160 (36) 357 (81) 270 (61) 161 (37) 204 (46)

Bijagós 142 137 (96) 46 (32) 135 (95) 132 (93) 119 (84) 135 (95) 132 (93) 120 (85) 95 (67)

Bolama 124 119 (96) 55 (44) 121 (98) 118 (95) 104 (84) 121 (98) 116 (94) 104 (84) 80 (65)

São Domingos 418 388 (93) 28 (7) 395 (95) 346 (83) 295 (71) 396 (95) 346 (83) 296 (71) 270 (65)

p-value for

same coverage

in all regions

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

BCG, bacillus Calmette-Guérin vaccine, OPV, oral polio vaccine, DTP, diphtheria+tetanus+pertussis vaccine, MV, measles vaccine.

Page 13 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




14

Table 3. Frequency of out-of-sequence vaccinations among 12-month-old and 24-month old children in rural Guinea-Bissau.

Region

BCG vaccinated children MV vaccinated children

12-month-old children 24-month old children 12-month-old children 24-month old children

n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%))

All 5146 2520 (49) 2770 (54) 3455 1783 (52) 2020 (58) 3567 924 (26) 1012 (28) 2981 875 (29) 1056 (35)

Oio 740 415 (56) 466 (63) 447 256 (57) 307 (69) 445 180 (40) 196 (44) 355 163 (46) 179 (50)

Biombo 765 364 (48) 397 (52) 409 223 (55) 250 (61) 494 120 (24) 133 (27) 327 110 (34) 123 (38)

Gabu 999 526 (53) 552 (55) 587 323 (55) 344 (59) 792 201 (25) 229 (29) 559 144 (26) 178 (32)

Cacheu 425 133 (31) 169 (40) 305 92 (30) 125 (41) 314 43 (14) 45 (14) 259 37 (14) 47 (18)

Bafata 803 353 (44) 412 (51) 477 221 (46) 258 (54) 612 128 (21) 132 (22) 460 97 (21) 106 (23)

Quinara 422 204 (48) 214 (51) 391 193 (49) 207 (53) 261 66 (25) 73 (28) 295 75 (25) 98 (33)

Tombali 348 258 (74) 263 (76) 327 252 (77) 275 (84) 204 110 (54) 124 (61) 281 159 (57) 214 (76)

Bijagós 137 28 (20) 39 (28) 107 24 (22) 29 (27) 95 7 (7) 8 (8) 102 11 (11) 13 (13)

Bolama 119 37 (31) 42 (35) 85 44 (52) 52 (61) 80 9 (11) 9 (11) 80 23 (29) 25 (31)

São Domingos 388 202 (52) 216 (56) 320 155 (48) 173 (54) 270 60 (22) 63 (23) 263 56 (21) 73 (28)

p-value for

same frequency

in all regions <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

BCG, bacillus Calmette-Guérin vaccine, DTP, diphtheria+tetanus+pertussis vaccine, MV, measles vaccine

Page 14 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




REFERENCES

(1) UNICEF. Expanding Immunization Coverage. 2011. Available from: http://www.unicef.org/immunization/index_coverage.html.

(2) UNICEF, WHO. Immunization summary - A statistical reference containing data through 2010. UNICEF, WHO 2012.

(3) WHO. GIVS : global immunization vision and strategy : 2006-2015. WHO 2005.

(4) WHO. WHO vaccine-preventable diseases : monitoring system : 2010 global summary. WHO 2010.

(5) WHO. Immunization coverage cluster survey: Reference manual. WHO 2005.

(6) Veirum JE, Sodemann M, Biai S, et al. Routine vaccinations associated with divergent effects on female and male mortality at the paediatric ward in Bissau, Guinea-Bissau. Vaccine 2005;23:1197-1204.

(7) Aaby P, Samb B, Simondon F, et al. Non-specific beneficial effect of measles immunisation: analysis of mortality studies from developing countries. BMJ 1995;311:481-485.

(8) Kabir Z, Long J, Reddaiah VP, et al. Non-specific effect of measles vaccination on overall child mortality in an area of rural India with high vaccination coverage: a population-based case-control study. Bull World Health Organ 2003;81:244-250.

(9) Aaby P, Martins C, Bale C, et al. Sex differences in the effect of vaccines on the risk of hospitalization due to measles in Guinea-bissau. Pediatr Infect Dis J 2010;29:324-328.

(10) Kristensen I, Aaby P, Jensen H. Routine vaccinations and child survival: follow up study in Guinea-Bissau, West Africa. BMJ 2000;321:1435-1438.

(11) Aaby P, Jensen H, Gomes J, et al. The introduction of diphtheria-tetanus-pertussis vaccine and child mortality in rural Guinea-Bissau: an observational study. Int J Epidemiol 2004;33:374-380.

(12) Agergaard J, Nante E, Poulstrup G, et al. Diphtheria-tetanus-pertussis vaccine administered simultaneously with measles vaccine is associated with increased morbidity and poor growth in girls. A randomised trial from Guinea-Bissau. Vaccine 2011;29:487-500.

(13) Aaby P, Jensen H, Rodrigues A, et al. Divergent female-male mortality ratios associated with different routine vaccinations among female-male twin pairs. Int J Epidemiol 2004;33:367-373.

(14) Aaby P, Vessari H, Nielsen J, et al. Sex differential effects of routine immunizations and childhood survival in rural Malawi. Pediatr Infect Dis J 2006;25:721-727.

(15) Aaby P, Jensen H, Walraven G. Age-specific changes in the female-male mortality ratio related to the pattern of vaccinations: an observational study from rural Gambia. Vaccine 2006;24:4701-4708.

(16) Aaby P, Benn CS, Nielsen J, et al. Testing the hypothesis that diphtheria-tetanus-pertussis vaccine has negative non-specific and sex-differential effects on child survival in high-mortality countries. BMJ Open 2012. In press.

Page 15 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



16

(17) Aaby P, Ravn H, Roth A, et al. Early diphtheria-tetanus-pertussis vaccination associated with higher female mortality and no difference in male mortality in a cohort of low birthweight children: an observational study within a randomised trial. Arch Dis Child 2012. In press.

(18) Aaby P, Jensen H, Samb B, et al. Differences in female-male mortality after high-titre measles vaccine and association with subsequent vaccination with diphtheria-tetanus-pertussis and inactivated poliovirus: reanalysis of West African studies. Lancet 2003;361:2183-2188.

(19) Biai S, Rodrigues A, Nielsen J, et al. Vaccination status and sequence of vaccinations as risk factors for hospitalisation among outpatients in a high mortality country. Vaccine 2011;29:3662-3669.

(20) Aaby P, Ibrahim SA, Libman MD, et al. The sequence of vaccinations and increased female mortality after high-titre measles vaccine: trials from rural Sudan and Kinshasa. Vaccine 2006;24:2764-2771.

(21) Aaby P, Biai S, Veirum JE, et al. DTP with or after measles vaccination is associated with increased in-hospital mortality in Guinea-Bissau. Vaccine 2007;25:1265-1269.

(22) Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003;3:21.

(23) Jahn A, Floyd S, Mwinuka V, et al. Ascertainment of childhood vaccination histories in northern Malawi. Trop Med Int Health 2008;13:129-138.

