- Split View
-
Views
-
Cite
Cite
Premila Webster, Joan Austoker, Does the English Breast Screening Programme's information leaflet improve women's knowledge about mammography screening? A before and after questionnaire survey, Journal of Public Health, Volume 29, Issue 2, June 2007, Pages 173–177, https://doi.org/10.1093/pubmed/fdm007
- Share Icon Share
ABSTRACT
To explore whether the English Breast Screening Programme's leaflet improved women's knowledge of breast cancer screening.
Before and after postal questionnaire survey.
A random sample of 100 women aged between 49 and 64 years registered with GPs in Oxfordshire on whom data from a prospective, questionnaire survey was available on knowledge and perception of breast cancer screening.
Women's knowledge of lifetime risk had improved significantly (p < 0.0001) after the leaflet; however, the ‘qualitative’ interpretation of this numeric risk varied. The proportion of women who said the purpose of screening was to enable simpler treatments had increased from 34 to 45% (p = 0.05). Thirty-two per cent who had previously responded that screening prevented breast cancer now responded correctly; 20% who responded correctly before the leaflet now responded incorrectly. None of the women thought that all screen-detected cancers could be cured; 95% of the women said they were very likely to attend breast screening if invited, and the rest said they were ‘fairly likely’ to attend. Additional information women wanted included: what causes breast cancer; percentage rates for survival with and without mammograms; and lifestyle advice on how to avoid breast cancer.
Although the leaflet had improved women's knowledge of the purpose of screening in some areas, some simple messages in the leaflet had not been understood by all women.
Introduction
Women invited for breast screening in the UK are sent an invitation by post. In order to make an informed choice it is important to provide information to those invited to attend screening programmes about its benefits and limitations. Statutory professional bodies like the GMC1 also require that women be given this information for ethical and medico-legal reasons.
The English Breast Screening Programme developed standardized information materials that are used by English Breast Screening Units. This was to ensure that all women invited for breast screening would get appropriate information, which for the first time included some information on the limitations of screening. The information on limitations included the possibility of recall of women who may not ultimately have cancer and the anxiety it may cause, the likelihood that cancers may be missed, the fact that not all screen-detected cancers can be cured and that mammography can be painful. Mention was also made of radiation risks. No information is given specifically on detection rate, sensitivity or specificity, or the lifetime risk of dying from breast cancer. This study aims to assess whether women's knowledge about mammography screening improved following the leaflet.
Methods
Sample selection
The respondents to a previous prospective, descriptive questionnaire survey to assess women's knowledge about breast cancer risk and an understanding of their perception of the purpose and implications of breast screening2 were used as the sampling frame. Owing to restraints of cost and time, the follow-up questionnaire could not be sent to all the respondents of the original survey so a random sample of 100 women was drawn from the respondents using computer-generated random numbers.
The sampling frame for the original questionnaire was women aged 49–64 years registered with general practitioners (GPs) in Oxfordshire as they would either have been invited or would expect an invitation to attend breast screening in the near future. A random sample of 1000 (8.3%) out of a possible 12,000 women was chosen as the largest sample that was possible to survey with the resources available. The response rate was 67.5% (n = 640/948), and in keeping with the population profile of this age group in Oxfordshire,3 the respondents were mainly white home-owning women but from a range of educational backgrounds. Ethics committee approval was obtained for the survey.
Questionnaire design
The purpose of this questionnaire was to identify whether knowledge about breast cancer risk and screening had changed following the leaflet. Questions about knowledge and views about breast screening information were repeated from the original questionnaire. These questions from the first survey were derived from validated questionnaires.4–7 Additional questions were included to evaluate women's understanding of information included in the leaflet on the following: On questions on knowledge and information about breast screening, respondents were required to choose from a list of pre-determined options and, where appropriate, more than one answer to a particular question was allowed. To questions on whether anything in the leaflet was unclear and difficult to understand or whether the respondents required additional information that was not included in the leaflet, the respondents could give free text responses. In addition there was space provided for comments.
not all screen-detected breast cancers can be cured;
not all breast cancers are picked up on screening;
the meaning of a ‘normal’ result;
the numerical rate of recall;
the possibility of breast cancer diagnosis following recall.
The questionnaire was posted to 100 women with a covering letter and a copy of the leaflet. Self-addressed, stamped envelopes were included to return the completed questionnaires. A reminder was sent out to non-responders after 3 weeks.
Data analysis
Statistical analyses were performed using SPSS 10.0 for Windows. McNemar's test for matched pairs was computed to examine associations between change in views before and after results for all the variables.
Results
Four questionnaires were returned without being completed (one women had died and three had moved). Seventy-five completed questionnaires were returned, giving a response rate of 78%.
