Abstract
Objective
To elicit women's preferences for delivery of the National Health Service (NHS) Breast Screening Programme.
Design
Interview survey.
Setting
Private households in the UK.
Population
Five hundred and ninety-seven women aged 45–75 interviewed as part of the National Statistics Omnibus Survey, December 2008 and January 2009.
Outcomes
Preferred setting (hospital or community) and preferred type of screening unit (mobile or permanent). Rated importance of distance, time, convenience and cost of travelling to a screening unit.
Results
Forty per cent of women did not mind whether the screening unit was based in a community or hospital setting, and 52% did not mind whether the unit was permanent or mobile. Among those who expressed a preference, 59% preferred a hospital to a community setting, and 62% preferred a permanent to a mobile unit. Many women (63%) said distance to a breast screening unit was an important factor, as was time to travel (58%). Among those with access to a car, 80% regarded the availability of parking as important, but only 40% regarded the cost of car travel as important. Among women with no access to a car, there was a similar pattern for ease of public transport (76%) and cost of public transport (48%).
Conclusions
Many women were unconcerned about the type and setting of breast screening units. Among those who were concerned, most preferred hospital over community settings and permanent over mobile units, but nonetheless most women said time and distance were important. Well-situated units with advanced publicity about public transport links and parking facilities may encourage greater uptake.
INTRODUCTION
The UK National Health Service (NHS) Breast Screening Programme was introduced in 1988 by the Department of Health following the Forrest report 1 with the aim of detecting breast cancer at an early stage and thus reducing breast cancer mortality. 2–4 It currently provides free breast screening every three years for all women over 50 and operates in over 80 breast screening units across the UK. Over two million women aged 50–70 are routinely invited for mammography each year in England, and around 73% of those invited attend (women aged over 70 may self-refer every 3 years if they wish). 5 Screening units are often based on hospital sites, but may also be situated in community clinics or non-health locations such as shopping centres, the rationale being that women may find them more convenient and pleasant, having less association with illness. Breast screening can also be delivered by purpose-built mobile units, which were introduced to increase geographical access, although they are generally based in a fixed location over a period of time.
In 2007, the Department of Health announced its policy to widen the age range for routine breast screening from 47 to 73 years. The Breast Screening Programme is implementing a phased expansion to accommodate this, which will eventually result in an extra 400,000 women invited for screening each year. 6 We conducted a national survey to elicit women's preferences regarding the way in which the NHS Breast Screening Programme is delivered in the context of the proposed upper and lower age extension. To our knowledge, this is the first study to report women's opinions regarding delivery of the Breast Screening Programme using a UK representative sample. The work was undertaken to inform the programme's management and planning decisions regarding the reorganization of services, and to help improve service delivery to encourage more women to attend.
METHODS
Participants and procedure
Sampling and data collection were conducted by the Office of National Statistics as part of their Omnibus Survey, a multipurpose survey carried out monthly in a representative sample of about 1800 adults living in private households in the UK. 7 The survey selects a new sample of 67 postal sectors each month with probability proportionate to size and stratified by region, socioeconomic classification and the proportion of people aged over 65. Within each sector, 30 addresses are chosen at random and one adult member (aged 16 or over) in each household is selected using a Kish Grid. A letter is sent to the household in advance and at least four attempts are made to conduct a face-to-face interview with the selected person. Proxy interviews are not undertaken.
A breast screening module was designed specifically for the study and included in the surveys in December 2008 and January 2009. It was administered to women in the current age range invited for screening (50–70) and those who would be included in proposed upward and downward age extensions. The response rate for the two surveys combined was 61% (2216/3653) of the eligible sample and the breast cancer screening module was administered to a total of 597 women aged 45–75.
Outcome measures
The module contained three preference questions: in which location women would prefer to have their mammogram (community setting, hospital setting, don't mind); which type of screening unit they would prefer (mobile unit, permanent unit, don't mind) and whether they would like to be invited with women from their GP surgery, with women from the same neighbourhood or whether they didn't mind. A further question asked where they attended their last mammogram (mobile unit, permanent unit, other, never had a mammogram). The women were then asked to rate the importance of six aspects of screening related to cost and convenience on a scale of 1 = not at all important to 4 = very important: the distance needed to travel; the time taken to get there; the availability of parking facilities; the cost of travelling by car; the ease of public transport; and the cost of public transport.