(24) Clark A, Sanderson C. Timing of children's vaccinations in 45 low-income and middle-income countries: an analysis of survey data. Lancet 2009;373:1543-1549.

(25) WHO, UNICEF. Review of national immunization coverage 1980-2008 - Guinea-Bissau. WHO 2009.

(26) Suarez L, Simpson DM, Smith DR. Errors and correlates in parental recall of child immunizations: effects on vaccination coverage estimates. Pediatrics 1997;99:E3.

(27) Valadez JJ, Weld LH. Maternal recall error of child vaccination status in a developing nation. Am J Public Health 1992;82:120-122.

(28) Luman ET, Worku A, Berhane Y, et al. Comparison of two survey methodologies to assess vaccination coverage. Int J Epidemiol 2007;36:633-641.

(29) Burton A, Monasch R, Lautenbach B, et al. WHO and UNICEF estimates of national infant immunization coverage: methods and processes. Bull World Health Organ 2009;87:535-541.

(30) WHO, UNICEF, and World Bank. State of the world's vaccines and immunization. 3rd ed. WHO 2009.

(31) WHO. World Health Statistics 2011 - Indicator compendium. WHO 2011.

(32) GAVI alliance. Guidelines on country proposals - For support for: New and Underused Vaccines. GAVI alliance 2011.

(33) Arevshatian L, Clements C, Lwanga S, et al. An evaluation of infant immunization in Africa: is a transformation in progress? Bull World Health Organ 2007;85:449-457.

Page 16 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



17

(34) Fisker A, Hornshøj L, Balde I, et al. Vaccination coverage before and after the introduction of pentavalent and yellow fever vaccines in Guinea-Bissau: Decline in the coverage for measles vaccine. Submitted.

(35) Aaby P, Benn CS, Nielsen J, et al. Testing the hypothesis that diphtheria-tetanus-pertussis vaccine has negative non-specific and sex-differential effects on child survival in high-mortality countries. BMJ Open 2012.

Page 17 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 1. The Expanded Program on Immunization schedule for Guinea-Bissau and the age interval for WHO survey visits: / BCG, bacillus Calmette-Guérin vaccine, OPV, oral polio vaccine, DTP,

diphtheria+tetanus+pertussis vaccine, MV, measles vaccine.

86x13mm (300 x 300 DPI)

Page 18 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 2. Flow chart of inclusion. / BHP, Bandim Health Project

94x38mm (300 x 300 DPI)

Page 19 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 3. Fully vaccinated 12-month-old children and 24-month-old children in rural Guinea-Bissau.

67x46mm (300 x 300 DPI)

Page 20 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 4. Vaccine coverage among 12-month-old children and 24-month-old children in rural Guinea-Bissau. / BCG, bacillus Calmette-Guérin vaccine, OPV, oral polio vaccine, DTP, diphtheria+tetanus+pertussis

vaccine, MV, measles vaccine. 65x44mm (300 x 300 DPI)

Page 21 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



STROBE 2007 (v4) checklist of items to be included in reports of observational studies in epidemiology*

Checklist for cohort, case-control, and cross-sectional studies (combined)

Section/Topic Item # Recommendation Reported on page #

Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 2

(b) Provide in the abstract an informative and balanced summary of what was done and what was found 2

Introduction

Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 4-5

Objectives 3 State specific objectives, including any pre-specified hypotheses 5

Methods

Study design 4 Present key elements of study design early in the paper 5

Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data

collection 5

Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of selection of participants. Describe

methods of follow-up

Case-control study—Give the eligibility criteria, and the sources and methods of case ascertainment and control

selection. Give the rationale for the choice of cases and controls

Cross-sectional study—Give the eligibility criteria, and the sources and methods of selection of participants

5-6

(b) Cohort study—For matched studies, give matching criteria and number of exposed and unexposed

Case-control study—For matched studies, give matching criteria and the number of controls per case N/A

Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic

criteria, if applicable 6

Data sources/ measurement 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe

comparability of assessment methods if there is more than one group 6

Bias 9 Describe any efforts to address potential sources of bias 6,9

Study size 10 Explain how the study size was arrived at 7, fig 2

Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen

and why 6-7

Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 7

(b) Describe any methods used to examine subgroups and interactions N/A

(c) Explain how missing data were addressed 6, fig2

(d) Cohort study—If applicable, explain how loss to follow-up was addressed

Case-control study—If applicable, explain how matching of cases and controls was addressed N/A

Page 22 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




Cross-sectional study—If applicable, describe analytical methods taking account of sampling strategy

(e) Describe any sensitivity analyses N/A

Results

Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility,

confirmed eligible, included in the study, completing follow-up, and analysed 7, fig 1-2

(b) Give reasons for non-participation at each stage 6, fig2

(c) Consider use of a flow diagram Fig2

Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and

potential confounders 5-6

(b) Indicate number of participants with missing data for each variable of interest Fig2

(c) Cohort study—Summarise follow-up time (eg, average and total amount) N/A

Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time N/A

Case-control study—Report numbers in each exposure category, or summary measures of exposure -

Cross-sectional study—Report numbers of outcome events or summary measures -

Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95%

confidence interval). Make clear which confounders were adjusted for and why they were included 7-8, tables 1-3

(b) Report category boundaries when continuous variables were categorized N/A

(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period N/A

Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses N/A

Discussion

Key results 18 Summarise key results with reference to study objectives 8

Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction

and magnitude of any potential bias 8-9

Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results

from similar studies, and other relevant evidence 9-10

Generalisability 21 Discuss the generalisability (external validity) of the study results 10

Other information

Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on

which the present article is based 11

*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.

Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE

checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at

http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.

Page 23 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




Vaccination Coverage and Out-of-Sequence Vaccinations in Rural Guinea-Bissau: An Observational Cohort Study

Journal: BMJ Open

Manuscript ID: bmjopen-2012-001509.R1

Article Type: Research

Date Submitted by the Author: 03-Sep-2012

Complete List of Authors: Hornshøj, Linda; Bandim Health Project, INDEPTH NETWORK, Benn, Christine; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut, ; Bandim Health Project, INDEPTH NETWORK, Fernandes, Manuel; Bandim Health Project, INDEPTH NETWORK, Rodrigues, Amabelia; Bandim Health Project, INDEPTH NETWORK, Aaby, Peter; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut, ; Bandim Health Project, INDEPTH

NETWORK, Fisker, Ane; Bandim Health Project; Research Center for Vitamins and Vaccines (CVIVA), Bandim Health Project, Statens Serum Institut,

<b>Primary Subject Heading</b>:

Public health

Secondary Subject Heading: Global health, Infectious diseases, Epidemiology, Health services research

Keywords:

Public health < INFECTIOUS DISEASES, International health services < HEALTH SERVICES ADMINISTRATION & MANAGEMENT, Epidemiology < INFECTIOUS DISEASES, Paediatric infectious disease & immunisation < PAEDIATRICS


BMJ Open on 2 F

ebruary 2019 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2012-001509 on 19 Novem

ber 2012. Dow

nloaded from



Vaccination Coverage and Out-of-Sequence Vaccinations in Rural Guinea-

Bissau: An Observational Cohort Study

Linda Hornshøja, Christine Stabell Benn

a,b, Manuel Fernandes

a, Amabelia Rodrigues

a,

Peter Aabya,b, Ane Bærent Fisker

a,b








Keywords:


Word count:


Page 1 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



2

ABSTRACT

Objective


which in Guinea-Bissau included bacillus Calmette-Guérin vaccine (BCG) at birth, 3


6, 10 and 14 weeks and measles vaccine (MV) at 9 months when this study was

conducted. The WHO assesses coverage by 12 months of age. The sequence of

vaccines may have an effect on child mortality, but is not considered in official

statistics or assessments of programme performance. We assessed vaccination

coverage and frequency of out-of-sequence vaccinations by 12 and 24 months of age.