Table 1 summarizes the demographic characteristics of the respondents. The demographics of this sample were similar to those who responded to the original prospective survey.
Variable . | Category . | Number . | % . |
---|---|---|---|
Age (in years) | 49–54 | 29/74 | 39 |
55–60 | 28/74 | 38 | |
60+ | 17/74 | 23 | |
Marital status | Married | 64/74 | 86 |
Widowed/divorced | 8/74 | 11 | |
Single | 2/74 | 3 | |
Education | Primary/secondary school | 37/74 | 50 |
College | 25/74 | 34 | |
University/post-graduate degree | 12/74 | 16 | |
Accommodation | Owned | 67/72 | 93 |
Rented council | 2/72 | 3 | |
Rented private | 2/72 | 3 | |
Other | 1/72 | 1 | |
Ethnic group | White | 74/74 | 100 |
Variable . | Category . | Number . | % . |
---|---|---|---|
Age (in years) | 49–54 | 29/74 | 39 |
55–60 | 28/74 | 38 | |
60+ | 17/74 | 23 | |
Marital status | Married | 64/74 | 86 |
Widowed/divorced | 8/74 | 11 | |
Single | 2/74 | 3 | |
Education | Primary/secondary school | 37/74 | 50 |
College | 25/74 | 34 | |
University/post-graduate degree | 12/74 | 16 | |
Accommodation | Owned | 67/72 | 93 |
Rented council | 2/72 | 3 | |
Rented private | 2/72 | 3 | |
Other | 1/72 | 1 | |
Ethnic group | White | 74/74 | 100 |
Variable . | Category . | Number . | % . |
---|---|---|---|
Age (in years) | 49–54 | 29/74 | 39 |
55–60 | 28/74 | 38 | |
60+ | 17/74 | 23 | |
Marital status | Married | 64/74 | 86 |
Widowed/divorced | 8/74 | 11 | |
Single | 2/74 | 3 | |
Education | Primary/secondary school | 37/74 | 50 |
College | 25/74 | 34 | |
University/post-graduate degree | 12/74 | 16 | |
Accommodation | Owned | 67/72 | 93 |
Rented council | 2/72 | 3 | |
Rented private | 2/72 | 3 | |
Other | 1/72 | 1 | |
Ethnic group | White | 74/74 | 100 |
Variable . | Category . | Number . | % . |
---|---|---|---|
Age (in years) | 49–54 | 29/74 | 39 |
55–60 | 28/74 | 38 | |
60+ | 17/74 | 23 | |
Marital status | Married | 64/74 | 86 |
Widowed/divorced | 8/74 | 11 | |
Single | 2/74 | 3 | |
Education | Primary/secondary school | 37/74 | 50 |
College | 25/74 | 34 | |
University/post-graduate degree | 12/74 | 16 | |
Accommodation | Owned | 67/72 | 93 |
Rented council | 2/72 | 3 | |
Rented private | 2/72 | 3 | |
Other | 1/72 | 1 | |
Ethnic group | White | 74/74 | 100 |
Of those invited 95% (n = 55/58) had attended screening. Of the total 61 respondents (58 invited to the National Screening Programme and three had private screening) who were screened, 12% had been recalled for further tests (n = 7/61) compared with 19% in the original sample. Three had been diagnosed with breast cancer; 10 had a family history of breast cancer (seven mother and three sister).
Change in knowledge and perceptions after reading the leaflet
Knowledge of lifetime risk
Before receiving the leaflet, of the 75 respondents, 29% (95% CI 19–39%) answered the numerical question correctly compared with 67% (95% CI 56–78%) after receiving the leaflet (p < 0.0001). Although two-thirds of the women now correctly answered the numerical breast cancer risk question, their interpretation of the 1 in 10 risk varied, with 60% interpreting this level of risk as inevitable or very likely and 40% interpreting the same risk as very unlikely or unlikely.
Purpose of screening
The leaflet contained the following information about breast screening:
Around half the cancers that are found at screening are still small enough to be removed from the breast. This means that the whole breast does not have to be removed.
Table 2 shows the change in views about the purpose of screening before and after the leaflet. The leaflet states that breast screening does not prevent cancer; 24 women (32%) who previously thought screening prevented breast cancer responded correctly after reading the leaflet that it did not. However, 15 women (20%) who previously thought that screening did not prevent cancer now erroneously believed that screening did prevent cancer.Breast screening reduces the risk of the women who attend dying from breast cancer.