Statistical analysis
Analyses were performed using Stata version 10.0 and two-sided significance tests were used throughout, taking P = 0.05 as significant. We used multivariate logistic regression analysis to examine differences in women's preferences and the importance of the cost and convenience of travel to a screening unit between: urban and rural locations; women with and without access to a car; and three age groups (45–54, 55–64, 65–75). The analysis adjusted for supplied weighting factors which calibrate the Omnibus Survey sample to Office for National Statistics population totals. 7 This corrects for the unequal probability of selection resulting from only one adult per household being selected and compensates to some extent for non-response bias. We present counts and percentages based on the unweighted sample and presented weighted odds ratios and 95% confidence intervals for the comparative statistical analyses. Car access was defined using the ONS standard classification variable, ‘Is there a car or van available for use in the household?’ (yes/no). The Office for National Statistics Urban Rural Classification for England and Wales, and the Scottish Government Urban Rural Classification were used to define urban and rural areas.
RESULTS
Hospital versus community setting
Forty percent of women did not mind whether the screening unit was based in a community or hospital setting. Among those who did express a preference, 59% preferred a hospital setting, so around one-third (35%) of the whole sample (Table 1). Preference for a hospital setting did not vary by urban/rural location or car access but it tended to be the preferred location for the youngest compared with the oldest women (71% versus 51%, P = 0.05). Among the younger women who expressed a preference, 64% (38/59) of those who had previously attended for a mammogram preferred a hospital setting compared with 79% (42/53) who had never attended (weighted OR [95% CI]: 2.46 [0.93 to 6.46]).
Preferences for breast screening setting by urban/rural location, car access and age group among women who expressed a preference
*Multivariate logistic regression analysis adjusting for urban/rural location, car access and age group, using sampling weights to calibrate the Omnibus Survey sample to Office of National Statistics population total
Permanent versus mobile unit
Around half (52%) of the women did not mind whether the unit was permanent or mobile. Among those who did express a preference, 62% preferred a permanent unit (Table 1), that is 30% of the whole sample. A higher proportion of women living in rural areas and women without access to a car favoured mobile units, though these findings were not statistically significant. Preference for a permanent unit decreased with age (80% versus 47%, P = 0.01). Among the younger women who expressed a preference, 74% (34/46) of those who had previously attended for a mammogram preferred a permanent unit compared with 87% (33/38) who had never attended (weighted OR [95% CI]: 2.75 [(0.73 to 10.4]).
Previous attendance
Overall, 19% percent of the women had never attended a mammogram but this did not vary between urban/rural location or by car access. This is relatively high because some women were too young to be invited; the first invitation is not issued until 49–52 years. Among those who had received a previous mammogram, 47% of urban and 24% of rural women had attended a permanent unit (42% overall). We found that 90% of women who previously attended a mobile unit expressed preference for a mobile unit and 73% of women who previously attended a permanent unit expressed preference for a permanent unit.
When women were asked whether they preferred to be invited to have a mammogram within the same time period as other women from their GP surgery or other women from their neighbourhood, almost all of them (554; 93%) responded that they did not mind so no further analysis was carried out.
Cost and convenience factors
The majority of women considered distance and time to be important factors when travelling to a breast screening unit (Table 2). These issues were of more concern to women in urban areas, those with no access to a car and younger women, though this was only statistically significant for the oldest versus youngest women (P = 0.04 for distance and P = 0.002 for time). Among those with access to a car, 80% regarded the availability of parking facilities as important, but only 40% regarded the cost as important (Table 3). There was a similar pattern among women with no access to a car with regard to the ease of public transport (76%) and cost of public transport (48%) (Table 3). Again, these issues appeared to be of more concern to the younger women, particularly among those younger women who had not attended previously for a mammogram (results not shown).
