- Email: [email protected]
Do women make an informed choice about participating in breast cancer screening? A survey among women invited for a first mammography screening examination
Patient Education and Counseling 89 (2012) 353–359
Contents lists available at SciVerse ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Informed Choice
Do women make an informed choice about participating in breast cancer screening? A survey among women invited for a first mammography screening examination Heleen van Agt a,*, Jacques Fracheboud a, Annemieke van der Steen b, Harry de Koning a a b
Department of Public Health, Erasmus MC University Medical Center, Rotterdam, The Netherlands Screening Organization South-West, Rotterdam, The Netherlands
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 January 2012 Received in revised form 19 July 2012 Accepted 9 August 2012
Objective: To determine the level of informed choice in women invited for breast cancer screening for the first time. Methods: To determine the content of decision-relevant knowledge, 16 experts were asked to judge whether each of 51 topics represented essential information to enable informed choices. To assess the level of informed choices, a questionnaire was then sent to all 460 invited women in the south-western part of the Netherlands who turned 50 in August 2008. Results: Of all 229 respondents, 95% were deemed to have sufficient knowledge as they answered at least 8 out of 13 items correctly. In 90% there was consistency between intention (not) to participate and attitude. As a result, 88% made an informed choice. Sixty-eight percent of women responded correctly on the item of over-diagnosis. Even if all non-respondents were assumed to have no knowledge, 50% of the total group invited to participate still had sufficient knowledge. Conclusions: Women were deemed to have sufficient relevant knowledge of the benefits and harms if they answered at least half of the items correctly. Practice implications: To further increase informed choices in breast cancer screening, information on some of the possible harms merits further attention. ß 2012 Elsevier Ireland Ltd. All rights reserved.
Keywords: Informed choice Informed decision-making Breast cancer screening Decision-relevant knowledge
1. Introduction Breast cancer screening has positive health effects in terms of reducing the risk of dying from breast cancer [1–3]. However, a negative effect of screening is that many women are advised to have a diagnostic assessment because of a positive screening result although they will never have breast cancer. The number of these women is of course much larger than those saved from a breast cancer death as a result of screening [3]. For population-based cancer screening programmes, at the population level the benefits should outweigh the harms [4]. However, individuals may still not want to be screened, because the balance between benefits and harms can be valued differently for their own personal situation. In several Western countries, informed decision-making about participating in screening, also in cancer screening, has become an explicit purpose [5,6]. An informed choice about participation in breast cancer screening requires that invited women have
* Corresponding author at: Department of Public Health, Erasmus MC University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Tel.: +31 10 7043718/7038460; fax: +31 10 7044724. E-mail address: [email protected] (H. van Agt). 0738-3991/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pec.2012.08.003
opportunities to weigh all possible favourable and unfavourable effects of screening so as to enable them to form an opinion and subsequently make an autonomous choice, free from external pressures or barriers [7]. There has been a shift from promoting the benefits of screening towards providing comprehensive information to enable people to make an informed choice [7–10]. Following Marteau, we defined an informed choice as one that is based on relevant knowledge while the decision-maker’s attitude is consistent with her actual screen behaviour [11]. Note that in applying the concept of informed choice, non-attendance can be a perfectly acceptable outcome of the deliberative process, if it is based on sufficient decision-relevant knowledge and consistent with the decision-maker’s attitude towards participating in the screening programmes [11]. In the Netherlands, breast cancer screening is offered free of charge every two years to all women in the 50–75 age brackets through a governmental funded screening programme. Regional screening organizations invite women to participate by sending them a personal letter. Information about the screening is provided in a leaflet enclosed with the invitation letter. However, it was not known whether women invited for breast cancer screening are able to make an informed decision about whether or not to participate in the screening programme.
354
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
The purpose of this study is to determine the level of informed choice in a representative sample of women who are invited for breast cancer screening for the first time. 2. Methods 2.1. Study design, participants and data collection In the Netherlands, women receive their first invitation to participate in breast cancer screening around their 50th birthday. To all 460 women in the regions of The Hague, Leiden and Delft who were newly eligible for the screening programme – mostly women who turned 50 in August 2008 – the screening organization (Bevolkingsonderzoek Zuid-West) sent an invitation letter, a standard information leaflet (version 2007) and the questionnaire to assess the level of informed choice. The content and presentation of information in the information leaflet was not based on decision-relevant knowledge for informed decision making about (non)participation in breast cancer screening. In a cover letter, women were asked to complete the questionnaire and return it to the screening organization; they were assured that not completing it would not have any consequences for their medical care. We did not send reminders. 2.2. Outcome measures 2.2.1. Knowledge The content of decision-relevant knowledge was determined in an expert consultation which aimed at the development of comprehensive, balanced and fair information about the favourable and unfavourable effects of breast cancer screening. Following the model of Irwig et al. [4], (1) we differentiated between essential information that is required for all women invited for breast cancer screening to make an informed choice, and (2) additional information that only some groups of women require to make an informed choice. Sixteen (out of 18 invited) experts participated in this consultative group, including the director of a screening organization, two general practitioners from the Dutch College of General Practitioners, one radiographer from a screening organization, one pathologist from a university medical centre, one radiologist from a general hospital, one radiologist from a university medical centre, one cancer surgeon specialist, one epidemiologist from the national cancer registry, one member of the Health Council of the Netherlands, one policymaker from the Ministry of Health, Welfare and Sport, two epidemiologists from a university medical centre, one psychologist specialized in risk communication from a university medical centre, one ethicist from a university ethics institute. The consumer perspective was included by a representative of the Dutch Breast Cancer Association representing Europe Donna in the Netherlands. This organization stands up for optimal quality of care for all people who have or had breast cancer, including the breast cancer screening programme, by stressing the patient perspective and by supporting patient’s self-management. A questionnaire was sent to the participating experts containing a list of topics compiled from national and international information materials [12] for a number of generic content domains [13,14] (Appendix I). Each topic was given a brief description based on recent evidence (Appendix II). Supplementary material related to this article found, in the online version, at http://dx.doi.org/10.1016/j.pec.2012.08.003. First, experts were asked to judge whether each topic represented information that was (1) essential, namely minimally required information that each woman need to make an informed choice, (2) additional, namely information that is not essential for all women, but only for those who need additional
information to make an informed choice or (3) unnecessary. To illustrate this, we explained that the essential information should be included in the national information leaflet provided to all invitees of breast cancer, and that the additional information might be presented elsewhere for example on a website. The standard leaflet should mention how to access the additional information. Secondly, they were asked whether the information in each category was sufficient. According to the experts, most of the topics contained essential and sufficient information for all invited women (Appendix II). In general, the experts recommended the use of simple texts without numerical values to present information on difficult topics such as false positives and over-diagnosis. The content of the knowledge measure was based on the essential information (third column of Appendix II). The aim was to develop a short and simple knowledge questionnaire, usable for large-scale assessment of informed choice, in a similar way as the knowledge measures developed for prenatal screening [13,15]. Items were formulated for generic content domains [14]. In consensus meetings among the researchers, the items were discussed, – if necessary – revised and ranked. Finally, 13 items were chosen, concerning the purpose of the screening (3), voluntariness of the screening (1), the disease being screened for (1), the likelihood of detection (1), the testing method (2), the meaning of a positive test result (1), the meaning of a negative test result (1), the unfavourable effects of the screening (1), the options following a positive diagnosis (1), and the possible findings after diagnostic assessment (1). Each item of the knowledge questionnaire consisted of a statement with response options ‘True’, ‘False’ and ‘Don’t know’. A score of 1 indicated a correct answer; a score of 0 indicated an incorrect answer. The correct answer was ‘True’ for items 1, 2, 3, 5, 8, 9, 11 and 12 and ‘False’ for items 4, 6, 7, 10 and 13. Invalid and ‘Don’t know’ responses were considered to be incorrect. The total score ranged from 0 (no correct answers) to 13 (all answers correct). There are still no agreed external criteria for the definition of ‘sufficient’ knowledge and ‘insufficient’ knowledge [16]. Therefore, following earlier research [16–18], we used the midpoint of the scale, being the most often used threshold to define sufficient knowledge. This means that scores above 7 (at least 8 correct answers) were classified as indicating sufficient knowledge about breast cancer screening and scores of 7 and lower were classified as indicating insufficient knowledge. In addition, to assess the effect of the chosen threshold on the percentages of sufficient knowledge, we used five alternative thresholds, requiring correct answers on 9, 10, 11, 12 or all 13 items respectively, to calculate these percentages. 2.2.2. Attitude and intentional screening uptake Attitudes towards attending the screening were assessed by four items based on a recent version of the Multidimensional Measure of Informed Choice (MMIC) in prenatal screening [19]: women were asked to rate their current response to ‘I feel attending the screening for breast cancer will be’ by using scores from 1 to 7 for four items anchored by ‘A bad thing/Not a bad thing’, ‘Beneficial/Not beneficial’, ‘Harmful/Not harmful’, and ‘A good thing/Not a good thing’. Responses were recoded (items 2 and 4) and summed up resulting in a score range of 0–24, with higher scores indicating more positive attitudes. The internal consistency of the attitude scale was 0.77 (Cronbach’s alpha). Scores >12 were classified as indicating positive attitudes towards undergoing the test, and scores 12 were classified as indicating negative attitudes towards undergoing the test [11,16]. Intentional screening uptake was assessed by the question ‘Do you intend to attend breast cancer screening?