Design



The Bandim Health Project’s rural Health and Demographic Surveillance site covers

258 randomly selected villages in all regions of Guinea-Bissau. Villages are visited

biannually and vaccination cards inspected to ascertain vaccination status. Between

2003 and 2009 vaccination status by 12 months of age was assessed for 5806 children

aged 12-23 months; vaccination status by 24 months of age was assessed for 3792

children aged 24-35 months.

Outcome measures


Results


EPI vaccinations. Many children received vaccines out-of-sequence: by 12 months of

age 54% of BCG vaccinated children had received DTP with or before BCG and 28%

of measles vaccinated children had received DTP with or after MV. By 24 months of

age the proportion of out-of-sequence vaccinations was 58% and 35%, respectively,

for BCG and MV.

Conclusions


continued to increase beyond 12 month of age. More than half of all children received

vaccinations out-of-sequence. This highlights the need to improve vaccination

services.

Page 2 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



3

ARTICLE SUMMARY

Article focus



- The third dose of diphtheria-tetanus-pertussis vaccine (DTP3) is used as a




consideration.

Key messages



- Limiting vaccination services to children less than 12 months of age will lower

the effective vaccination coverage considerably.




surveillance has many advantages compared with EPI household surveys.

- Though the study was representative for rural Guinea-Bissau, it may not

necessarily be representative of other low-income settings.

Page 3 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



4

INTRODUCTION





countries the core vaccines in the EPI vaccines are bacillus Calmette-Guérin vaccine

(BCG) and oral polio vaccine (OPV) at birth, diphtheria- tetanus-pertussis vaccine

(DTP) and OPV at 6, 10 and 14 weeks and measles vaccine (MV) at 9 months of age

though new vaccines are now being added in many places.[2]




by 2010.[3]





including supplemental activities such as campaigns with OPV and MV. The

administrative data are based on the number of vaccine doses administered to the

estimated target population, most commonly the estimated number of children



children so that all children included have had 12 months to receive all vaccinations,

i.e. being between 12–23 months of age at the time of the survey as shown in figure

1.[4,5]

Studies from Guinea-Bissau and other low-income countries suggest that the routine

vaccinations have effects on mortality beyond that of protecting against the target

disease. These effects are referred to as non-specific effects (NSE). BCG and MV

have beneficial effects in lowering overall child morbidity and mortality,[6-18]

whereas DTP may be associated with increased morbidity and mortality.[7-

9,11,12,17-21] The NSEs depend on the most recent vaccine and giving vaccines out-

Page 4 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



5

of-sequence may therefore affect child mortality. Worryingly several studies have

shown that DTP given together with MV or after MV is associated with increased

child morbidity and mortality.[9-13,19,21-23]




month-old children.

METHODS

Setting







all ten regions. The study population in the HDSS is based on 182 randomly selected

clusters of 100 women of fertile age and their children below five years of age living

in a total of 258 villages. In the cluster areas all new households are registered and in

all households new women and children taking up residence are added. The HDSS

cohort is therefore continuously growing.

Study population





children born from the 1st January 2004 until 31st of December 2005, registered before


were included.

Vaccinations

Page 5 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



6

Guinea-Bissau follows the standard EPI. Children in the present study were to be

vaccinated in accordance with the EPI schedule shown in figure 1, which at the time

included only BCG, OPV0-3, DTP1-3 and MV. The booster MV (MV2) is not part of

the routine vaccination schedule as the MV coverage in Guinea-Bissau is far from the

80% criterion for implementation of MV2.[24] MV campaigns have been conducted

in 2006 and 2009.









MV and children receive a vaccination card free of charge. Children were only

included in the two cohorts for this study if they had reached 12 or 24 months of age

before the BHP nurses began administering vaccines.



vaccination coverage by 12 months of age in children aged 12-23 months at the time

of a home visit. Vaccination status was determined based on the first visit among

children aged 12-23 months in 2005-2009 at which vaccination status could be

assessed by inspecting the vaccination card. If the child did not have a vaccination

card at the time of our visits, it was defined as unvaccinated if the mother confirmed

that the child had never received any vaccines. If we were not able to verify a child’s

vaccinations at any visit in the interval of 12-23 months of age either by 1) seeing the

vaccination card or 2) ascertaining that the child had never received vaccinations, the

child was not included in the analysis. The vaccination status by 24 months of age

was determined using the same method for children visited at 24-35 months of age.

Definitions

Page 6 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



7


at birth (OPV0), because very few children receive OPV0 in the rural areas. Out-of-

sequence vaccinations were 1) BCG administered together with any dose of DTP, 2)

any dose of DTP before BCG, 3) MV with any dose of DTP and 4) any dose of DTP

after MV.



regression with robust variance estimates.[25] Regional coverages were compared

using chi-2-tests. Data was analysed using STATA 11.0.

RESULTS



by 12 months of age (68%) than by 24 months of age (59%), RR=1.15 (95% CI 1.12-

1.18).


By 12 months of age only 50% of children were fully vaccinated if reception of OPV0

was disregarded. If OPV0 was included in the definition of “fully vaccinated” only

9% of 12-month-old children were fully vaccinated. There were large regional

differences: only 26% of children were fully vaccinated in the region of Tombali

compared with 65% in Gabu and Bafata, the regions with highest coverage (table 1).

By 24 months of age the percentage of fully vaccinated children was 65%, a 29%

(95%CI =25-34%) increase. The increase ranged from 20% to 87% in the different

regions (table 1).



with large regional differences (table 2). The drop-out rate between DTP1 and DTP3

was 24% and the dropout rate between BCG and MV was 31%. There was only a

slight increase in the vaccines the children receive early in life when comparing

vaccine coverage at 12 and 24 months of age, but the increase for vaccines given later

in life was much higher. BCG, OPV1 and DTP1 increased by 3%, whereas OPV3

Page 7 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



8

increased by 12%, DTP3 by 11% and MV increased by 28% (figure 3). By 24 months

of age, the drop-out rate between DTP1 and DTP3 was 20% and 13% between BCG

and MV. The DTP3 coverage was higher than the MV coverage by 12 months of age

in all regions, except for Tombali. By 24 months of age the overall MV coverage was

higher than DTP3, but in four out of ten regions (Cacheu, Quinara, Bijagós and

Bolama) the MV coverage was still lower than the DTP3 coverage (data not shown).












DISCUSSION

Main results

There were two main observations. First, only half of all children were fully

vaccinated by 12 months of age and 65% by 24 months of age (table 1, figure 2).

Estimating vaccination coverage only by 12 months of age therefore underestimates

the final coverage considerably. Many children had initiated the vaccination program,

but not received all the EPI vaccines by 12 months of age. This was reflected in high

coverage of vaccines scheduled early in life, with little increased from 12 to 24

months of age, whereas the increase in vaccines scheduled later in infancy was much

higher: MV increased by 28% (25-31%) (figure 3).