The purpose of screening is to: . | Before (95% CI), n = 75 . | After (95% CI), n = 75 . | McNemars test, two-sided P-value . |
---|---|---|---|
Enable simpler treatments | 34% (24–46) | 48% (36–60) | 0.05 |
Prevent breast cancer deaths | 79% (68–87) | 83% (72–90) | 0.7 |
Prevent breast cancer developing | 48% (36–60) | 36% (25–48) | 0.2 |
The purpose of screening is to: . | Before (95% CI), n = 75 . | After (95% CI), n = 75 . | McNemars test, two-sided P-value . |
---|---|---|---|
Enable simpler treatments | 34% (24–46) | 48% (36–60) | 0.05 |
Prevent breast cancer deaths | 79% (68–87) | 83% (72–90) | 0.7 |
Prevent breast cancer developing | 48% (36–60) | 36% (25–48) | 0.2 |
The purpose of screening is to: . | Before (95% CI), n = 75 . | After (95% CI), n = 75 . | McNemars test, two-sided P-value . |
---|---|---|---|
Enable simpler treatments | 34% (24–46) | 48% (36–60) | 0.05 |
Prevent breast cancer deaths | 79% (68–87) | 83% (72–90) | 0.7 |
Prevent breast cancer developing | 48% (36–60) | 36% (25–48) | 0.2 |
The purpose of screening is to: . | Before (95% CI), n = 75 . | After (95% CI), n = 75 . | McNemars test, two-sided P-value . |
---|---|---|---|
Enable simpler treatments | 34% (24–46) | 48% (36–60) | 0.05 |
Prevent breast cancer deaths | 79% (68–87) | 83% (72–90) | 0.7 |
Prevent breast cancer developing | 48% (36–60) | 36% (25–48) | 0.2 |
The only statistically significant result (p = 0.05) was the increase in the number of women who did not think that the purpose of screening was to enable simpler treatments (34%) to believing that it did, after reading the leaflet (45%).
Variables were also examined for associations with age and educational status; none of the associations were statistically significant.
False negative
Under the heading ‘How reliable is breast cancer screening’ the leaflet explained that:
In addition, in the summary, the leaflet stated:Mammography is the most reliable way of detecting breast cancer early but, like other screening tests, it is not perfect.
In both surveys women were asked how often they thought that breast screening missed breast cancer. One woman said ‘never’ before the leaflet and none of the women responded ‘never’ after the leaflet. There were no statistically significant associations between change in views.Screening may miss some breast cancers.
Recall for further tests
The leaflet included the following information under the heading: What does it mean if I am called back?'
One woman (1%) in each time period (i.e. before/after) responded ‘never’ when asked how often women were recalled. There were no statistically significant associations.Some women (about one in every 20 that are screened) are called back because the appearance of the x-ray suggests that more tests are needed.
In response to the question on the numerical proportion of women recalled, 47% (95% CI 35–59%) of the 72 women who answered this question responded correctly that 1 in 20 women are recalled; 18% (95% CI 10–29%) overestimated the recall rate, 24% (95% CI 14–35%) underestimated it and 11% (95% CI 5–21%) did not know.
Comparisons were carried out between different age groups (p = 0.771) and educational levels (p = 0.06). The respondents were also asked what proportion of the women recalled would have breast cancer. One woman responded all of them and one most of them.
Finding a breast lump between screening
Seventy out of the 74 respondents (95% CI 87–99%) said both before and after the leaflet that they would visit their GP immediately if they found a breast lump three months after a mammography. Three who previously said they would visit their GP if the problem persisted for more than a month said they would visit their GP immediately and one said both times that that she would visit her GP if the problem persists for more than one month.
Views about screening
The proportion of women who thought that they would need to attend screening periodically even if it was normal the first time had increased from 88% (95% CI 78–94%) to 96% (95% CI 89–99%) after reading the leaflet. Compared with 96% (95% CI 89–99%) who disagreed or strongly disagreed before the leaflet to the statement that screening takes too much time, all respondents now disagreed or strongly disagreed with this statement.
McNemar's test for matched pairs was not significant for any of the variables.
Evaluating women's understanding of additional information in the leaflet
Screen detected breast cancer and cure
The following statement was included in the summary of the advantages and limitations of breast screening:
Women were asked approximately, how many screen-detected cancers could be cured. Seventy-five women responded to this question. None of the women thought that all cancers or none of the cancers found by screening can be cured.Not all breast cancers that are found at screening can be cured
Meaning of a ‘normal’ result
In the leaflet, women are informed that screening can miss cancers and that cancer can occur in the interval between screens. Women were asked what they thought a normal result means. Ninety-one percent of the 74 women who responded to this question (95% CI 85–98%) said that a ‘normal’ result means that they are highly unlikely or unlikely to have cancer. Seven percent (95% CI 1–13%) said it means that they definitely do not have cancer. One woman said it meant that it was likely she may have cancer and one did not know.