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
2.2.3. Informed choice Women’s choices were classified as informed if they had 1) sufficient knowledge about the screening and 2) if there was consistency between their attitudes towards attending the screening and actual or intentional screening uptake, using the MMIC [11,16]. So, an informed choice to decline screening was defined as one in which women with negative attitudes towards attending the screening (score 12) and sufficient knowledge (score > 4) intended not to take part in the screening; an informed choice to accept screening was defined as one in which women with positive attitudes towards attending the screening (score > 12) and with sufficient knowledge (score > 4) intended to take part in the screening. Women with insufficient knowledge (according to the criterion used) or those intending to act inconsistently with their attitudes were classified as making uninformed choices. Women whose intention to act was inconsistent with their attitude were either those who intended not to participate in screening while holding positive attitudes towards attendance (score > 12) or those who intended to participate while holding negative attitudes towards attendance (score 12). In addition, women were asked to indicate their age and highest attained level of education, classified as low (elementary school, lower vocational education and junior general secondary education), intermediate (intermediate vocational, senior general secondary) and high (higher vocational and university) (www.cbs.nl/en-GB). 2.3. Analyses The percentages of questions in the knowledge questionnaire answered correctly and the percentages of sufficient knowledge were calculated in the group of respondents. To estimate the minimum level of knowledge in the total group of invited women
355
Table 1 Mean value of total knowledge score by level of education of the respondents to the questionnaire (n = 229). Educational level Low (total) Elementary Lower vocational Junior general secondary Intermediate (total) Intermediate vocational Senior general secondary High (total) Higher vocational University Total Missing
Number
Mean (SD) of total knowledge score 10.0 9.0 10.2 10.1 11.4 11.2 11.5 11.4 11.1 11.9 10.9
9 29 38 47 35 42 20 220 9
(1.9) (2.3) (1.6) (2.0) (1.5) (1.7) (1.2) (1.4) (1.5) (1.2) (1.7)
One-way ANOVA (F(6, 213) = 7.21, p < 0.0001).
(n = 460), we extrapolated these percentages to the total group, assuming that non-respondents (n = 231) did not know the answers to any of the questions. The percentage of women whose attitudes towards attending the screening were consistent with intentional screening uptake was calculated by counting the number of women with positive attitudes who intended to attend and the number of women with negative attitudes who intended not to attend. Finally, the percentages of informed choice were calculated by adding up the numbers of women whose attitudes towards attending the screening were consistent with intentional screening uptake and those with sufficient knowledge. Percentages of informed choice were calculated according to six different criteria for sufficient knowledge.
Table 2 Decision-relevant knowledge among women invited for breast cancer screening (n = 229). Knowledge item
Women who responded to the questionnaire (n = 229)
Total of women who received the questionnaire (n = 460)
Number/% with correct answers
Number/% with correct answers
1. The purpose of breast cancer screening is to detect breast cancer at an early stage. 2. If breast cancer is discovered in time, then amputation of the breast (mastectomy) may be avoided. 3. Thanks to screening, fewer women die from breast cancer. 4. Breast cancer is less common in women in the Netherlands than other types of cancer in women. 5. In breast cancer screening an X-ray is made of each breast. 6. If the X-rays of the breasts show suspicious abnormalities, that means I definitely have breast cancer. 7. If the breast X-rays do not show any suspicious abnormalities, that means I definitely do not have breast cancer. 8. The making of breast X-rays can be painful. 9. The screening may lead to the detection or treatment of slow-growing breast cancers that would otherwise never have led to complaints. 10. Screening always detects an early stage breast cancer. 11. If the breast X-rays show suspicious breast abnormalities, I will be referred to a hospital for further research. 12. The follow-up in the hospital can give certainty about whether I have cancer or not. 13. Participation in the screening is compulsory.
228/99.6%
228/49.6%
206/90.0%
206/44.8%
199/86.9% 129/56.3%
199/43.3% 129/28.0%
217/94.8% 215/93.9%
217/47.2% 215/46.3%
164/71.6%
164/35.7%
188/82.1% 153/66.8%
188/40.9% 153/33.3%
149/65.1% 211/92.1%
149/32.4% 211/45.9%
201/87.8%
201/43.7%
222/96.9%
222/48.3%
Knowledge about breast cancer screening is deemed sufficient if respondent correctly answered:
Number/% with sufficient knowledge
Number/% with sufficient knowledge
At least 8 out of 13 items At least 9 out of 13 items At least 10 out of 13 items At least 11 out of 13 items At least 12 out of 13 items All 13 items
218/95.2% 205/89.5% 185/80.8% 141/61.6% 98/42.8% 42/18.3%
47.4% 44.6% 40.2% 30.7% 21.3% 9.1%
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
356
3. Results 3.1. Response and characteristics of the responding women Of the 460 women invited for breast cancer screening, 229 (49.8%) completed the postal questionnaire. The respondents’ ages were 49 (11.5%), 50 (49.8%), 51 (38.3) and 60 (0.4%) (new resident). In 34.5% of respondents the educational level was low, 37.3% had an intermediate level of education and 28.2% had a high level of education.
Table 3 Consistency between attitudes towards attending the screening and intentional screening uptake among the respondents.
Positive attitude Negative attitude Total Consistency No consistency a
3.2. Knowledge The mean total score was 10.9 (standard deviation 1.7). The internal consistency of the scale was 0.55 (Cronbach’s alpha). Scores significantly increased in proportion to attained level of education (Table 1) (F(6, 213) = 7.21, p < 0.0001), reflecting validity. More than 80% of respondents answered 9 out of 13 questions correctly (Table 2, 1st column). The other four questions, concerning the prevalence of the disease (item 4), the likelihood of
Intentional screening uptake
No intentional screening uptake or unsure about it
Total
179 14 193
6 0 6
185 14 199a 179/199 (89.9%) 20/199 (10.1%)
30 responses were invalid (incomplete or inconsistent).
detection (item 10), potential over-diagnosis (item 9) and false negative results (item 7), were answered correctly by 56.3%, 65.1%, 66.8% and 71.6% of respondents, respectively. Between 95 and 81% of respondents demonstrated sufficient knowledge according to criteria I, II and III by answering eight to ten out of thirteen items correctly. However, only 18.3% of respondents answered all items correctly (criterion VI).