Second, a substantial number of children received out-of-sequence vaccinations. More

than half of all BCG vaccinated children received DTP with or before BCG and a

third of MV vaccinated children received DTP with or after MV (table 3). As

Page 8 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



9

vaccination coverage increased among 24-month-old children, so did the frequency of

out-of-sequence vaccinations (tables 1,3).


The HDSS clusters are randomly selected in all regions. The sample size is substantial

and the data obtained is therefore representative of the population in Guinea-Bissau

excluding the urban area of Bissau. The villages have been visited biannually for

years, thus offering the opportunity to verify and correct ambiguous data. We believe

that the continued surveillance contributes to more reliable data than other types of

surveys. For example, in this study, children with previously registered vaccinations

could not be misclassified as unvaccinated at a later visit if the vaccinations card had

been lost. A study from Malawi showed a BCG scar rate of 70% among children

without documents, making it clear how vaccination coverage can be underestimated

if children without vaccination cards are classified as unvaccinated.[26] We also

registered pregnancies and births and could thereby determine date of birth fairly

precisely if there was no written record.






from 45 low- and middle-income countries.[27]




coverage to be: BCG 87%, OPV3 60%, DTP3 59% and MV 71%, with a 26%

DTP1/DTP3 dropout rate.[28] When we compared their results to ours, BCG

coverage did not differ, but the WHO OPV3 and DTP3 coverage was 10-13% (5-

18%) lower and the WHO MV coverage was 16% (11-21%) higher than our estimates

(table 2,[28]). The higher MV coverage may in part be explained by our study not

including data from the urban Bissau. In Bissau the coverage is higher as vaccinations

are more accessible. However, surprisingly higher coverage of all antigens was not

Page 9 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



10

found in the WHO survey. If the WHO survey underestimated the true coverage it

could in part be due to the use of history, i.e. maternal recall, to confirm vaccination

status. The validity of recall is correlated to the number of doses received and the age

of the child. Mothers tend to recall vaccinations more accurately for younger children

with fewer doses and the use of recall seems to underestimate vaccination

coverage.[29,30] In the WHO survey, visits are made to randomly selected houses

according to the survey method.[5] Unlike in our study, there are no follow-up visits

to households in which the vaccination status for a child can not be determined at the

time of the visit. The WHO EPI survey methodology has previously been criticised

for lacking methodological rigour and for introducing bias.[31,32]

Implications

The vaccination coverage estimates used by WHO and United Nations Children’s

Fund (UNICEF) emphasises the coverage by 12 months of age, not taking into

account the timing of the vaccines received or vaccines received after 12 months of

age.[33,34] Also, DTP3 is used by funding agencies to evaluate programme

performance[35] and improvements in DTP3 coverage is often linked to additional

funding.[36] This emphasis on vaccinations given during the first year of life and on

DTP3 rather than MV may have unwarranted consequences. First, there is less focus

on MV[37,38] and many African countries now have a lower coverage of MV than

DTP3.[2] Second, the current donor focus on vaccination status by 12 months of age

may lead to vaccination providers limiting vaccinations only to children less 12

months of age. This has happened in Guinea-Bissau recently, after the completion of

this study.[38] Our data showed that MV coverage increased by 28% during the

second year of life. Not vaccinating children after 12 months of age provides only a 3-

month-window for children to receive routine MV and based on our data it will lead

to a considerably lower MV coverage. Added together the emphasis on early

vaccinations and on DTP3 rather on MV is likely to lead to more children with DTP

as their last vaccination and an increasing number of children not receiving MV. This

may be particularly unfortunate as several studies show that MV lowers overall child

morbidity and mortality[6,8-14,16-18] whereas DTP is associated with increased

morbidity and mortality, especially for girls.[7-9,11,12,17-21]

Page 10 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



11

Currently the evaluation of the vaccination program does not take the frequency of

out-of-sequence vaccinations into account. Several studies have indicated that out-of-

sequence MV and DTP vaccinations have a negative effect on both child morbidity

and mortality[9-13,19,21-23]. The frequency of out-of-sequence vaccinations should

therefore be included as an indicator of the performance of routine vaccination

programmes.

CONCLUSION






vaccinations. We suggest that vaccination coverage should also be reported by 24

months of age to encourage vaccination of older children and that the proportion of

children receiving DTP with or after MV should be the central performance indicator.

The focus on vaccinations to children less than 12 months of age[38] may lead to a

stop of vaccination of older children as experienced recently in Guinea-Bissau. This

will obviously restrain many children from completing their vaccination schedule and

this negative effect will be strongest for MV. This will lower herd immunity to

measles infections and increase the risk of epidemics. Since all studies show that MV

administered as the most recent vaccine has a beneficial effect on child survival[6,8-

14,16-18] it may also have unfortunate consequences that an increasing proportion of

children will have DTP as their most recent vaccination.[7-9,11,12,17-21]

Funding







Page 11 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



12


paper.

Competing interests

None declared.

Contributors






Ethical approval

No ethical approval has been sought for this study as its data is observational and

derived from the Bandim Health Project’s HDSS which has been in place in Guinea-

Bissau since 1978 and is conducted by request from the Guinean Ministry of Health.



Page 12 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from




Region



n

Fully vaccinated

excl. OPV0 (n(%))

n

Fully vaccinated

excl. OPV0 n(%))

All 5806 2903 (50) 2446 3792 (65)

Oio 908 261 (29) 212 546 (39)

Biombo 898 406 (45) 261 482 (54)

Gabu 1071 695 (65) 510 614 (83)

Cacheu 457 265 (58) 322 224 (70)

Bafata 855 552 (65) 492 422 (86)

Quinara 493 219 (44) 431 252 (58)

Tombali 440 115 (26) 365 178 (49)

Bijagós 142 88 (62) 113 96 (85)

Bolama 124 75 (60) 94 75 (79)

São Domingos 418 227 (54) 333 217 (65)

p-value for same

coverage in all

regions <0.0001 <0.0001


Page 13 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




14


Region n

Vaccine (n(%))


All 5806 5146 (89) 781 (13) 5311 (91) 4756 (82) 3836 (66) 5338 (92) 4806 (83) 3968 (68) 3567 (61)

Oio 908 740 (82) 39 (4) 787 (87) 623 (69) 393 (43) 798 (88) 669 (74) 483 (53) 445 (49)

Biombo 898 765 (85) 202 (22) 793 (88) 696 (78) 554 (62) 801 (89) 701 (78) 558 (62) 494 (55)

Gabu 1071 999 (93) 136 (13) 1020 (95) 984 (92) 870 (81) 1023 (96) 987 (92) 882 (82) 792 (74)

Cacheu 457 425 (93) 92 (20) 440 (96) 412 (90) 345 (75) 442 (97) 415 (91) 359 (79) 314 (69)

Bafata 855 803 (94) 97 (11) 816 (95) 783 (92) 689 (81) 818 (96) 785 (92) 704 (82) 612 (72)

Quinara 493 422 (86) 30 (6) 447 (91) 387 (79) 307 (62) 447 (91) 385 (78) 301 (61) 261 (53)

Tombali 440 348 (79) 56 (13) 357 (81) 275 (63) 160 (36) 357 (81) 270 (61) 161 (37) 204 (46)

