IMPROVING EARLY DIAGNOSIS OF LUNG CANCER: THE IMPACT OF

REGIONAL AND NATIONAL AWARENESS CAMPAIGNS

Lucy Ironmonger, Ella Ohuma, Michael D Peake, Abigail Bentley, Monika Ciurej & Nick Ormiston-Smith June 2014

Contents

OUTLINE:

– Background and objective

– Methods

– Key results

– Discussion

• Conclusions

• Limitations

• Further work

2

BACKGROUND

1

Background: Lung Cancer in England

– Most common cause of cancer death

– Around 28,000 deaths per year

– Long-term survival is poor

& little improvement in recent years

4

5 year age-standardised net survival

Background: Lung Cancer in England

– Survival is worse than in many other countries

Coleman et al. 2011

Reasons:

• Differences in quality/access to treatment

• Later diagnosis Walters et al. 2013 & Holmberg et al. 2010

To improve survival rates in England > more can be done to improve earlier stage at diagnosis

5

0%

5%

10%

15%

20%

Australia Canada Denmark Norway Sweden England

Rel

ativ

e su

rviv

al 5 year age-standardised relative survival

Background: Public Awareness Campaigns

– Small community-based intervention in Doncaster Athey et al. 2012

– Department of Health funded local pilot interventions

– Aim: raise awareness of persistent cough as a lung cancer symptom & encourage GP visits

– Regional pilot: East & West Midlands

Oct– Nov 2011

– National campaign: England

May–June 2012

6

Cancer Research UK were commissioned to carry out the campaign evaluations

OBJECTIVE:

Evaluate the impact of the regional and

national lung cancer campaigns

METHODS

2

9

Public awareness GP

presentations Urgent GP referrals GP-referred

chest x-rays Diagnosis

Stage

Treatment

Survival

Methods

Methods – Data collection

– Data analysis: differences in metrics between campaign months & a pre-campaign period were analysed

– Compared to controls when possible

10

Metric Data provider Type of data

• Public awareness of lung cancer symptoms

TNS BMRB (A market research organisation)

Sample of population

• Patients GPs recorded as presenting with a cough

Mayden (Healthcare IT specialists)

Sample of GP practices

• Urgent GP referrals for suspected lung cancer

East Midlands KIT from the Cancer Waiting Times database (NHS England)

National dataset

• Chest x-rays Diagnostic Imaging Dataset team (NHS England)

National dataset

• Diagnosis, staging & treatment National Lung Cancer Audit team (Health & Social Care Info Centre)

National dataset •One-year survival

(available for regional pilot only)

KEY RESULTS

3

Key Results

– ↑ public awareness of lung cancer symptoms

e.g. Prompted awareness of cough for 3+ weeks

12

19% 18%

34% 33%

22% 20%

0%

5%

10%

15%

20%

25%

30%

35%

40%

Perc

ent

of

resp

on

den

ts

Pre-campaign

Post-Campaign

Pilot area Control area National campaign Regional campaign

* *

* = significantly different to pre-campaign

– ↑ people with a cough going to see their GP

– ↑ urgent GP referrals for suspected lung cancer

– ↑ GP-referred chest x-rays

– ↑ number of lung cancers diagnosed

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

Regional National Regional National Regional National Regional National

GP attendance Urgent GP referrals GP-referred chest x-rays Cases diagnosed

Control area /time period

Campaign area /time period

* *

*

*

* *

*

*

*

* *

†

†

Key Results

13

Not available

*

†

Percent change between campaign/control months and previous time period:

*

† †

*= significant change †= significant difference to control

Key Results – Evidence of a stage shift at diagnosis

• ↑ proportion of non small cell lung cancers diagnosed at stage I (p<0.001)

• ↓ proportion diagnosed at stage IV (p<0.001)

• No change for the control period (p=0.404 & p=0.244)

– Increase in surgical resection rate

• 2.3 percentage point increase in proportion receiving resection (p<0.001)

• No evidence of increase for the control (p=0.425)

14

-4

-3

-2

-1

0

1

2

3

4

Stage I Stage II Stage III Stage IV

Perc

enta

ge p

oin

t ch

ange

b

etw

een

20

11

& 2

01

2

Control (Feb-April)

Campaign (May-July)

*

* * = significant change

Key Results

15

– One-year age-standardised crude survival

Regional campaign:

• Increase of 4 percentage points in the pilot area (p=0.024)

• Increase of 2 percentage points in the control area (p=0.034)

But no evidence of a difference in magnitude (p=0.425)

DISCUSSION

4

Conclusions

– “Whole system” response: increases for metrics across the pathway

– First data to show a shift in stage at diagnosis following a lung cancer awareness campaign

– Surgical resection = treatment most likely to improve long-term survival

Expect an increase in proportion receiving surgery will lead to a reduction in lung cancer mortality rates

17

Limitations

– No strict control available

– Lead-time bias

• An increase in survival might be due to patients diagnosed earlier but their overall survival time will be the same

18

Further work

– Impact on emergency presentations

– Impact on long-term survival and mortality rates

– Cost-effectiveness

– Negative & other positive impacts

– Longevity of impact

– Effect of repeating the campaign

19

– Continued evaluation of other Be Clear on Cancer campaigns

– Comparisons across campaigns

Further work

20

Acknowledgments

We would like to thank the following organisations:

• Cancer Research UK

• Department of Health

• Diagnostic Imaging Dataset team

• East Midlands Knowledge & Intelligence Team

• Mayden

• National Lung Cancer Audit team and the Health & Social Care Information Centre

•NHS England

• Public Health England

• TNS BMRB

Since this presentation, these results have been published in the

British Journal of Cancer:

Ironmonger L, Ohuma E, Ormiston-Smith N, Gildea C, Thomson CS, Peake MD. An evaluation of the impact of large scale interventions to raise public awareness of a lung cancer symptom. Br J Cancer, 2014 [Epub ahead of print] doi:10.1038/bjc.2014.596