Table 4 Informed choice among women invited for breast cancer screening using different criteria for sufficient decision-relevant knowledge. Knowledge
Attitude
Intention to participate
I. Sufficient knowledge: if at least 8 out of 13 items are answered correctly Sufficient Positive 175 Negative 13 Insufficient Positive 4 Negative 1 Total 193 Informed choice Uninformed choice II. Sufficient knowledge: if at least 9 out of 13 items are answered correctly Sufficient Positive 165 Negative 11 Insufficient Positive 14 Negative 3 Total 193 Informed choice Uninformed choice III. Sufficient knowledge: if at least 10 out of 13 items are answered correctly Sufficient Positive 150 Negative 9 Insufficient Positive 29 Negative 5 Total 193 Informed choice Uninformed choice IV. Sufficient knowledge: if at least 11 out of 13 items are answered correctly Sufficient Positive 115 Negative 6 Insufficient Positive 64 Negative 8 Total 193 Informed choice Uninformed choice V. Sufficient knowledge: if at least 12 out of 13 items are answered correctly Sufficient Positive 81 Negative 5 Insufficient Positive 98 Negative 9 Total 193 Informed choice Uninformed choice VI. Sufficient knowledge: if all 13 items are answered correctly Sufficient Positive 37 Negative 3 Insufficient Positive 142 Negative 11 Total 193 Informed choice Uninformed choice
No intention to participate or unsure about it
Total
6 0 0 0 6
181 13 4 1 199 175 (87.9%) 24 (12.1%)
6 0 0 0 6
171 11 14 3 199 165 (82.9%) 34 (17.1%)
5 0 1 0 6
155 9 30 5 199 150 (75.4%) 49 (24.6%)
5 0 1 0 6
120 6 65 8 199 115 (57.8%) 84 (42.2%)
5 0 1
86 5 99 9 199 81 (40.7%) 118 (60.3%)
6
1 0 5 0 6
38 3 147 11 199 37 (18.6%) 162 (81.4%)
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
If we assume, hypothetically, that non-respondents did not know the answers to any of the items in the questionnaire, then the percentages of correct answers in the total group of women were much lower, ranging between 28 and 49.5% (Table 2, 2nd column). In the total group, the percentages of sufficient knowledge ranged from 9% to 47%, depending on the criterion used. 3.3. Attitudes and intentional screening uptake Thirty responses were incomplete or inconsistent and therefore invalid. Almost all respondents had positive attitudes towards attending the screening, and within this group almost all were planning to actually participate in the screening (90%) (Table 3). Six women had a positive attitude but no intention to participate in the screening. Although 14 had a negative attitude towards attending the screening, all of these were nevertheless planning to participate. So, among 20 women (10%) intention was not consistent with attitude. 3.4. Informed choice When measured against the first and most common criterion of sufficient knowledge, 175 women (88%) had sufficient knowledge as well as an intention (not) to participate that was consistent with their attitude and therefore made an informed decision (Table 4). A total of 24 women (12%) did not make an informed choice. Within this group, 19 did not make an informed choice as a result of inconsistencies between attitude and intentional uptake, four did not make an informed choice as a result of insufficient knowledge and one did not make an informed choice as a result of insufficient knowledge and inconsistency between attitude and intentional uptake. With 10 items answered correctly out of 13, the share of informed choices is still 75%, but with 11 out of 13 items it drops to 58%. Application of the ‘all items correct’ criterion further reduces the share of informed choices to 19%. 4. Discussion and conclusion 4.1. Discussion A recent study concluded that in several European countries information brochures about breast cancer screening lack information that is essential in order to arrive at an informed decision [20]. Moreover, information on the harms and benefits of the screening must be understood and should be unbiased [21]. Apparently, presenting information about the harms and benefits of breast cancer screening in an understandable and balanced way is a complex matter [22]. We assessed the level of decision-relevant knowledge for informed decisions on participation in breast cancer screening among a sample of women that was representative for the Dutch population in terms of the distribution of level of education. Most first-time invitees for breast cancer screening had sufficient knowledge (according to the current criterion), were positive about attending the screening and planned to participate in it. As a consequence, most of these women made an informed choice about screening uptake. Almost all respondents knew the purpose of breast cancer screening. Nevertheless, we found that, overall, respondents knew less about some of the unfavourable effects of screening than about other topics. Almost one third of the responding women thought that a normal mammogram guaranteed absence of breast cancer, and one third were not aware of the risk of over-diagnosis. If we assume that women did not respond to the questionnaire because they did not know the answers, then, with the criterion used, only
357
half of the total group of invited women had sufficient knowledge. This proportion can be considered the minimum level of knowledge in this population. Women in our study received the national information leaflet that was used for that purpose at the time, which had not been drawn up on the basis of our expertdefined information. For example, it did mention the possibility of over-diagnosis. Also note that not everyone who receives an information leaflet is likely to read it thoroughly, if at all. To our knowledge there have been no other studies so far that assessed the level of informed choice among first-time invitees for breast cancer screening in the general population. Some studies did evaluate the effects of information materials on knowledge and informed choice in breast cancer screening [23,17,24,25]. However, the focus in these studies was different, aimed at enhancing decision-making among women in age groups for which the screening was not initially recommended. Our study assessed the level of informed decision-making within the current practice of breast cancer screening. Previous studies also differed in terms of the definition and operationalisation of the content of decisionrelevant knowledge. Although the knowledge measure in most of these studies included items about some of the risks and benefits of screening, often numerical, these measures were not based on expert-defined decision-relevant knowledge as in our study. A major strength of our study is that the content of the knowledge measure was based on an agreed minimum of decisionrelevant knowledge derived from relevant experts and recent evidence, which has not previously been defined in a systematic way. According to these experts, many topics constitute essential information for all invited women, including information about the unfavourable effects of screening such as false positive and false negative results, and over-diagnosis, if presented in a simple and understandable way. Gøtzsche et al. concluded that information leaflets fail to address many of the unfavourable effects of screening [26]. As compared with the information leaflet developed by Gøtzsche et al. (www.cochrane.dk), our expert-based information contained almost all topics concerning the unfavourable effects of screening, including over-diagnosis. There were some differences in the numerical information presented, due to country-specific data, but overall the content of the information too was very similar. For example, the text about over-diagnosis says: ‘if we detect breast cancer, you will usually receive treatment. However, breast cancer sometimes grows so slowly that it would not have caused you any problems during your life. In that case, the treatment would have been unnecessary.’. Because of the current evidence in the Netherlands, the only topic not included in the expert-based information was extra surgery and after-treatment. In a Dutch study, it was found that the proportion of mastectomies among the target age group had decreased by 25% in 2009, two years after the full implementation of the nation-wide screening programme for women aged 50–69 years [27]. More importantly, mastectomy rates per 100,000 women aged 50–69 years decreased after the prevalence peak in the initial stage of the programme, in contrast to younger and older breast cancer patients in whom mastectomy rates remained constant or even slightly increased [28]. Whereas Wald recommended that information in leaflets should be quantitative and specific rather than general [11], our experts advised to restrict information to simple messages and to minimize the use of numerical values. For that reason, we did not include numerical items. Although Gøtzsche’s leaflet did contain numerical information, this was presented separately from the basic message. This study has some limitations. Despite the use of anonymous questionnaires and pre-stamped envelope to return the questionnaire, the response rate was 50%, probably due to the timing of the study (which coincided with the summer holidays) and the fact
358
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
that we did not send reminders. However, this is in accordance with other survey research among general populations in the Netherlands, where a trend towards lower response rates was found [29]. Educational level might be correlated with the level of sufficient knowledge and informed choice. However, there was probably no responder bias related to this as the distribution of educational level reflects that of the general Dutch population, of which 60% have an intermediate level of education (VMBO) maximally. Within the context of this study to measure informed choice anonymously, it was not feasible to ask women’s consent for obtaining individual data on actual screening uptake from registries to link with their informed choice outcomes. Therefore, we did not measure actual screening uptake, which in the Netherlands is 80% for women invited for breast cancer screening for the first time [28]. Possible selective response may have resulted in an overrepresentation of women with positive feelings about the screening. It is also possible that the measurement of attitudes was limited. Attitude refers to one’s like or dislike towards a specific entity, and measurement should therefore be specific to context, action and time [30]. Attitude is assumed to have a cognitive, affective and behavioural component, which should be included in the measure [30]. Although we adopted the attitude measure from the frequently used MMIC [19], the instruction may not be specific enough. Possibly, the affective component in the attitude measure was underrepresented. For example, there were no items asking whether respondents feel that the screening is unpleasant or embarrassing. There are some issues for further research. General agreement is needed on the content of decision-relevant knowledge for informed choice about participating in breast cancer screening and on who should define this knowledge [31]. Ideally, a knowledge measure based on well-defined decision-relevant knowledge will enable valid measurements regarding whether a person has sufficient knowledge or not. Given the experts’ opinion that information about many topics is essential, probably a stricter criterion for correct answers should be used if the items represent the minimally required knowledge. Psychometric methods as used in the development of educational tests might help to construct reliable and valid knowledge measures. To develop information materials or other interventions to support informed decision making in breast cancer screening, further steps are needed. However, the deliberation process of how people should achieve good decisions remains poorly understood [32]. Therefore, future research should focus on how information should be presented to enable the acquirement of knowledge that can be used for personal deliberation of options to enable a free and autonomous choice. There are many decision-making theories that may be of help in this research [33]. 4.2. Conclusion Of all responding first-time invitees for breast cancer screening, 88% made an informed choice about participation, as measured against the current criterion for sufficient knowledge. Women had sufficient relevant knowledge about the benefits and harms in general, but less knowledge about some specific harms, namely the possibility of a false negative mammogram or over-diagnosis. 4.3. Practice implications To further increase informed choices in breast cancer screening, information on possible harms is an issue that merits special attention. Information materials based on well-defined decisionrelevant knowledge are essential. A summary of expert-defined information is now available.