Bijagós 142 137 (96) 46 (32) 135 (95) 132 (93) 119 (84) 135 (95) 132 (93) 120 (85) 95 (67)

Bolama 124 119 (96) 55 (44) 121 (98) 118 (95) 104 (84) 121 (98) 116 (94) 104 (84) 80 (65)

São Domingos 418 388 (93) 28 (7) 395 (95) 346 (83) 295 (71) 396 (95) 346 (83) 296 (71) 270 (65)

p-value for

same coverage

in all regions

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001


Page 14 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




15


Region



n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%))

All 5146 2520 (49) 2770 (54) 3455 1783 (52) 2020 (58) 3567 924 (26) 1012 (28) 2981 875 (29) 1056 (35)

Oio 740 415 (56) 466 (63) 447 256 (57) 307 (69) 445 180 (40) 196 (44) 355 163 (46) 179 (50)

Biombo 765 364 (48) 397 (52) 409 223 (55) 250 (61) 494 120 (24) 133 (27) 327 110 (34) 123 (38)

Gabu 999 526 (53) 552 (55) 587 323 (55) 344 (59) 792 201 (25) 229 (29) 559 144 (26) 178 (32)

Cacheu 425 133 (31) 169 (40) 305 92 (30) 125 (41) 314 43 (14) 45 (14) 259 37 (14) 47 (18)

Bafata 803 353 (44) 412 (51) 477 221 (46) 258 (54) 612 128 (21) 132 (22) 460 97 (21) 106 (23)

Quinara 422 204 (48) 214 (51) 391 193 (49) 207 (53) 261 66 (25) 73 (28) 295 75 (25) 98 (33)

Tombali 348 258 (74) 263 (76) 327 252 (77) 275 (84) 204 110 (54) 124 (61) 281 159 (57) 214 (76)

Bijagós 137 28 (20) 39 (28) 107 24 (22) 29 (27) 95 7 (7) 8 (8) 102 11 (11) 13 (13)

Bolama 119 37 (31) 42 (35) 85 44 (52) 52 (61) 80 9 (11) 9 (11) 80 23 (29) 25 (31)

São Domingos 388 202 (52) 216 (56) 320 155 (48) 173 (54) 270 60 (22) 63 (23) 263 56 (21) 73 (28)

p-value for

same frequency

in all regions <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001


Page 15 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




REFERENCES


(2) UNICEF, WHO. Immunization summary - A statistical reference containing data through 2010. WHO 2012.












(14) Breiman RF, Streatfield PK, Phelan M, et al. Effect of infant immunisation on childhood mortality in rural Bangladesh: analysis of health and demographic surveillance data. Lancet 2004;364:2204-2211.

(15) El-Zein M, Parent ME, Benedetti A, et al. Does BCG vaccination protect against the development of childhood asthma? A systematic review and meta-analysis of epidemiological studies. Int J Epidemiol 2010;39:469-486.

Page 16 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



17




(19) Aaby P, Benn C, Nielsen J, et al. Testing the hypothesis that diphtheria-tetanus-pertussis vaccine has negative non-specific and sex-differential effects on child survival in high-mortality countries. BMJ Open 2012;2:e000707.

(20) Aaby P, Ravn H, Roth A, et al. Early diphtheria-tetanus-pertussis vaccination associated with higher female mortality and no difference in male mortality in a cohort of low birthweight children: an observational study within a randomised trial. Arch Dis Child 2012;97:685-691.




(24) WHO. Weekly epidemiological record. 2009;35:349-360.








Page 17 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



18


(33) WHO, UNICEF, and World Bank. State of the world's vaccines and immunization. 3rd ed. WHO 2009.




(37) Aaby P, Benn CS. Assessment of childhood immunisation coverage. Lancet 2009;373:1428.

(38) Fisker A, Hornshøj L, Rodrigues A, et al. The new decade of vaccines: Improving DTP-3 coverage and reducing coverage for measles vaccine. An observational study of the introduction of new vaccines in Guinea-Bissau. Submitted 2012.

Page 18 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Vaccination Coverage and Frequency of Out-of-Sequence Vaccinations in Rural

Guinea-Bissau: An Observational Cohort Study

Linda Hornshøja, Christine Stabell Benna,b Ane Bærent Fiskera,b, Manuel Fernandesa,

Amabelia Rodriguesa, Christine Stabell Benn

a,b, Peter Aaby

a,b, Ane Bærent Fisker

a,b








Keywords:


Word count:


Page 19 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



2

ABSTRACT

Objective


which in Guinea-Bissau includeds bacillus Calmette-Guérin vaccine (BCG) at birth, 3


6, 10 and 14 weeks and measles vaccine (MV) at 9 months when this study was

conducted. The WHO assesses coverage by 12 months of age. The sequence of

vaccines may have an effect on child mortality, but is not considered in official

statistics or assessments of programme performance. We assessed vaccination

coverage and frequency of out-of-sequence vaccinations by 12 and 24 months of age.

Design



The Bandim Health Project’s rural Health and Demographic Surveillance site System

in rural Guinea-Bissau was used, it covers 258 randomly selected villages in all

regions of Guinea-Bissau. Villages are visited biannually and vaccination cards

inspected to ascertain vaccination status. Between 2003 and 2009 vaccination status

by 12 months of age was assessed for 5806 children aged 12-23 months; vaccination

status and by 24 months of age was assessed for 3792 children aged 24-35 months.

Outcome measures


Results


EPI vaccinations. Many children received vaccines out-of-sequence: by 12 months of

age 54% of BCG vaccinated children had received DTP with or before BCG and 28%

of measles vaccinated children had received DTP with or after MV. by 12 months of

age, bBy 24 months of age the proportion of out-of-sequence vaccinations was 58%

and 35%, respectively, for BCG and MV.

Conclusions


continued to increase beyond 12 month of age. The current practice of limiting

vaccinations to children below 12 months of age will lower overall coverage

considerably. More than half of all children received vaccinations out-of-sequence.

This highlights the need to improve vaccination services.

Page 20 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



3

ARTICLE SUMMARY

Article focus



- The third dose of diphtheria-tetanus-pertussis vaccine (DTP3) is used as a




considerationaccount.

Key messages



- Limiting vaccination servicess to children less thanbelow 12 months of age

will lower the effectiveoverall vaccination coverage considerably.




surveillance has many advantages compared with EPIother household surveys.

- Though the study was representative for limited to rural Guinea-Bissau, it and

may not necessarily be representative of other low-income settings.

Page 21 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



4

INTRODUCTION





countries the core vaccines in the standard schedule for EPI vaccines are includes

bacillus Calmette-Guérin vaccine (BCG) and oral polio vaccine (OPV) at birth,

diphtheria- tetanus-pertussis vaccine (DTP) and OPV at 6, 10 and 14 weeks and

measles vaccine (MV) at 9 months of age though new vaccines are now being added

in many places.[2]




by 2010.[3]





including supplemental activities such as OPV and MV campaigns with OPV and

MV. The administrative data are based on the number of vaccine doses administered

to the estimated target population, most commonly the estimated number of children



children so that all children included have had 12 months to receive all vaccinations,

(i.e. being between 12–23 months of age at the time of the survey as shown in figure

1).[4,5]

Studies from Guinea-Bissau and other low-income countries suggest that the routine

vaccinations have effects on mortality beyond that of protecting against the target

disease. These effects are referred to as non-specific effects (NSE). BCG and MV

have beneficial effects in lowering overall child morbidity and mortality,[6-18][6-10]

whereas DTP may be associated with increased morbidity and mortality.[7-

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Page 22 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



5

9,11,12,17-21][6,9-17] The NSEs depend on the most recent vaccine and giving

vaccines out-of-sequence may therefore affect child mortality. Worryingly several

studies have shown that DTP given together with MV or after MV is associated with

increased child morbidity and mortality.[9-13,19,21-23][12,18-21]




month-old children.

