Almost half of all breast cancers are diagnosed in women aged over 65 years and, when older women are diagnosed, their cancers are more likely to be at a more advanced stage than those in younger women in the UK (Lyratzopoulos et al, 2012; Cancer Research UK, 2016). Low awareness of breast cancer symptoms, particularly non-lump symptoms, and the increasing risk of breast cancer with age can mean that women are less likely to check their breasts, leading to a higher probability of delayed presentation (Ramirez et al, 1999; Linsell et al, 2008; Koo et al, 2017).
The authors developed the Promoting Early Presentation (PEP) Intervention, a scripted one-to-one interaction between a health professional and a woman, which is supported by an information booklet for the woman to take home. The PEP Intervention targets older women's knowledge, motivation, confidence and skills to present promptly with breast cancer symptoms (Burgess et al, 2008; 2009).
In a randomised controlled trial in the NHS Breast Screening Programme, delivered after the final invited mammogram, the PEP Intervention increased older women's breast cancer awareness more than usual care (Forbes et al, 2011; Kaushal et al, 2016). The PEP Intervention also increased breast cancer awareness to a similar degree in routine clinical practice delivered by NHS radiographers in the NHS Breast Screening Programme and by practice nurses in primary care (Forbes et al, 2012; Campbell et al, 2016; Dodd et al, 2017). In 2013, the All Party Parliamentary Group on Breast Cancer recommended that the PEP Intervention should be rolled out more widely to assess its effect on survival and be tested in other settings (All Party Parliamentary Group on Breast Cancer, 2013).
Aims
The aim of this project was to evaluate an adapted version of the PEP Intervention for use in a primary care setting using the seasonal flu vaccination campaign as a vehicle for delivery. Influenza vaccination for patients over the age of 65 is part of the Quality and Outcomes Framework (the incentive scheme for general practices) and therefore actively promoted by practices, leading to high uptake among patients aged over 65 years (Doran et al, 2008).
The authors developed a shorter, 1–minute version of the PEP Intervention specifically for use in a primary care setting during flu vaccination appointments. Using the flu campaign as a vehicle for the intervention allowed a large proportion of women over 70 to be targeted, including those who had never attended breast screening. The authors evaluated the effect of the training on health professionals' breast cancer awareness and confidence to deliver the intervention, the effect of the PEP Intervention on women's breast cancer awareness, attendances and two-week wait referrals for breast symptoms among older women. The views of health professionals on the acceptability and feasibility of delivering the PEP Intervention in this setting were also collected.
Methods
Participants
Six general practices in southeast London with a high proportion of female patients aged 70 years and above and a high uptake of the flu vaccination took part in the project. Women with a history of breast cancer or dementia were excluded from the study (377 out of 3457 women).
PEP Intervention
The intervention used in this study was based on PEP, which is a scripted one-to-one interaction with a health professional, takes about 1 minute to deliver and is supported by a booklet (Burgess et al, 2008; 2009; Linsell et al, 2009).
The script sets out the increasing risk of developing breast cancer with age, describes breast cancer symptoms and advises on how to look for breast changes, and encourages women to present promptly on noticing any breast change.
The booklet Looking after your breasts also includes an action planning page (where the woman can write out her plans for breast checking and what she will do if she notices a change) and information on how to make an appointment for breast screening. The health professional refers to the booklet during the intervention and gives it to the woman to take home.
Training health professionals and evaluation of this
In September and October 2011, 20 nurses and 20 GPs were trained to deliver the intervention during attendance for flu vaccination during the 2011–2012 campaign. Training sessions took place in GP practices and lasted 1 hour.
The health professionals' knowledge of breast cancer and confidence to promote early presentation of breast cancer were measured before training and up to 1 month after training. The authors asked female health professionals about their confidence to notice a change in their breasts.
The questionnaire was based on a validated measure of breast cancer awareness the authors had previously developed (Linsell et al, 2009; Omar et al, 2010). Implementation began immediately after each practice's training and continued until 31 December 2011.
Measuring the effect on breast cancer awareness in older women
GP practices carried out a postal survey of their registered female population aged 70 years and above in September 2011 and repeated it in January 2012.