Acknowledgements Van Agt, Fracheboud, Van der Steen and De Koning have no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous three years, any other relationships or activities that could appear to have influenced the submitted work. Only De Koning declares financial activities outside the submitted work, consisting of a grant from a German insurance company, SCOR Global Life, paid to his institution for forecasting age specific incidences for carcinoma in situ of the breast and invasive breast cancer under the assumption of various screening scenarios. We would like to thank all experts who participated in the expert consultation. We are grateful to Mr. E.J. Kamp, communication adviser of Bevolkingsonderzoek Zuid-West (formerly BoBWest), for his kind assistance in the data collection. Sadly, he died in 2010. This research was funded by the Centre for Population Screening (CvB) of the National Institute for Public Health and the Environment (RIVM). The CvB was not in any way involved in the study design, in the collection, analysis and interpretation of the data, in the writing of the report or in the decision to submit the paper for publication. References [1] Berry DA, Cronin KA, Plevritis SK, Fryback DG, Clarke L, Zelen M, et al. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 2005;353:1784–92. [2] Blanks RG, Moss SM, McGahan CE, Quinn MJ, Babb PJ. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1990–8: comparison of observed with predicted mortality. Brit Med J 2000;321:665–9. [3] Otto SJ, Fracheboud J, Looman CW, Broeders MJ, Boer R, Hendriks JH, et al. Initiation of population-based mammography screening in Dutch municipalities and effect on breast-cancer mortality: a systematic review. Lancet 2003;361:1411–7. [4] Irwig L, McCaffery K, Salkeld G, Bossuyt P. Informed choice for screening: implications for evaluation. Brit Med J 2006;332:1148–50. [5] Minister of Health Letter about screening [Kaderbrief screening]. The Hague; 2008 [6] General Medical Council Seeking Patients’ Consent: the Ethical Considerations. London; 1998 [7] Beauchamp TL, Childress JF. Principles of biomedical ethics, 5th edition, Oxford University Press; 2001. [8] Raffle AE. Information about screening – is it to achieve high uptake or to ensure informed choice? Health Expect 2001;4:92–8. [9] Marteau TM, Kinmonth AL. Screening for cardiovascular risk: public health imperative or matter for individual informed choice? Brit Med J 2002;325: 78–80. [10] Austoker J. Gaining informed consent for screening is difficult—but many misconceptions need to be undone. Brit Med J 1999;319:722–3. [11] Marteau TM, Dormandy E, Michie S. A measure of informed choice. Health Expect 2001;4:99–108. [12] National Cancer Institute NCI. Designing print materials: a communications guide for breast cancer screening. Bethesda (MD): NIH; 2007. [13] Schoonen HM, van Agt HM, Essink-Bot ML, Wildschut HI, Steegers EA, de Koning HJ. Informed decision-making in prenatal screening for Down’s syndrome: what knowledge is relevant? Patient Educ Couns 2011;84:265–70. [14] Wald N. Information leaflets in medical screening. J Med Screen 2006;13:109. [15] Schoonen HM, Essink-Bot ML, Van Agt HM, Wildschut HI, Steegers EA, De Koning HJ. Informed decision-making about the fetal anomaly scan: what knowledge is relevant? Ultrasound Obstet Gynecol 2011;37:649–57. [16] Michie S, Dormandy E, Marteau TM. The multi-dimensional measure of informed choice: a validation study. Patient Educ Couns 2002;48:87–91. [17] Mathieu E, Barratt AL, McGeechan K, Davey HM, Howard K, Houssami N. Helping women make choices about mammography screening: an online randomized trial of a decision aid for 40-year-old women. Patient Educ Couns 2010;61:63–72. [18] Dormandy E, Michie S, Hooper R, Marteau TM. Informed choice in antenatal Down syndrome screening: a cluster-randomised trial of combined versus separate visit testing. Patient Educ Couns 2006;61:56–64. [19] Dormandy E, Tsui EY, Marteau TM. Development of a measure of informed choice suitable for use in low literacy populations. Patient Educ Couns 2007;66:278–95. [20] Gummersbach E, Piccoliori G, Zerbe CO, Altiner A, Othman C, Rose C, et al. Are women getting relevant information about mammography screening for an
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
[21] [22]
[23]
[24]
[25]
[26]
informed consent: a critical appraisal of information brochures used for screening invitation in Germany, Italy, Spain and France. Eur J Public Health 2010;20:409–14. Fox R. Informed choice in screening programmes: do leaflets help? A critical literature review. J Public Health (Oxf) 2006;28:309–17. Zapka JG, Geller BM, Bulliard JL, Fracheboud J, Sancho-Garnier H, BallardBarbash R, et al. Print information to inform decisions about mammography screening participation in 16 countries with population-based programs. Patient Educ Couns 2006;63:126–37. Mathieu E, Barratt A, Davey HM, McGeechan K, Howard K, Houssami N. Informed choice in mammography screening: a randomized trial of a decision aid for 70-year-old women. Arch Intern Med 2007;167:2039–46. Lewis CL, Pignone MP, Sheridan SL, Downs SM, Kinsinger LS. A randomized trial of three videos that differ in the framing of information about mammography in women 40 to 49 years old. J Gen Intern Med 2003;18:875–83. Haakenson CP, Vickers KS, Cha SS, Vachon CM, Thielen JM, Kircher KJ, et al. Efficacy of a simple, low-cost educational intervention in improving knowledge about risks and benefits of screening mammography. Mayo Clin Proc 2006;81:783–91. Gøtzsche PC, Jorgensen KJ. The breast screening programme and misinforming the public. J R Soc Med 2011;104:361–9.
359
[27] Fracheboud J, Groenewoud JH, de Koning HJ. Fifteen years of population-based breast cancer screening in the Netherlands. Semin Breast Dis 2007;10:72–82. [28] National Evaluation Team for Breast cancer screening in the Netherlands N. National evaluation of breast cancer screening in the Netherlands 1990–2007. Rotterdam: Erasmus University Medical Center, Dept. of Public Health. Report No.: Twelfth report; 2009 [29] Smeets H, Janssen JPG. Using National registrations to correct for selctive nonresponse. Political preference of ethnic groups. In: Statistics Canada Symposium 2001. 2001. p. 1–9. [30] Ajzen I. Attitudes, personality, and behavior. Buckingham: Open University Press; 1988. [31] Green JM, Hewison J, Bekker HL, Bryant LD, Cuckle HS. Psychosocial aspects of genetic screening of pregnant women and newborns: a systematic review. Health Technol Assess 2004;8:1–109. iii, ix–x. [32] Elwyn G, Frosch D, Volandes AE, Edwards A, Montori VM. Investing in deliberation: a definition and classification of decision support interventions for people facing difficult health decisions. Med Decis Making 2010;30: 701–11. [33] Durand MA, Stiel M, Boivin J, Elwyn G. Where is the theory? Evaluating the theoretical frameworks described in decision support technologies. Patient Educ Couns 2008;71:125–35.