METHODS

Setting







258 villages in all ten regions. The study population in the HDSS is based on 182

randomly selected clusters of 100 women of fertile age and their children below five

years of age living in a total of 258 villages. In the cluster areas all new households

are registered and in all households new women and children taking up residence are

added. The HDSS cohort is therefore continuously growing.

Study population





children born from the 1st January 2004 until 31

st of December 2005, registered before


were included.

Vaccinations

Page 23 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



6

Guinea-Bissau follows the standard EPI. and cChildren in the present study were to be

vaccinated in accordance with the EPI schedule shown in figure 1, which at the time

included only BCG, OPV0-3, DTP1-3 and MV. The booster MV (MV2) is not part of

the routine vaccination schedule as the MV coverage in Guinea-Bissau is far from the

80% criterion for implementation of MV2.[24] MV campaigns have been conducted

in 2006 and 2009.









MV and children receive a vaccination card free of charge. CIn selection of the two

cohorts for this study, children were only included in the two cohorts for this study if

they had reached 12 or 24 months of age before the BHP nurses began administering

vaccines.



vaccination coverage by 12 months of age in children aged 12-23 months at the time

of a home visits. Vaccination status was determined based on the first visit among

children aged 12-23 months in 2005-2009, at which vaccination status could be

assessed by inspecting the vaccination card. If the child did not have a vaccination

card at the time of our visits, it was defined as unvaccinated if the mother confirmed

that the child had never received any vaccines. If we were not able to verify a child’s

vaccinations at any visit in the interval of 12-23 months of age either by 1) by seeing

the vaccination card or 2) ascertaining that the child had never received vaccinations,

the child was not included in the analysis. The vaccination status by 24 months of age

was determined using the same method for children visited at 24-35 months of age.

Definitions

Page 24 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



7


at birth (OPV0), because very few children receive d OPV0 in the rural areas.

Out-of-sequence vaccinations were 1) BCG administered together with any dose of

DTP, 2) any dose of DTP with or before BCG, 3) MV with any dose of DTP and 4)

any dose of DTP with or after MV.



regression with robust variance estimates.[25][22] Regional Ccoverages werefor

different regions was compared using chi-2-tests. Data was analysed using STATA

11.0.

RESULTS



by 12 months of age (68%) than by 24 months of age (59%), RR=1.15 (95% CI

=1.12-1.18).


By 12 months of age only 50% of children were fully vaccinated if reception whether

or not a child had received of OPV0 was disregarded. If OPV0 was included in the

definition of “fully vaccinated” only 9% of 12-month-old children were fully

vaccinated. There were large regional differences: only 26% of children were fully

vaccinated in the region of Tombali compared with 65% in Gabu and Bafata, the

regions with highest coverage (table 1). By 24 months of age the percentage of fully

vaccinated children was 65%, a 29% (95%CI =25-34%) increase. The increase ranged

from 20% to 87% in the different regions (table 1, figure 3).



with large regional differences (table 2). The drop-out rate between DTP1 and DTP3

was 24% and the dropout rate between BCG and MV was 31%. There was only a

slight increase in the vaccines the children receive early in life when comparing

vaccine coverage at 12 and 24 months of age, but the increase for vaccines given later

Page 25 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



8

in life was much higher. BCG, OPV1 and DTP1 increased by 3%, whereas OPV3

increased by 12%, DTP3 by 11% and MV increased by 28% (figure 34). By 24

months of age, the drop-out rate between DTP1 and DTP3 was 20% and 13%

between BCG and MV. The DTP3 coverage was higher than the MV coverage by 12

months of age in all regions, except for Tombali. By 24 months of age the overall MV

coverage was higher than DTP3, but in four out of ten regions (Cacheu, Quinara,

Bijagós and Bolama) the MV coverage was still lower than the DTP3 coverage (data

not shown).












DISCUSSION

Main results

There were two main observations. FirstVaccinations are important in lowering

overall child mortality. In our study, only half of all children were fully vaccinated by

12 months of age and 65% by 24 months of age (table 1, figure 2). Only eEstimating

vaccination coverage only amongby 12- months of age-old children therefore

underestimates the final coverage considerably. Many children had initiated the

vaccination program, but not received all the EPI vaccines by 12 months of age. This

was reflected in high coverage of vaccines scheduled early in life, with little increased

from 12 to 24 months of age, whereas the increase in vaccines scheduled later in

infancy was much higher: MV increased by 28% (25-31%) (figure 34). Regional

Page 26 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



9

differences in vaccination coverage and frequency of out-of-sequence vaccinations

were large (tables 1,3).

Second, Aa substantial number of children received out-of-sequence vaccinations.

More than half of all BCG vaccinated children received DTP with or before BCG and

a third of MV vaccinated children received DTP with or after MV (table 3). As

vaccination coverage increased among 24-month-old children, so did the frequency of

out-of-sequence vaccinations (tables 1,3).


The HDSS clusters are randomly selected in all regions. Tand the sample size is

substantial and the data obtained is therefore representative of the population in

Guinea-Bissau excluding the urban area of Bissau. The villages have been visited

biannually for years, thus offering the opportunity to verify and correct ambiguous

data. We believe that the continued surveillance contributes to more reliable data than

other types of surveys. For example, in this study, children with previously registered

vaccinations could not be misclassified as unvaccinated at a later visit if the

vaccinations card had been lost. A study from Malawi showed a BCG scar rate of

70% among children without documents, making it clear how vaccination coverage

can be underestimated if children without vaccination cards are simply classified as

unvaccinated.[26][23] We also registered pregnancies and births and could thereby

determine date of birth fairly precisely if there was no written record.






from 45 low- and middle-income countries.[27][24]




Field Code Changed

Field Code Changed

Page 27 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



10

coverage to be: BCG 87%, OPV3 60%, DTP3 59% and MV 71%, with a 26%

DTP1/DTP3 dropout rate.[28][25] When we compared their results to ours, BCG

coverage did not differ, but the WHO OPV3 and DTP3 coverage was 10-13% (5-

18%) lower and the WHO MV coverage was 16% (11-21%) higher than our estimates

(table 2,[28][25]). The higher MV coverage may in part be explained by our study not

including data from the urban Bissau. In Bissau the coverage is higher as vaccinations

are more accessible. However, surprisingly higher coverage of all antigens was not

found in the WHO survey. If the WHO survey underestimated the true coverage it

could in part be due to the use of history, i.e. maternal recall, to confirm vaccination

status. The validity of recall is correlated to the number of doses received and the age

of the child. Mothers tend to recall vaccinations more accurately for younger children

with fewer doses and the use of recall seems to underestimate vaccination

coverage.[29,30][26,27] In the WHO survey, visits are made to randomly selected

houses according to the survey method.[5] Unlike in our study, there are no follow-up

visits to households in which the vaccination status for a child can not be determined

at the time of the visit. The WHO EPI survey methodology has previously been

criticised for lacking methodological rigour and for introducing bias.[31,32][28,29]