This contained 10 questions that measured knowledge of breast cancer, confidence in being able to detect symptoms and barriers to symptomatic presentation, adapting other measures and incorporating evidence from qualitative studies (Burgess et al, 2006; Stubbings et al, 2009; Linsell et al, 2010;). It also asked for information on age, postcode and ethnicity.
Number of attendances for breast symptoms and urgent referrals
The practice managers collected data on the number of attendances for breast symptoms and the number of two-week wait breast referrals, using Read Codes, per quarter from January 2010 to June 2012.
Feedback from health professionals
A purposive sample of health professionals were interviewed about their views of the training, their experiences of delivering the intervention, the acceptability of the intervention to patients and the feasibility of implementing the intervention into routine practice. A mixture of GPs and nurses from all practices who were highly involved in delivering the intervention were selected at the end of the project.
Analysis
This study examined changes in health professionals' knowledge, confidence and breast checking behaviour by calculating medians and using the Wilcoxon matched pairs and McNemar tests (Swinscow and Campbell, 1997).
In the older women registered with the practices, the authors examined change in knowledge of breast cancer symptoms between the first survey (September 2011) and the second survey (January 2012) using c2 tests (the data were not paired because data to identify respondents was not available) (Swinscow and Campbell, 1997). Change in confidence to check breasts and reported barriers to symptomatic presentation were examined using univariate logistic regression. All analyses were conducted using SPSS version 15.
The authors compared the number of breast symptom attendances and two-week wait referrals made by the practice before and after the intervention (July to September 2011 and January to March 2012) and over the same time periods in the previous year. Interviews with the health professionals were analysed qualitatively using thematic analysis to identify common patterns and themes (Braun and Clarke, 2006).
Ethical approval
This project was a health service evaluation administered by the general practices and so did not require approval from an ethics committee.
Results
Effect of training on health professionals' knowledge and confidence
The response to the questionnaire about knowledge and confidence to promote early presentation of breast cancer was 100% before training and 73% after training. There were no significant differences between responders and non-responders to the second questionnaire by sex, profession (GP/nurse) or baseline breast cancer awareness, confidence to promote early presentation of breast cancer or confidence to tackle barriers to symptomatic presentation. Health professionals' knowledge of breast cancer symptoms, knowledge that risk of breast cancer increases with age, reported breast checking, confidence to notice a change in their breasts (women only), confidence to promote early presentation of breast cancer and tackle barriers to symptomatic presentation increased after training (Table 1).
| Before training | After training | P for difference* | |
|---|---|---|---|
| Number of breast cancer symptoms recognised (median out of 11) | 8 | 11 | <0.001 |
| Number of non-lump breast cancer symptoms recognised (median out of 9) | 7 | 9 | 0.001 |
| Identified a woman aged 70 years or above who was at most risk of breast cancer, rather than a younger woman or a woman of any age (%) | 52 | 85 | 0.012 |
| Knew that risk of breast cancer increases after the age of 70 rather than decreases or stays the same (%) | 66 | 97 | 0.004 |
| Confident to promote early presentation (median out of 50†) | 40 | 45 | <0.001 |
| Confident to tackle barriers to presentation (median out of 40§) | 35 | 36 | 0.008 |
| Reported checking breasts at least once a month (% of 22 women) | 32 | 68 | 0.008 |
| Confident to notice changes in breasts (% of 22 women) | 73 | 100 | 0.031 |
Effect of the campaign on older women's awareness
Response to the first survey was 36% (n=1102) and to the second survey was 39% (n=1249). Sixty per cent of participants were aged 70–79 years, and most were of white ethnic group (Table 2).
| Baseline n (%) | Follow-up n (%) | |
|---|---|---|
| Total number of participants | 1102 | 1249 |
| Age | Data available for 1089 participants | Data available for 1222 participants |
| 70–79 | 647 (59.4) | 732 (59.1) |
| 80+ | 442 (40.6) | 506 (40.9) |
| Ethnicity | Data available for 1053 participants | Data available for 1173 participants |
| White | 926 (87.9) | 1018 (86.8) |
| Mixed | 17 (1.6) | 21 (1.8) |
| Asian | 24 (2.3) | 35 (3.0) |
| Black | 79 (7.5) | 88 (7.5) |
In the second survey, the median score for knowledge of breast cancer symptoms was higher than in the first survey (5/11 versus 4/11, p=0.005) and the median score for knowledge of non-lump breast cancer symptoms was also higher (3/9 versus 2/9, p=0.002). Knowledge of seven of the 11 symptoms was higher in the second survey (Table 3), in particular change in nipple position (+6.5%), pain in breast or armpit (+6.4%) and nipple discharge or bleeding (+5.6%).