Contents lists available at SciVerse ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Informed Choice
Do women make an informed choice about participating in breast cancer screening? A survey among women invited for a first mammography screening examination Heleen van Agt a,*, Jacques Fracheboud a, Annemieke van der Steen b, Harry de Koning a a b
Department of Public Health, Erasmus MC University Medical Center, Rotterdam, The Netherlands Screening Organization South-West, Rotterdam, The Netherlands
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 January 2012 Received in revised form 19 July 2012 Accepted 9 August 2012
Objective: To determine the level of informed choice in women invited for breast cancer screening for the first time. Methods: To determine the content of decision-relevant knowledge, 16 experts were asked to judge whether each of 51 topics represented essential information to enable informed choices. To assess the level of informed choices, a questionnaire was then sent to all 460 invited women in the south-western part of the Netherlands who turned 50 in August 2008. Results: Of all 229 respondents, 95% were deemed to have sufficient knowledge as they answered at least 8 out of 13 items correctly. In 90% there was consistency between intention (not) to participate and attitude. As a result, 88% made an informed choice. Sixty-eight percent of women responded correctly on the item of over-diagnosis. Even if all non-respondents were assumed to have no knowledge, 50% of the total group invited to participate still had sufficient knowledge. Conclusions: Women were deemed to have sufficient relevant knowledge of the benefits and harms if they answered at least half of the items correctly. Practice implications: To further increase informed choices in breast cancer screening, information on some of the possible harms merits further attention. ß 2012 Elsevier Ireland Ltd. All rights reserved.
Keywords: Informed choice Informed decision-making Breast cancer screening Decision-relevant knowledge
1. Introduction Breast cancer screening has positive health effects in terms of reducing the risk of dying from breast cancer [1–3]. However, a negative effect of screening is that many women are advised to have a diagnostic assessment because of a positive screening result although they will never have breast cancer. The number of these women is of course much larger than those saved from a breast cancer death as a result of screening [3]. For population-based cancer screening programmes, at the population level the benefits should outweigh the harms [4]. However, individuals may still not want to be screened, because the balance between benefits and harms can be valued differently for their own personal situation. In several Western countries, informed decision-making about participating in screening, also in cancer screening, has become an explicit purpose [5,6]. An informed choice about participation in breast cancer screening requires that invited women have
* Corresponding author at: Department of Public Health, Erasmus MC University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Tel.: +31 10 7043718/7038460; fax: +31 10 7044724. E-mail address: [email protected] (H. van Agt). 0738-3991/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pec.2012.08.003
opportunities to weigh all possible favourable and unfavourable effects of screening so as to enable them to form an opinion and subsequently make an autonomous choice, free from external pressures or barriers [7]. There has been a shift from promoting the benefits of screening towards providing comprehensive information to enable people to make an informed choice [7–10]. Following Marteau, we defined an informed choice as one that is based on relevant knowledge while the decision-maker’s attitude is consistent with her actual screen behaviour [11]. Note that in applying the concept of informed choice, non-attendance can be a perfectly acceptable outcome of the deliberative process, if it is based on sufficient decision-relevant knowledge and consistent with the decision-maker’s attitude towards participating in the screening programmes [11]. In the Netherlands, breast cancer screening is offered free of charge every two years to all women in the 50–75 age brackets through a governmental funded screening programme. Regional screening organizations invite women to participate by sending them a personal letter. Information about the screening is provided in a leaflet enclosed with the invitation letter. However, it was not known whether women invited for breast cancer screening are able to make an informed decision about whether or not to participate in the screening programme.
354
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
The purpose of this study is to determine the level of informed choice in a representative sample of women who are invited for breast cancer screening for the first time. 2. Methods 2.1. Study design, participants and data collection In the Netherlands, women receive their first invitation to participate in breast cancer screening around their 50th birthday. To all 460 women in the regions of The Hague, Leiden and Delft who were newly eligible for the screening programme – mostly women who turned 50 in August 2008 – the screening organization (Bevolkingsonderzoek Zuid-West) sent an invitation letter, a standard information leaflet (version 2007) and the questionnaire to assess the level of informed choice. The content and presentation of information in the information leaflet was not based on decision-relevant knowledge for informed decision making about (non)participation in breast cancer screening. In a cover letter, women were asked to complete the questionnaire and return it to the screening organization; they were assured that not completing it would not have any consequences for their medical care. We did not send reminders. 2.2. Outcome measures 2.2.1. Knowledge The content of decision-relevant knowledge was determined in an expert consultation which aimed at the development of comprehensive, balanced and fair information about the favourable and unfavourable effects of breast cancer screening. Following the model of Irwig et al. [4], (1) we differentiated between essential information that is required for all women invited for breast cancer screening to make an informed choice, and (2) additional information that only some groups of women require to make an informed choice. Sixteen (out of 18 invited) experts participated in this consultative group, including the director of a screening organization, two general practitioners from the Dutch College of General Practitioners, one radiographer from a screening organization, one pathologist from a university medical centre, one radiologist from a general hospital, one radiologist from a university medical centre, one cancer surgeon specialist, one epidemiologist from the national cancer registry, one member of the Health Council of the Netherlands, one policymaker from the Ministry of Health, Welfare and Sport, two epidemiologists from a university medical centre, one psychologist specialized in risk communication from a university medical centre, one ethicist from a university ethics institute. The consumer perspective was included by a representative of the Dutch Breast Cancer Association representing Europe Donna in the Netherlands. This organization stands up for optimal quality of care for all people who have or had breast cancer, including the breast cancer screening programme, by stressing the patient perspective and by supporting patient’s self-management. A questionnaire was sent to the participating experts containing a list of topics compiled from national and international information materials [12] for a number of generic content domains [13,14] (Appendix I). Each topic was given a brief description based on recent evidence (Appendix II). Supplementary material related to this article found, in the online version, at http://dx.doi.org/10.1016/j.pec.2012.08.003. First, experts were asked to judge whether each topic represented information that was (1) essential, namely minimally required information that each woman need to make an informed choice, (2) additional, namely information that is not essential for all women, but only for those who need additional
information to make an informed choice or (3) unnecessary. To illustrate this, we explained that the essential information should be included in the national information leaflet provided to all invitees of breast cancer, and that the additional information might be presented elsewhere for example on a website. The standard leaflet should mention how to access the additional information. Secondly, they were asked whether the information in each category was sufficient. According to the experts, most of the topics contained essential and sufficient information for all invited women (Appendix II). In general, the experts recommended the use of simple texts without numerical values to present information on difficult topics such as false positives and over-diagnosis. The content of the knowledge measure was based on the essential information (third column of Appendix II). The aim was to develop a short and simple knowledge questionnaire, usable for large-scale assessment of informed choice, in a similar way as the knowledge measures developed for prenatal screening [13,15]. Items were formulated for generic content domains [14]. In consensus meetings among the researchers, the items were discussed, – if necessary – revised and ranked. Finally, 13 items were chosen, concerning the purpose of the screening (3), voluntariness of the screening (1), the disease being screened for (1), the likelihood of detection (1), the testing method (2), the meaning of a positive test result (1), the meaning of a negative test result (1), the unfavourable effects of the screening (1), the options following a positive diagnosis (1), and the possible findings after diagnostic assessment (1). Each item of the knowledge questionnaire consisted of a statement with response options ‘True’, ‘False’ and ‘Don’t know’. A score of 1 indicated a correct answer; a score of 0 indicated an incorrect answer. The correct answer was ‘True’ for items 1, 2, 3, 5, 8, 9, 11 and 12 and ‘False’ for items 4, 6, 7, 10 and 13. Invalid and ‘Don’t know’ responses were considered to be incorrect. The total score ranged from 0 (no correct answers) to 13 (all answers correct). There are still no agreed external criteria for the definition of ‘sufficient’ knowledge and ‘insufficient’ knowledge [16]. Therefore, following earlier research [16–18], we used the midpoint of the scale, being the most often used threshold to define sufficient knowledge. This means that scores above 7 (at least 8 correct answers) were classified as indicating sufficient knowledge about breast cancer screening and scores of 7 and lower were classified as indicating insufficient knowledge. In addition, to assess the effect of the chosen threshold on the percentages of sufficient knowledge, we used five alternative thresholds, requiring correct answers on 9, 10, 11, 12 or all 13 items respectively, to calculate these percentages. 2.2.2. Attitude and intentional screening uptake Attitudes towards attending the screening were assessed by four items based on a recent version of the Multidimensional Measure of Informed Choice (MMIC) in prenatal screening [19]: women were asked to rate their current response to ‘I feel attending the screening for breast cancer will be’ by using scores from 1 to 7 for four items anchored by ‘A bad thing/Not a bad thing’, ‘Beneficial/Not beneficial’, ‘Harmful/Not harmful’, and ‘A good thing/Not a good thing’. Responses were recoded (items 2 and 4) and summed up resulting in a score range of 0–24, with higher scores indicating more positive attitudes. The internal consistency of the attitude scale was 0.77 (Cronbach’s alpha). Scores >12 were classified as indicating positive attitudes towards undergoing the test, and scores 12 were classified as indicating negative attitudes towards undergoing the test [11,16]. Intentional screening uptake was assessed by the question ‘Do you intend to attend breast cancer screening?