Importance of distance and time to breast screening unit by urban/rural location, car access and age group
*Multivariate logistic regression analysis adjusting for urban/rural location, car access and age group, using sampling weights to calibrate the Omnibus Survey sample to Office of National Statistics population totals
Importance of the cost and convenience of travelling to breast screening unit by urban/rural location and age group
*Multivariate logistic regression analysis adjusting for urban/rural location and age group, using sampling weights to calibrate the Omnibus Survey sample to Office of National Statistics population totals
DISCUSSION
Summary of main findings
This survey has shown that women do not have any overwhelming preferences for the setting or type of breast screening unit. The majority of women were far more concerned about distance, time and convenience of travel to a screening unit, particularly those with no access to a car. Less than half of women expressed strong concerns about the cost of travel. Only women in rural areas with no access to a car were more likely to favour a mobile over a permanent unit. Younger women were generally more concerned about cost and convenience issues compared with older women, particularly among those younger women who had not attended previously for a mammogram.
Strengths and limitations of the study
To our knowledge, this study is the first national survey asking women about their preferences regarding the delivery of breast screening services and what factors are important to them when travelling to their screening unit. While the Omnibus Survey is designed to select a representative sample of the UK population, 39% of the people selected to take part in the surveys used in this analysis declined to take part or could not be contacted. We weighted our results by age, sex and region using Office for National Statistics population estimates to allow for this.
Our study does not allow us to rank the relative importance of setting, permanent or mobile, or the various cost and convenience issues. It was also not possible to examine the factors influencing the women's preferences, for example, whether permanent units were preferred because they are perceived to be more spacious, better equipped, more accessible or have better parking facilities than a mobile unit. We found that many women expressed a preference for the type of unit they had most recently attended, so it is possible that the preference data merely reflect previous experience or a lack of knowledge about the other options available within the service. This is particularly the case for the younger women who had not yet been invited for screening; they have no experience on which to base their preferences.
Relationship to other research
Socioeconomic deprivation 8–12 and geographical access 8,11,13 are known to be associated with the uptake of breast screening; women in deprived areas and living further away from screening locations are less likely to attend. Access to a car is another factor that has been linked to increased uptake. 8 A recent survey found that women in households with a car were more likely to have had a mammogram than those in households with no car; however it was not a significant predictor for attendance to cervical screening, which usually takes place in a woman's local GP surgery. 14
The results from a qualitative study on women's views of a breast screening service showed a preference for a local, easily accessible unit with free car parking facilities and a ‘cosy, nonclinical atmosphere’. 15 Maheswaran et al. 16 reported increased uptake at non-health sites, such as shopping centres or leisure centre car parks. Our finding that convenience of access to a screening unit with good car parking facilities was important to the women is consistent with research showing that these factors influence a woman's decision to attend.
The greater concern among younger women about cost and convenience of travel may reflect greater work and young family commitments than older women, or the lower attendance among older women, particularly those who are no longer routinely invited; 5 they may feel less affected by these issues. It also suggests that lower attendance in older women is not likely to be due to cost and convenience factors.
Our finding that women who expressed a preference preferred hospital over community settings for breast screening services was surprising, going against the received wisdom about organization of UK breast screening services since their inception in the late 1980s. 1
Policy implications and further research
Since a large proportion of women are not concerned about the setting and type of breast screening unit, decisions about improving or expanding the service should focus on convenient access with good public transport links and parking facilities. Advance publicity about these services may encourage greater uptake. Hospital locations should not always be assumed to have cheap and easily available parking, and non-health sites such as shopping centres, or the new locally based polyclinics, may more readily meet these requirements. There is also a need for mobile units for the women without access to a car in rural areas. Future research should examine which factors influence women in their preferred choice of setting and type of unit.
Footnotes
ACKNOWLEDGEMENTS
We thank the women who participated in the Omnibus Survey and completed the Breast Screening Module. The Omnibus Survey is carried out by the Office of National Statistics which bears no responsibility for this analysis and interpretation of the data. The co-authors wish to acknowledge with appreciation the contribution made to this paper by Joan Austoker, who passed away on 19 January 2010.