| First survey (%) n=922 | Second survey (%) n=1055 | Difference (%) | p (χ2) | Odds ratio (95% confidence interval) | |
|---|---|---|---|---|---|
| Change in nipple position | 365 (39.6) | 486 (46.1) | +6.5 | 0.004 | 1.30 (1.09–1.56) |
| Pain in breast or armpit | 516 (56.0) | 658 (62.4) | +6.4 | 0.004 | 1.30 (1.09–1.56) |
| Nipple discharge or bleeding | 543 (58.9) | 680 (64.5) | +5.6 | 0.01 | 1.27 (1.06–1.52) |
| Lump or thickening in the armpit | 648 (70.3) | 794 (75.3) | +5.0 | 0.01 | 1.28 (1.06–1.57) |
| Skin puckering or dimpling | 376 (40.8) | 480 (45.5) | +4.7 | 0.04 | 1.21 (1.01–1.45) |
| Nipple rash | 134 (14.5) | 199 (18.9) | +4.4 | 0.01 | 1.37 (1.08–1.74) |
| Skin redness | 157 (17.0) | 220 (20.9) | +3.9 | 0.03 | 1.28 (1.02–1.61) |
| Nipple pulling in | 436 (47.3) | 540 (51.2) | +3.9 | 0.08 | 1.17 (0.98–1.40) |
| Changes in shape of breast or nipple | 479 (52.0) | 574 (55.4) | +3.4 | 0.13 | 1.15 (0.96–1.37) |
| Change in size of breast or nipple | 362 (39.3) | 439 (41.6) | +2.3 | 0.29 | 1.10 (0.92–1.32) |
| Lump or thickening in the breast | 824 (89.4) | 923 (87.5) | -1.9 | 0.23 | 0.83 (0.63–1.10) |
The proportion of women reporting that they checked their breasts at least once a month was higher in the second survey (49% vs 43%, p=0.01), as was women's confidence to know which breast changes to look for and to remember to check their breasts. Knowledge of age-related risk was not significantly different between the two surveys (Table 4).
| Baseline n (%) | Follow-up n (%) | p (χ2) | Odds ratio (95% confidence interval | |
|---|---|---|---|---|
| Breast checking | ||||
| Checks breasts at least once a month | 471/1085 (43.4) | 602/1228 (49.0) | 0.01 | 1.25 (1.06–1.48) |
| Confident to know what breast changes to look for | 730/1033 (70.7) | 899/1184 (75.9) | 0.005 | 1.31 (1.08–1.58) |
| Confident to know how to check breasts | 730/1015 (71.9) | 889/1178 (75.5) | 0.06 | 1.20 (0.99–1.45) |
| Confident to remember to check breasts | 590/1010 (58.4) | 754/1151 (65.5) | 0.001 | 1.35 (1.14–1.61) |
| Age-related risk | ||||
| A 70-year-old woman | 129/1057 (12.2) | 138/1200 (11.5) | 0.63 | 0.94 (0.73–1.21) |
| After the age of 70 a woman is more likely to get breast cancer | 240/1052 (22.8) | 245/1200 (20.4) | 0.16 | 0.86 (0.71–1.06) |
| Emotional barriers | ||||
| Worrying about any treatment they might have to have | 421/984 (42.8) | 486/1120 (43.4) | 0.79 | 1.02 (0.86–1.22) |
| Worrying about what the doctor might find | 384/993 (38.7) | 460/1122 (41.0) | 0.28 | 1.10 (0.93–1.31) |
| Not feeling confident talking to the doctor about a symptom | 163/975 (16.7) | 183/1102 (16.6) | 0.95 | 0.99 (0.79–1.25) |
| Embarrassed to go and see the doctor | 134/1001 (13.4) | 130/1121 (11.6) | 0.22 | 0.85 (0.66–1.10) |
| Service barriers | ||||
| Not being able to see their usual doctor | 339/971 (34.9) | 381/1111 (34.3) | 0.77 | 0.97 (0.81–1.17) |
| Finding it difficult to make an appointment | 299/982 (30.4) | 369/1115 (33.1) | 0.19 | 1.13 (0.94–1.36) |
| Worrying about wasting the doctor's time | 243/997 (24.4) | 267/1117 (23.9) | 0.79 | 0.97 (0.80–1.19) |
| Finding the doctor difficult to talk to | 118/973 (12.1) | 148/1096 (13.5) | 0.36 | 1.13 (0.87–1.46) |
| Practical barriers | ||||
| Too many other things to worry about | 157/956 (16.4) | 159/1097 (14.5) | 0.23 | 0.86 (0.68–1.10) |
| Finding it difficult to arrange transport | 128/973 (13.2) | 153/1101 (13.9) | 0.63 | 1.07 (0.83–1.38) |
| Too busy to make time to go to the doctor | 77/964 (8.0) | 80/1081 (7.4) | 0.59 | 0.91 (0.66–1.27) |
There was no significant difference in the proportion of women reporting barriers that might put them off going to a doctor between the two surveys (median number of barriers reported 2/11 versus 2/11). The most commonly identified barrier was ‘Worrying about any treatment I might have to have’ (43%).