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
2.2.3. Informed choice Women’s choices were classified as informed if they had 1) sufficient knowledge about the screening and 2) if there was consistency between their attitudes towards attending the screening and actual or intentional screening uptake, using the MMIC [11,16]. So, an informed choice to decline screening was defined as one in which women with negative attitudes towards attending the screening (score 12) and sufficient knowledge (score > 4) intended not to take part in the screening; an informed choice to accept screening was defined as one in which women with positive attitudes towards attending the screening (score > 12) and with sufficient knowledge (score > 4) intended to take part in the screening. Women with insufficient knowledge (according to the criterion used) or those intending to act inconsistently with their attitudes were classified as making uninformed choices. Women whose intention to act was inconsistent with their attitude were either those who intended not to participate in screening while holding positive attitudes towards attendance (score > 12) or those who intended to participate while holding negative attitudes towards attendance (score 12). In addition, women were asked to indicate their age and highest attained level of education, classified as low (elementary school, lower vocational education and junior general secondary education), intermediate (intermediate vocational, senior general secondary) and high (higher vocational and university) (www.cbs.nl/en-GB). 2.3. Analyses The percentages of questions in the knowledge questionnaire answered correctly and the percentages of sufficient knowledge were calculated in the group of respondents. To estimate the minimum level of knowledge in the total group of invited women
355
Table 1 Mean value of total knowledge score by level of education of the respondents to the questionnaire (n = 229). Educational level Low (total) Elementary Lower vocational Junior general secondary Intermediate (total) Intermediate vocational Senior general secondary High (total) Higher vocational University Total Missing
Number
Mean (SD) of total knowledge score 10.0 9.0 10.2 10.1 11.4 11.2 11.5 11.4 11.1 11.9 10.9
9 29 38 47 35 42 20 220 9
(1.9) (2.3) (1.6) (2.0) (1.5) (1.7) (1.2) (1.4) (1.5) (1.2) (1.7)
One-way ANOVA (F(6, 213) = 7.21, p < 0.0001).
(n = 460), we extrapolated these percentages to the total group, assuming that non-respondents (n = 231) did not know the answers to any of the questions. The percentage of women whose attitudes towards attending the screening were consistent with intentional screening uptake was calculated by counting the number of women with positive attitudes who intended to attend and the number of women with negative attitudes who intended not to attend. Finally, the percentages of informed choice were calculated by adding up the numbers of women whose attitudes towards attending the screening were consistent with intentional screening uptake and those with sufficient knowledge. Percentages of informed choice were calculated according to six different criteria for sufficient knowledge.
Table 2 Decision-relevant knowledge among women invited for breast cancer screening (n = 229). Knowledge item
Women who responded to the questionnaire (n = 229)
Total of women who received the questionnaire (n = 460)
Number/% with correct answers
Number/% with correct answers
1. The purpose of breast cancer screening is to detect breast cancer at an early stage. 2. If breast cancer is discovered in time, then amputation of the breast (mastectomy) may be avoided. 3. Thanks to screening, fewer women die from breast cancer. 4. Breast cancer is less common in women in the Netherlands than other types of cancer in women. 5. In breast cancer screening an X-ray is made of each breast. 6. If the X-rays of the breasts show suspicious abnormalities, that means I definitely have breast cancer. 7. If the breast X-rays do not show any suspicious abnormalities, that means I definitely do not have breast cancer. 8. The making of breast X-rays can be painful. 9. The screening may lead to the detection or treatment of slow-growing breast cancers that would otherwise never have led to complaints. 10. Screening always detects an early stage breast cancer. 11. If the breast X-rays show suspicious breast abnormalities, I will be referred to a hospital for further research. 12. The follow-up in the hospital can give certainty about whether I have cancer or not. 13. Participation in the screening is compulsory.
228/99.6%
228/49.6%
206/90.0%
206/44.8%
199/86.9% 129/56.3%
199/43.3% 129/28.0%
217/94.8% 215/93.9%
217/47.2% 215/46.3%
164/71.6%
164/35.7%
188/82.1% 153/66.8%
188/40.9% 153/33.3%
149/65.1% 211/92.1%
149/32.4% 211/45.9%
201/87.8%
201/43.7%
222/96.9%
222/48.3%
Knowledge about breast cancer screening is deemed sufficient if respondent correctly answered:
Number/% with sufficient knowledge
Number/% with sufficient knowledge
At least 8 out of 13 items At least 9 out of 13 items At least 10 out of 13 items At least 11 out of 13 items At least 12 out of 13 items All 13 items
218/95.2% 205/89.5% 185/80.8% 141/61.6% 98/42.8% 42/18.3%
47.4% 44.6% 40.2% 30.7% 21.3% 9.1%
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
356
3. Results 3.1. Response and characteristics of the responding women Of the 460 women invited for breast cancer screening, 229 (49.8%) completed the postal questionnaire. The respondents’ ages were 49 (11.5%), 50 (49.8%), 51 (38.3) and 60 (0.4%) (new resident). In 34.5% of respondents the educational level was low, 37.3% had an intermediate level of education and 28.2% had a high level of education.
Table 3 Consistency between attitudes towards attending the screening and intentional screening uptake among the respondents.
Positive attitude Negative attitude Total Consistency No consistency a
3.2. Knowledge The mean total score was 10.9 (standard deviation 1.7). The internal consistency of the scale was 0.55 (Cronbach’s alpha). Scores significantly increased in proportion to attained level of education (Table 1) (F(6, 213) = 7.21, p < 0.0001), reflecting validity. More than 80% of respondents answered 9 out of 13 questions correctly (Table 2, 1st column). The other four questions, concerning the prevalence of the disease (item 4), the likelihood of
Intentional screening uptake
No intentional screening uptake or unsure about it
Total
179 14 193
6 0 6
185 14 199a 179/199 (89.9%) 20/199 (10.1%)
30 responses were invalid (incomplete or inconsistent).
detection (item 10), potential over-diagnosis (item 9) and false negative results (item 7), were answered correctly by 56.3%, 65.1%, 66.8% and 71.6% of respondents, respectively. Between 95 and 81% of respondents demonstrated sufficient knowledge according to criteria I, II and III by answering eight to ten out of thirteen items correctly. However, only 18.3% of respondents answered all items correctly (criterion VI).