Implications

At presentThe vaccination coverage estimates used by WHO and United Nations

Children’s Fund (UNICEF) emphasises the coverage of DTP3 by 12 months of age,

not taking into account the timing of the vaccines received or vaccines received at a

later ageafter 12 months of age.[33,34][30,31] This has two consequences, which

may have large implications. First, the emphasis on DTP3 leads to more focus on

Also, DTP3 and less focus on measles vaccine. It is often used by funding agencies

such as the Global Alliance for Vaccines and Immunizations (GAVI) to evaluate

programme performance[35][32] and improvements in DTP3 coverage of vaccination

coverage is often linked to additional funding.[36][33] This emphasis on vaccinations

given during the first year of life and on DTP3 rather than MV may have unwarranted

consequences. First, there is less focus on MV[37,38] andAs a result many African

countries now have a lower coverage of MV than DTP3.[2] Second, the current donor

focus on vaccination status by 12 months of age may lead to vaccination providers

limiting vaccinations only to children less 12 months of age. This has happened in

Guinea-Bissau recently, after the completion of this study.[38] This is unfortunate.

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Field Code Changed

Page 28 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



11

Several studies show that MV lowers overall child morbidity and mortality[6-10,13-

15] and several studies suggest that DTP is associated with increased female

morbidity and mortality.[6,9-15]

Second, oOur data was from before 2007, and it showed that MV coverage increased

by 28% during the second year of life. However, the current donor focus on

vaccination status by 12 months of age leads to many countries, including Guinea-

Bissau, have now stopped vaccinating children above 12 months of age, because they

do not count in the statistics. The current policy not to vaccinate children above 12

months of age gives children only a 3-month-time-window to be measles vaccinated.

This will undoubtedly lead to a lower MV coverage and an increasing number of

children being susceptible to measles infectionNot vaccinating children after 12

months of age provides only a 3-month-window for children to receive routine MV

and based on our data it will lead to a considerably lower MV coverage. Added

together the emphasis on early vaccinations and on DTP3 rather on MV is likely to

lead to more children with DTP as their last vaccination and an increasing number of

children not receiving MV. This may be particularly unfortunate as several studies

show that MV lowers overall child morbidity and mortality[6,8-14,16-18] whereas

DTP is associated with increased morbidity and mortality, especially for girls.[7-

9,11,12,17-21]

Currently the evaluation of the vaccination program does not take the frequency of

out-of-sequence vaccinations into account. Several studies have indicated that out-of-

sequence MV and DTP vaccinations have a negative effect on both child morbidity

and mortality[9-13,19,21-23]. The frequency of out-of-sequence vaccinations should

therefore be included as an indicator of the performance of routine vaccination

programmes.

.

CONCLUSION






Page 29 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



12

vaccinations. As several studies suggest that out-of-sequence vaccinations have an

impact on overall survival[18-21] such deviations from the schedule need to be taken

into account and should be reported as performance indicators for vaccination

programmes in likening with specific antigen coverage.

We suggest that vaccination coverage should also be reported by 24 months of age to

encourage vaccination of older children and that the proportion of children receiving

DTP with or after MV should be the central performance indicator.

The focus current drive to limit on vaccinations to children less thanbelow 12 months

of age[38][34] may lead to a stop of vaccination of older children as experienced

recently in Guinea-Bissau. This will obviously restrainprevent many children from

completing their vaccination schedule and this negative effect will be strongest for

MVmeasles vaccination. This will lower heard immunity to measles infections and

increase the risk of epidemics. Since all studies show that MVmeasles vaccination

administered as the most recent vaccine has a beneficial effect on child survival[6,8-

14,16-18][6-8,10,14] it may also will have unfortunate consequences that an

increasing proportion of children will have DTP as their most recent vaccination.[7-

9,11,12,17-21] [10,12,15,17,35]

Funding








paper.

Competing interests

None declared.

Contributors

Page 30 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



13






Ethical approval

No ethical approval has been sought for this study as its data is observational and

derived from the Bandim Health Project’s HDSS which has been in place in Guinea-

Bissau since 1978 and is conducted by request from the Guinean Ministry of Health.



Page 31 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from




Region



n

Fully vaccinated

excl. OPV0 (n(%))

n

Fully vaccinated

excl. OPV0 n(%))

All 5806 2903 (50) 2446 3792 (65)

Oio 908 261 (29) 212 546 (39)

Biombo 898 406 (45) 261 482 (54)

Gabu 1071 695 (65) 510 614 (83)

Cacheu 457 265 (58) 322 224 (70)

Bafata 855 552 (65) 492 422 (86)

Quinara 493 219 (44) 431 252 (58)

Tombali 440 115 (26) 365 178 (49)

Bijagós 142 88 (62) 113 96 (85)

Bolama 124 75 (60) 94 75 (79)

São Domingos 418 227 (54) 333 217 (65)

p-value for same

coverage in all

regions <0.0001 <0.0001


Page 32 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




15


Region n

Vaccine (n(%))


All 5806 5146 (89) 781 (13) 5311 (91) 4756 (82) 3836 (66) 5338 (92) 4806 (83) 3968 (68) 3567 (61)

Oio 908 740 (82) 39 (4) 787 (87) 623 (69) 393 (43) 798 (88) 669 (74) 483 (53) 445 (49)

Biombo 898 765 (85) 202 (22) 793 (88) 696 (78) 554 (62) 801 (89) 701 (78) 558 (62) 494 (55)

Gabu 1071 999 (93) 136 (13) 1020 (95) 984 (92) 870 (81) 1023 (96) 987 (92) 882 (82) 792 (74)

Cacheu 457 425 (93) 92 (20) 440 (96) 412 (90) 345 (75) 442 (97) 415 (91) 359 (79) 314 (69)

Bafata 855 803 (94) 97 (11) 816 (95) 783 (92) 689 (81) 818 (96) 785 (92) 704 (82) 612 (72)

Quinara 493 422 (86) 30 (6) 447 (91) 387 (79) 307 (62) 447 (91) 385 (78) 301 (61) 261 (53)

Tombali 440 348 (79) 56 (13) 357 (81) 275 (63) 160 (36) 357 (81) 270 (61) 161 (37) 204 (46)

Bijagós 142 137 (96) 46 (32) 135 (95) 132 (93) 119 (84) 135 (95) 132 (93) 120 (85) 95 (67)

Bolama 124 119 (96) 55 (44) 121 (98) 118 (95) 104 (84) 121 (98) 116 (94) 104 (84) 80 (65)

São Domingos 418 388 (93) 28 (7) 395 (95) 346 (83) 295 (71) 396 (95) 346 (83) 296 (71) 270 (65)

p-value for

same coverage

in all regions

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001


Page 33 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




16


Region



n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with BCG

(n(%))

Received DTP

with or before

BCG (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%)) n

Received DTP

with MV

(n(%))

Received DTP

with or after

MV (n(%))

All 5146 2520 (49) 2770 (54) 3455 1783 (52) 2020 (58) 3567 924 (26) 1012 (28) 2981 875 (29) 1056 (35)