Uptake intention
Women were asked, if invited to go for breast screening, how likely was it that they would attend. Five percent of the 74 who replied (95% CI 0–10%) responded fairly likely, the rest 95% (95% CI 90–100%) responded very likely.
Women were asked whether they would you like any other information that was not included in the leaflet. Of the 70 respondents who replied, 74% (95% CI 64–84%) said no, 14% (95% CI 6–20%) said yes and 12% (95% CI 4–20%) said they did not know. Those who said they would like other information were asked to specify what other information they wanted. Seven out of the 10 respondents who responded ‘yes’ said what specific information they wanted; this included: Women were also asked if there was anything in the leaflet that they found unclear or difficult to understand; 4% of the 70 respondents (95% CI 1–9%) said yes. Two of the three specified what was unclear or difficult to understand. One respondent said the statement ‘breast screening reduces the risk of the women who attend dying from breast cancer’ was confusing. Another woman responded that the leaflet included a statement ‘breast screening saves an estimated 1250 lives each year in this country’, but she did not know how many cases of breast cancer there were and how many die. The respondent also wanted to know how many lives would be saved if all women who were invited to screening attended. Two women made additional comments: one respondent wanted more explanation why women over 70 are not invited for screening; another said the section on ‘what if I need treatment?’ was too brief and vague.
what causes breast cancer;
more information on breast self examination;
more information on recall;
is more screening necessary if there is a family history;
percentage rates for survival with and without mammograms;
lifestyle advice on how to avoid breast cancer;
how hormone replacement therapy (HRT) affects breast tissues;
comparison between private and NHS screening.
Discussion
Main finding of this study
The leaflet significantly increased the knowledge of numeric lifetime risk; however, the qualitative interpretation of this risk varied widely. Women interpret the risk according to their own value judgements. Therefore, while one woman may interpret a 1 in 10 risk as an average risk, another may interpret the same risk as inevitable. Work done in this area has emphasized that the interpretation of numerical risk information is subject to a wide range of biases.8–10
Although the leaflet had improved women's knowledge of the purpose of screening to some extent it was surprising to note that even the message that screening does not prevent cancer had not been understood by all women. Although the leaflet seems to clearly state that breast screening does not prevent cancer 15 women (20%) who previously thought that screening did not prevent cancer, now erroneously believed that screening did prevent cancer.
Ninety-five percent of the women said they were very likely to attend breast screening if invited, the rest said they were ‘fairly likely’ to attend.
What is already known on this topic
Most of the research in this area is an evaluation of the content of information leaflets.
What this study adds
This study, by assessing whether the information leaflet had any impact on knowledge, perceptions and understanding of breast screening, shows that:
women's knowledge of lifetime risk had improved significantly (p < 0.0001) after the leaflet; however, the ‘qualitative’ interpretation of this numeric risk varied;
the proportion of women who said the purpose of screening was to enable simpler treatments had increased from 34 to 45% (p = 0.05);
none of the women thought that all screen-detected cancers could be cured;
although the leaflet had improved women's knowledge in some areas some simple messages in the leaflet had not been understood by all women;
women were very likely to attend breast screening if invited.
Limitations of this study
This survey was undertaken on a relatively small sample of 100 women. However, the response rate was high and baseline information on knowledge and perceptions was available on this sample. This allowed comparisons to be made and helped to gather information on the impact of the leaflet on knowledge and perception of breast screening.
It would have been useful to gain an insight into the reasons for the lack of understanding of some apparently direct and simple messages. In order to do this a sub-group would need to have been interviewed. This was not possible, partly due to the complexities and constraints of data protection.
For the first time since breast screening was introduced in the UK, the National Screening Programme had decided to develop a standard information leaflet that included some information on disbenefits. While there have been improvements, there are still gaps in the information provided. There is lack of clear information on detection rate, sensitivity and specificity. Although these areas are cursorily mentioned, the information is either vague or inadequate. The leaflet has only partly dealt with the perennial problem of epidemiological information and how to present it. While it clearly states and has subsequently increased the knowledge of women about lifetime risk, the information on lifetime survival with and without screening is not addressed. While this concept is not as easy to get across as lifetime risk and is complicated by the different breast cancers that could be detected, i.e. DCIS, invasive cancer, etc., it is still important that this information is imparted to assist women in making an informed choice about attending breast screening.
This survey is only the first small step in determining the usefulness of the information leaflet. In order to gain a better insight into women's understanding of breast screening more work needs to be done. Efforts must be made to raise awareness of the pros and cons of breast screening to assist women to make an informed choice. Policy makers should target those who may not be as well informed and perhaps use other channels, like the media, creatively, to disseminate information about screening.