Table 4 Informed choice among women invited for breast cancer screening using different criteria for sufficient decision-relevant knowledge. Knowledge
Attitude
Intention to participate
I. Sufficient knowledge: if at least 8 out of 13 items are answered correctly Sufficient Positive 175 Negative 13 Insufficient Positive 4 Negative 1 Total 193 Informed choice Uninformed choice II. Sufficient knowledge: if at least 9 out of 13 items are answered correctly Sufficient Positive 165 Negative 11 Insufficient Positive 14 Negative 3 Total 193 Informed choice Uninformed choice III. Sufficient knowledge: if at least 10 out of 13 items are answered correctly Sufficient Positive 150 Negative 9 Insufficient Positive 29 Negative 5 Total 193 Informed choice Uninformed choice IV. Sufficient knowledge: if at least 11 out of 13 items are answered correctly Sufficient Positive 115 Negative 6 Insufficient Positive 64 Negative 8 Total 193 Informed choice Uninformed choice V. Sufficient knowledge: if at least 12 out of 13 items are answered correctly Sufficient Positive 81 Negative 5 Insufficient Positive 98 Negative 9 Total 193 Informed choice Uninformed choice VI. Sufficient knowledge: if all 13 items are answered correctly Sufficient Positive 37 Negative 3 Insufficient Positive 142 Negative 11 Total 193 Informed choice Uninformed choice
No intention to participate or unsure about it
Total
6 0 0 0 6
181 13 4 1 199 175 (87.9%) 24 (12.1%)
6 0 0 0 6
171 11 14 3 199 165 (82.9%) 34 (17.1%)
5 0 1 0 6
155 9 30 5 199 150 (75.4%) 49 (24.6%)
5 0 1 0 6
120 6 65 8 199 115 (57.8%) 84 (42.2%)
5 0 1
86 5 99 9 199 81 (40.7%) 118 (60.3%)
6
1 0 5 0 6
38 3 147 11 199 37 (18.6%) 162 (81.4%)
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
If we assume, hypothetically, that non-respondents did not know the answers to any of the items in the questionnaire, then the percentages of correct answers in the total group of women were much lower, ranging between 28 and 49.5% (Table 2, 2nd column). In the total group, the percentages of sufficient knowledge ranged from 9% to 47%, depending on the criterion used. 3.3. Attitudes and intentional screening uptake Thirty responses were incomplete or inconsistent and therefore invalid. Almost all respondents had positive attitudes towards attending the screening, and within this group almost all were planning to actually participate in the screening (90%) (Table 3). Six women had a positive attitude but no intention to participate in the screening. Although 14 had a negative attitude towards attending the screening, all of these were nevertheless planning to participate. So, among 20 women (10%) intention was not consistent with attitude. 3.4. Informed choice When measured against the first and most common criterion of sufficient knowledge, 175 women (88%) had sufficient knowledge as well as an intention (not) to participate that was consistent with their attitude and therefore made an informed decision (Table 4). A total of 24 women (12%) did not make an informed choice. Within this group, 19 did not make an informed choice as a result of inconsistencies between attitude and intentional uptake, four did not make an informed choice as a result of insufficient knowledge and one did not make an informed choice as a result of insufficient knowledge and inconsistency between attitude and intentional uptake. With 10 items answered correctly out of 13, the share of informed choices is still 75%, but with 11 out of 13 items it drops to 58%. Application of the ‘all items correct’ criterion further reduces the share of informed choices to 19%. 4. Discussion and conclusion 4.1. Discussion A recent study concluded that in several European countries information brochures about breast cancer screening lack information that is essential in order to arrive at an informed decision [20]. Moreover, information on the harms and benefits of the screening must be understood and should be unbiased [21]. Apparently, presenting information about the harms and benefits of breast cancer screening in an understandable and balanced way is a complex matter [22]. We assessed the level of decision-relevant knowledge for informed decisions on participation in breast cancer screening among a sample of women that was representative for the Dutch population in terms of the distribution of level of education. Most first-time invitees for breast cancer screening had sufficient knowledge (according to the current criterion), were positive about attending the screening and planned to participate in it. As a consequence, most of these women made an informed choice about screening uptake. Almost all respondents knew the purpose of breast cancer screening. Nevertheless, we found that, overall, respondents knew less about some of the unfavourable effects of screening than about other topics. Almost one third of the responding women thought that a normal mammogram guaranteed absence of breast cancer, and one third were not aware of the risk of over-diagnosis. If we assume that women did not respond to the questionnaire because they did not know the answers, then, with the criterion used, only
357
half of the total group of invited women had sufficient knowledge. This proportion can be considered the minimum level of knowledge in this population. Women in our study received the national information leaflet that was used for that purpose at the time, which had not been drawn up on the basis of our expertdefined information. For example, it did mention the possibility of over-diagnosis. Also note that not everyone who receives an information leaflet is likely to read it thoroughly, if at all. To our knowledge there have been no other studies so far that assessed the level of informed choice among first-time invitees for breast cancer screening in the general population. Some studies did evaluate the effects of information materials on knowledge and informed choice in breast cancer screening [23,17,24,25]. However, the focus in these studies was different, aimed at enhancing decision-making among women in age groups for which the screening was not initially recommended. Our study assessed the level of informed decision-making within the current practice of breast cancer screening. Previous studies also differed in terms of the definition and operationalisation of the content of decisionrelevant knowledge. Although the knowledge measure in most of these studies included items about some of the risks and benefits of screening, often numerical, these measures were not based on expert-defined decision-relevant knowledge as in our study. A major strength of our study is that the content of the knowledge measure was based on an agreed minimum of decisionrelevant knowledge derived from relevant experts and recent evidence, which has not previously been defined in a systematic way. According to these experts, many topics constitute essential information for all invited women, including information about the unfavourable effects of screening such as false positive and false negative results, and over-diagnosis, if presented in a simple and understandable way. Gøtzsche et al. concluded that information leaflets fail to address many of the unfavourable effects of screening [26]. As compared with the information leaflet developed by Gøtzsche et al. (www.cochrane.dk), our expert-based information contained almost all topics concerning the unfavourable effects of screening, including over-diagnosis. There were some differences in the numerical information presented, due to country-specific data, but overall the content of the information too was very similar. For example, the text about over-diagnosis says: ‘if we detect breast cancer, you will usually receive treatment. However, breast cancer sometimes grows so slowly that it would not have caused you any problems during your life. In that case, the treatment would have been unnecessary.’. Because of the current evidence in the Netherlands, the only topic not included in the expert-based information was extra surgery and after-treatment. In a Dutch study, it was found that the proportion of mastectomies among the target age group had decreased by 25% in 2009, two years after the full implementation of the nation-wide screening programme for women aged 50–69 years [27]. More importantly, mastectomy rates per 100,000 women aged 50–69 years decreased after the prevalence peak in the initial stage of the programme, in contrast to younger and older breast cancer patients in whom mastectomy rates remained constant or even slightly increased [28]. Whereas Wald recommended that information in leaflets should be quantitative and specific rather than general [11], our experts advised to restrict information to simple messages and to minimize the use of numerical values. For that reason, we did not include numerical items. Although Gøtzsche’s leaflet did contain numerical information, this was presented separately from the basic message. This study has some limitations. Despite the use of anonymous questionnaires and pre-stamped envelope to return the questionnaire, the response rate was 50%, probably due to the timing of the study (which coincided with the summer holidays) and the fact
358
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
that we did not send reminders. However, this is in accordance with other survey research among general populations in the Netherlands, where a trend towards lower response rates was found [29]. Educational level might be correlated with the level of sufficient knowledge and informed choice. However, there was probably no responder bias related to this as the distribution of educational level reflects that of the general Dutch population, of which 60% have an intermediate level of education (VMBO) maximally. Within the context of this study to measure informed choice anonymously, it was not feasible to ask women’s consent for obtaining individual data on actual screening uptake from registries to link with their informed choice outcomes. Therefore, we did not measure actual screening uptake, which in the Netherlands is 80% for women invited for breast cancer screening for the first time [28]. Possible selective response may have resulted in an overrepresentation of women with positive feelings about the screening. It is also possible that the measurement of attitudes was limited. Attitude refers to one’s like or dislike towards a specific entity, and measurement should therefore be specific to context, action and time [30]. Attitude is assumed to have a cognitive, affective and behavioural component, which should be included in the measure [30]. Although we adopted the attitude measure from the frequently used MMIC [19], the instruction may not be specific enough. Possibly, the affective component in the attitude measure was underrepresented. For example, there were no items asking whether respondents feel that the screening is unpleasant or embarrassing. There are some issues for further research. General agreement is needed on the content of decision-relevant knowledge for informed choice about participating in breast cancer screening and on who should define this knowledge [31]. Ideally, a knowledge measure based on well-defined decision-relevant knowledge will enable valid measurements regarding whether a person has sufficient knowledge or not. Given the experts’ opinion that information about many topics is essential, probably a stricter criterion for correct answers should be used if the items represent the minimally required knowledge. Psychometric methods as used in the development of educational tests might help to construct reliable and valid knowledge measures. To develop information materials or other interventions to support informed decision making in breast cancer screening, further steps are needed. However, the deliberation process of how people should achieve good decisions remains poorly understood [32]. Therefore, future research should focus on how information should be presented to enable the acquirement of knowledge that can be used for personal deliberation of options to enable a free and autonomous choice. There are many decision-making theories that may be of help in this research [33]. 4.2. Conclusion Of all responding first-time invitees for breast cancer screening, 88% made an informed choice about participation, as measured against the current criterion for sufficient knowledge. Women had sufficient relevant knowledge about the benefits and harms in general, but less knowledge about some specific harms, namely the possibility of a false negative mammogram or over-diagnosis. 4.3. Practice implications To further increase informed choices in breast cancer screening, information on possible harms is an issue that merits special attention. Information materials based on well-defined decisionrelevant knowledge are essential. A summary of expert-defined information is now available.