Oio 740 415 (56) 466 (63) 447 256 (57) 307 (69) 445 180 (40) 196 (44) 355 163 (46) 179 (50)

Biombo 765 364 (48) 397 (52) 409 223 (55) 250 (61) 494 120 (24) 133 (27) 327 110 (34) 123 (38)

Gabu 999 526 (53) 552 (55) 587 323 (55) 344 (59) 792 201 (25) 229 (29) 559 144 (26) 178 (32)

Cacheu 425 133 (31) 169 (40) 305 92 (30) 125 (41) 314 43 (14) 45 (14) 259 37 (14) 47 (18)

Bafata 803 353 (44) 412 (51) 477 221 (46) 258 (54) 612 128 (21) 132 (22) 460 97 (21) 106 (23)

Quinara 422 204 (48) 214 (51) 391 193 (49) 207 (53) 261 66 (25) 73 (28) 295 75 (25) 98 (33)

Tombali 348 258 (74) 263 (76) 327 252 (77) 275 (84) 204 110 (54) 124 (61) 281 159 (57) 214 (76)

Bijagós 137 28 (20) 39 (28) 107 24 (22) 29 (27) 95 7 (7) 8 (8) 102 11 (11) 13 (13)

Bolama 119 37 (31) 42 (35) 85 44 (52) 52 (61) 80 9 (11) 9 (11) 80 23 (29) 25 (31)

São Domingos 388 202 (52) 216 (56) 320 155 (48) 173 (54) 270 60 (22) 63 (23) 263 56 (21) 73 (28)

p-value for

same frequency

in all regions <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001


Page 34 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




REFERENCES


(2) UNICEF, WHO. Immunization summary - A statistical reference containing data through 2010. WHO 2012.












(14) Breiman RF, Streatfield PK, Phelan M, et al. Effect of infant immunisation on childhood mortality in rural Bangladesh: analysis of health and demographic surveillance data. Lancet 2004;364:2204-2211.

(15) El-Zein M, Parent ME, Benedetti A, et al. Does BCG vaccination protect against the development of childhood asthma? A systematic review and meta-analysis of epidemiological studies. Int J Epidemiol 2010;39:469-486.

Formatted: Space After: 12 pt

Page 35 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



18




(19) Aaby P, Benn C, Nielsen J, et al. Testing the hypothesis that diphtheria-tetanus-pertussis vaccine has negative non-specific and sex-differential effects on child survival in high-mortality countries. BMJ Open 2012;2:e000707.

(20) Aaby P, Ravn H, Roth A, et al. Early diphtheria-tetanus-pertussis vaccination associated with higher female mortality and no difference in male mortality in a cohort of low birthweight children: an observational study within a randomised trial. Arch Dis Child 2012;97:685-691.




(24) WHO. Weekly epidemiological record. 2009;35:349-360.








Page 36 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



19


(33) WHO, UNICEF, and World Bank. State of the world's vaccines and immunization. 3rd ed.

WHO 2009.




(37) Aaby P, Benn CS. Assessment of childhood immunisation coverage. Lancet 2009;373:1428.

(38) Fisker A, Hornshøj L, Rodrigues A, et al. The new decade of vaccines: Improving DTP-3 coverage and reducing coverage for measles vaccine. An observational study of the introduction of new vaccines in Guinea-Bissau. Submitted 2012.

Formatted: Tab stops: 0.5", Right + 0.63",

Left + Not at 0.38"

Page 37 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 1. The Expanded Program on Immunization schedule for Guinea-Bissau and the age interval for WHO survey visits: / BCG, bacillus Calmette-Guérin vaccine, OPV, oral polio vaccine, DTP,

diphtheria+tetanus+pertussis vaccine, MV, measles vaccine.

167x90mm (300 x 300 DPI)

Page 38 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 2. Flow chart of inclusion. / BHP, Bandim Health Project

132x90mm (300 x 300 DPI)

Page 39 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



Figure 3. Vaccine coverage among 12-month-old children and 24-month-old children in rural Guinea-Bissau./BCG, bacillus Calmette-Guérin vaccine, OPV, oral polio vaccine, DTP, diphtheria+tetanus+pertussis

vaccine, MV, measles vaccine.

221x90mm (300 x 300 DPI)

Page 40 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


ovember 2012. D

ownloaded from



STROBE 2007 (v4) checklist of items to be included in reports of observational studies in epidemiology*

Checklist for cohort, case-control, and cross-sectional studies (combined)

Section/Topic Item # Recommendation Reported on page #

Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 1-2

(b) Provide in the abstract an informative and balanced summary of what was done and what was found 2

Introduction

Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 4-5

Objectives 3 State specific objectives, including any pre-specified hypotheses 5

Methods

Study design 4 Present key elements of study design early in the paper 5

Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data

collection 5

Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of selection of participants. Describe

methods of follow-up

Case-control study—Give the eligibility criteria, and the sources and methods of case ascertainment and control

selection. Give the rationale for the choice of cases and controls

Cross-sectional study—Give the eligibility criteria, and the sources and methods of selection of participants

5-6

(b) Cohort study—For matched studies, give matching criteria and number of exposed and unexposed

Case-control study—For matched studies, give matching criteria and the number of controls per case N/A

Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic

criteria, if applicable 7

Data sources/ measurement 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe

comparability of assessment methods if there is more than one group 6-7

Bias 9 Describe any efforts to address potential sources of bias 6,9

Study size 10 Explain how the study size was arrived at 5-6, fig 2

Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen

and why 5-6

Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 7

(b) Describe any methods used to examine subgroups and interactions N/A

(c) Explain how missing data were addressed 6, fig2

(d) Cohort study—If applicable, explain how loss to follow-up was addressed

Case-control study—If applicable, explain how matching of cases and controls was addressed N/A

Page 41 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960




Cross-sectional study—If applicable, describe analytical methods taking account of sampling strategy

(e) Describe any sensitivity analyses N/A

Results

Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility,

confirmed eligible, included in the study, completing follow-up, and analysed 7, fig 1-2

(b) Give reasons for non-participation at each stage 6, fig2

(c) Consider use of a flow diagram fig2

Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and

potential confounders 5-6

(b) Indicate number of participants with missing data for each variable of interest fig2

(c) Cohort study—Summarise follow-up time (eg, average and total amount) N/A

Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time N/A

Case-control study—Report numbers in each exposure category, or summary measures of exposure -

Cross-sectional study—Report numbers of outcome events or summary measures -

Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95%

confidence interval). Make clear which confounders were adjusted for and why they were included 7-8, tables 1-3

(b) Report category boundaries when continuous variables were categorized N/A

(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period N/A

Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses N/A

Discussion

Key results 18 Summarise key results with reference to study objectives 8-9

Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction

and magnitude of any potential bias 9

Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results

from similar studies, and other relevant evidence 9-11

Generalisability 21 Discuss the generalisability (external validity) of the study results 10-11

Other information

Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on

which the present article is based 11

*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.

Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE

checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at

http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.

Page 42 of 42


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



Documents

BMJ Open Hornshøja, Ane Bærent Fiskera,b, Manuel Fernandesa, Amabelia Rodriguesa, Christine Stabell Benna,b, Peter Aabya,b a) Bandim Health Project, INDEPTH Network, Apartado 861,