Attendance for symptoms and emergency referrals
Between July and September 2011 (the quarter before the seasonal flu campaign), 5/3338 (0.15%) registered women aged 70 years and above attended one of the six practices with breast symptoms; between October and December 2011 (the quarter of the seasonal flu campaign), 11/3432 (0.32%) women attended and, between January and March 2012, 12/3324 (0.36%) women attended. However, for the same period the year before, between January and March 2011, 15/3296 (0.46%) of women aged 70 years and above attended with breast symptoms.
Findings were similar regarding rates of two-week wait referrals for breast symptoms, which for every quarter from October to December 2010 and January to March 2012 were between 0.30% and 0.52% of women per quarter, with no clear increase during or after the campaign.
Health professionals' perceptions of the intervention
The authors interviewed three practice managers, three GPs and four nurses. Health professionals thought the key messages of the intervention were important and that it would help increase breast awareness in this group of women.
‘I thought it was an excellent idea, because it makes older people more aware of breast cancer … I think they forget to check their breasts and they forget to look after their bodies as they get older.’
‘The booklet is brilliant because it was actually a tool to go through and it was a prompt [to go through with the patient] … having those pictures, the little graph showing risk going up and what to do and how to phone the practice I thought was very clear.’
Health professionals reported that women valued the content on the risk of breast cancer increasing with age, breast cancer symptoms and how to promote early presentation of breast cancer.
‘There were some other women who actually when you asked them “what do you think your risk of breast cancer is as you get older?” They were shocked to find that their risk just carried on going up.’
‘The majority of them [the women] were very positive and quite a few were quite pleased and shocked to find that they could actually phone the phone number and book themselves on [for breast screening] if they wanted to. Not a lot of people realised that, so that's nice having the number at the back of the book. And the book is very visual. They liked the book.’
Most health professionals reported that the intervention was easy to deliver, that they covered all the areas of the script in the time allowed and that it did not slow down flu vaccination clinics.
‘In the beginning, I thought “are we really going to be able to do this within the five minutes [for the flu vaccination appointment] that we've got?” but it was surprising how easy it was to actually bring it up [the intervention] while you are getting it [the vaccine] out of the cupboard.’
It was reported that the intervention sometimes took longer than 1 minute if women asked questions and some opportunities to deliver the intervention were missed because of busy days and not having the booklets available.
Practice managers did not report any negative effect on overall waiting times.
Discussion
Summary of main findings
Training health professionals to deliver a brief intervention to promote early presentation of breast cancer among older women increased their knowledge and confidence to promote early presentation of breast cancer.
There was evidence that the intervention increased knowledge of breast cancer symptoms, confidence to check their breasts and frequency of breast checking among older women registered with the practices. The intervention did not increase knowledge of age-related risk of breast cancer nor reduce barriers to symptomatic presentation.