Acknowledgements Van Agt, Fracheboud, Van der Steen and De Koning have no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous three years, any other relationships or activities that could appear to have influenced the submitted work. Only De Koning declares financial activities outside the submitted work, consisting of a grant from a German insurance company, SCOR Global Life, paid to his institution for forecasting age specific incidences for carcinoma in situ of the breast and invasive breast cancer under the assumption of various screening scenarios. We would like to thank all experts who participated in the expert consultation. We are grateful to Mr. E.J. Kamp, communication adviser of Bevolkingsonderzoek Zuid-West (formerly BoBWest), for his kind assistance in the data collection. Sadly, he died in 2010. This research was funded by the Centre for Population Screening (CvB) of the National Institute for Public Health and the Environment (RIVM). The CvB was not in any way involved in the study design, in the collection, analysis and interpretation of the data, in the writing of the report or in the decision to submit the paper for publication. References [1] Berry DA, Cronin KA, Plevritis SK, Fryback DG, Clarke L, Zelen M, et al. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 2005;353:1784–92. [2] Blanks RG, Moss SM, McGahan CE, Quinn MJ, Babb PJ. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1990–8: comparison of observed with predicted mortality. Brit Med J 2000;321:665–9. [3] Otto SJ, Fracheboud J, Looman CW, Broeders MJ, Boer R, Hendriks JH, et al. Initiation of population-based mammography screening in Dutch municipalities and effect on breast-cancer mortality: a systematic review. Lancet 2003;361:1411–7. [4] Irwig L, McCaffery K, Salkeld G, Bossuyt P. Informed choice for screening: implications for evaluation. Brit Med J 2006;332:1148–50. [5] Minister of Health Letter about screening [Kaderbrief screening]. The Hague; 2008 [6] General Medical Council Seeking Patients’ Consent: the Ethical Considerations. London; 1998 [7] Beauchamp TL, Childress JF. Principles of biomedical ethics, 5th edition, Oxford University Press; 2001. [8] Raffle AE. Information about screening – is it to achieve high uptake or to ensure informed choice? Health Expect 2001;4:92–8. [9] Marteau TM, Kinmonth AL. Screening for cardiovascular risk: public health imperative or matter for individual informed choice? Brit Med J 2002;325: 78–80. [10] Austoker J. Gaining informed consent for screening is difficult—but many misconceptions need to be undone. Brit Med J 1999;319:722–3. [11] Marteau TM, Dormandy E, Michie S. A measure of informed choice. Health Expect 2001;4:99–108. [12] National Cancer Institute NCI. Designing print materials: a communications guide for breast cancer screening. Bethesda (MD): NIH; 2007. [13] Schoonen HM, van Agt HM, Essink-Bot ML, Wildschut HI, Steegers EA, de Koning HJ. Informed decision-making in prenatal screening for Down’s syndrome: what knowledge is relevant? Patient Educ Couns 2011;84:265–70. [14] Wald N. Information leaflets in medical screening. J Med Screen 2006;13:109. [15] Schoonen HM, Essink-Bot ML, Van Agt HM, Wildschut HI, Steegers EA, De Koning HJ. Informed decision-making about the fetal anomaly scan: what knowledge is relevant? Ultrasound Obstet Gynecol 2011;37:649–57. [16] Michie S, Dormandy E, Marteau TM. The multi-dimensional measure of informed choice: a validation study. Patient Educ Couns 2002;48:87–91. [17] Mathieu E, Barratt AL, McGeechan K, Davey HM, Howard K, Houssami N. Helping women make choices about mammography screening: an online randomized trial of a decision aid for 40-year-old women. Patient Educ Couns 2010;61:63–72. [18] Dormandy E, Michie S, Hooper R, Marteau TM. Informed choice in antenatal Down syndrome screening: a cluster-randomised trial of combined versus separate visit testing. Patient Educ Couns 2006;61:56–64. [19] Dormandy E, Tsui EY, Marteau TM. Development of a measure of informed choice suitable for use in low literacy populations. Patient Educ Couns 2007;66:278–95. [20] Gummersbach E, Piccoliori G, Zerbe CO, Altiner A, Othman C, Rose C, et al. Are women getting relevant information about mammography screening for an
H. van Agt et al. / Patient Education and Counseling 89 (2012) 353–359
[21] [22]
[23]
[24]
[25]
[26]
informed consent: a critical appraisal of information brochures used for screening invitation in Germany, Italy, Spain and France. Eur J Public Health 2010;20:409–14. Fox R. Informed choice in screening programmes: do leaflets help? A critical literature review. J Public Health (Oxf) 2006;28:309–17. Zapka JG, Geller BM, Bulliard JL, Fracheboud J, Sancho-Garnier H, BallardBarbash R, et al. Print information to inform decisions about mammography screening participation in 16 countries with population-based programs. Patient Educ Couns 2006;63:126–37. Mathieu E, Barratt A, Davey HM, McGeechan K, Howard K, Houssami N. Informed choice in mammography screening: a randomized trial of a decision aid for 70-year-old women. Arch Intern Med 2007;167:2039–46. Lewis CL, Pignone MP, Sheridan SL, Downs SM, Kinsinger LS. A randomized trial of three videos that differ in the framing of information about mammography in women 40 to 49 years old. J Gen Intern Med 2003;18:875–83. Haakenson CP, Vickers KS, Cha SS, Vachon CM, Thielen JM, Kircher KJ, et al. Efficacy of a simple, low-cost educational intervention in improving knowledge about risks and benefits of screening mammography. Mayo Clin Proc 2006;81:783–91. Gøtzsche PC, Jorgensen KJ. The breast screening programme and misinforming the public. J R Soc Med 2011;104:361–9.
359
[27] Fracheboud J, Groenewoud JH, de Koning HJ. Fifteen years of population-based breast cancer screening in the Netherlands. Semin Breast Dis 2007;10:72–82. [28] National Evaluation Team for Breast cancer screening in the Netherlands N. National evaluation of breast cancer screening in the Netherlands 1990–2007. Rotterdam: Erasmus University Medical Center, Dept. of Public Health. Report No.: Twelfth report; 2009 [29] Smeets H, Janssen JPG. Using National registrations to correct for selctive nonresponse. Political preference of ethnic groups. In: Statistics Canada Symposium 2001. 2001. p. 1–9. [30] Ajzen I. Attitudes, personality, and behavior. Buckingham: Open University Press; 1988. [31] Green JM, Hewison J, Bekker HL, Bryant LD, Cuckle HS. Psychosocial aspects of genetic screening of pregnant women and newborns: a systematic review. Health Technol Assess 2004;8:1–109. iii, ix–x. [32] Elwyn G, Frosch D, Volandes AE, Edwards A, Montori VM. Investing in deliberation: a definition and classification of decision support interventions for people facing difficult health decisions. Med Decis Making 2010;30: 701–11. [33] Durand MA, Stiel M, Boivin J, Elwyn G. Where is the theory? Evaluating the theoretical frameworks described in decision support technologies. Patient Educ Couns 2008;71:125–35.