The intervention was considered worthwhile and feasible by health professionals and practice managers. There were no reports of disruption to clinics or complaints from staff or patients. There was some evidence of a small increase in the number of women attending for breast symptoms or two-week wait referrals for breast symptoms in the short term, suggesting that it would not put an unnecessary burden on the health service.
Strengths and limitations of the study
This study is, to the authors' knowledge, the first of its kind to evaluate an intervention to promote early presentation of breast cancer among older women in primary care using the seasonal influenza campaign as a vehicle. Uptake of training by health professionals was good and their response to the second survey, which evaluated their knowledge and confidence, was reasonable.
The sample size for the evaluation of the effect on breast cancer awareness among older women was large. The response rate was relatively low but comparable to similar NHS postal surveys (Campbell et al, 2009). The authors did not have any data about non-responders as they were unable to access it. Therefore the results of the survey may not be representative of all the women aged 70 years and above registered at the practices. The authors were unable to identify whether the same women responded to the first and second survey; however, the age and ethnic distribution of the survey participants were similar to the two surveys, which suggests it was unlikely that major bias was introduced through having different women responding to each survey.
Because this was an uncontrolled study, it cannot be said with certainty that the increase in breast cancer awareness was because of the intervention; however, there were no other national or local campaigns to increase breast cancer awareness during the period that could have been responsible. The authors were also unable to evaluate the quality of the delivery of the intervention. It is likely that the quality would have varied by health professional.
The questionnaire itself may have led to the finding of greater breast cancer awareness in the second survey (an impact known as the ‘mere measurement’ effect) (Godin et al, 2008). This is unlikely to explain the findings as women in the usual care group of a randomised controlled trial of the PEP Intervention completed this questionnaire at multiple time points but this resulted in limited changes in breast cancer awareness (Forbes et al, 2011).
Comparisons with other literature
A systematic review of interventions to promote cancer awareness and early presentation of symptoms (search date 2008) found that few interventions (in breast or other cancers) had been rigorously tested so there is limited evidence of their effectiveness (Austoker et al, 2009).
Since then, there have been significant developments in this field. A randomised controlled trial of the PEP Intervention in the NHS Breast Screening Programme found it increased knowledge of breast cancer symptoms and awareness that the risk of breast cancer increases with age, and reported breast checking 2 years later (Forbes et al, 2011). Implementation of this intervention during routine clinical practice in the NHS Breast Screening Programme and primary care have shown similar effectiveness (Forbes et al, 2012; Campbell et al, 2016; Dodd et al, 2017).
Unlike previous studies of the PEP Intervention, this study found that delivering the adapted version during the seasonal flu campaign did not increase knowledge of age-related risk. However, it did find increases in knowledge of breast cancer symptoms and breast checking. It may be that there is insufficient time during the 1–minute intervention to deliver this message effectively, or that health professionals were not sufficiently equipped to deliver it. A similar one-to-one intervention to promote early presentation of lung cancer symptoms in general practice found it reduced time to consultation after symptom onset and an increase in consultations for lung symptoms (Smith et al, 2013).
Since this study, the Department of Health and the NHS implemented regional cancer awareness media campaigns (‘Be Clear on Cancer’) in 2012 to promote early presentation of breast cancer among older women. These led to an 8% increase in the number of two-week wait referrals for breast symptoms, which prompted rollout of the campaigns across England in 2014 and 2015 (National Cancer Registration and Analysis Service 2016). However, it is not clear whether media campaigns led to sustained changes in breast cancer awareness.
Conclusions
This study found it was feasible and effective to train health professionals to deliver a 1–minute intervention to promote early presentation of breast cancer, which is likely to have led to increased breast cancer awareness in older women.
The seasonal flu campaign provides an opportunity to promote earlier presentation of breast cancer among older women in primary care that does not lead to excessive demand on health services. Approaches of this kind could contribute to promoting early presentation of breast cancer and improving breast cancer awareness among older women.
Although this work was conducted a number of years ago, the authors believe that the conclusions still apply as early diagnosis of breast cancer is still a public health priority (NHS England, 2017).