- Email: [email protected]
Comparing patients’ and their partners’ preferences for adjuvant chemotherapy in early breast cancer
Patient Education and Counseling 72 (2008) 239–245 www.elsevier.com/locate/pateducou
Comparing patients’ and their partners’ preferences for adjuvant chemotherapy in early breast cancer Vlatka M. Duric a,*, Phyllis N. Butow b, Louise Sharpe b, Stephane Heritier a, Frances Boyle c, Jane Beith d, Nicholas R.C. Wilcken e, Alan S. Coates f,g, R. John Simes a, Martin R. Stockler a,d,* a
NHMRC Clinical Trials Centre, University of Sydney, NSW, Australia b School of Psychology, University of Sydney, NSW, Australia c Mater Hospital, North Sydney, NSW, Australia d Sydney Cancer Centre, NSW, Australia e Nepean Hospital, NSW, Australia f The Cancer Council, Australia g School of Public Health, University of Sydney, NSW, Australia
Received 18 July 2007; received in revised form 4 February 2008; accepted 23 February 2008
Abstract Objective: Preferences of women who had completed adjuvant chemotherapy for early breast cancer were compared with those of their partners by determining the smallest benefits they judged sufficient to make adjuvant chemotherapy worthwhile. Methods: Forty-six women and their partners were interviewed separately, 3–34 months after completing adjuvant chemotherapy. Preferences were elicited using four sets of validated, hypothetical trade-off scenarios about the possible benefits of adjuvant chemotherapy on survival times (5 and 15 years) and survival rates (65% and 85% at 5 years). Agreement within couples was assessed with the intraclass correlation coefficient (ICC). Associations between baseline characteristics, preferences, and agreement within couples were assessed with linear regression after normal score transformation. Results: The mean age of the women was 57 years and of their partners’ was 60. Most couples were married (91%). Benefits of an extra 1 day or 0.1% were judged sufficient to make adjuvant chemotherapy worthwhile by 59–72% of women and 54–59% of partners. Agreement was exact in 35–41% of couples and approximate in 59–83%. Agreement was better for scenarios with a worse prognosis (ICC 0.67 and 0.35) than for scenarios with a better prognosis (ICC 0.13 and 0.05). Having dependent children was associated with partners requiring larger benefits but patients requiring smaller benefits to make adjuvant chemotherapy worthwhile (interaction P = 0.001). Conclusion: Patients’ and partners’ preferences for adjuvant chemotherapy differed and were influenced by considerations other than length and quality of life. Practice implications: Clinicians who are aware of these differences can tailor discussions to ensure that all interested parties understand and agree on the goals and benefits of treatment. # 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Adjuvant chemotherapy; Early breast cancer; Patient preferences; Partner preferences
1. Introduction
* Corresponding authors at: NHMRC Clinical Trials Centre, University of Sydney, Locked Bag 77, Camperdown, NSW 2050, Australia. Tel.: +61 2 9562 5000; fax: +61 2 9565 1863. E-mail addresses: [email protected] (V.M. Duric), [email protected] (M.R. Stockler). 0738-3991/$ – see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2008.02.021
Medical decision-making is often depicted as a dialogue between doctor and patient. In reality, however, others are often involved in decision-making [1,2]. For example, the decision to participate in a clinical trial was found to be influenced by approximately 10 different people, including the family doctor, the surgeon, the spouse or life partner, and family members [3]. Patients’ families may actively participate in decision-making,
240
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
or at least influence it. Partners probably have the most intimate knowledge of a patients’ circumstances and also share concerns such as parenting, work aspirations and financial issues. Clinical practice guidelines recommend that clinicians invite and endorse the involvement of partners and other supporters in medical decision-making [4]. Previous studies of preferences for treatments have compared patients’ preferences with those of health professionals and lay people and have found that they differ substantially [5]. Some research has focused on partners as proxies for assessing patients’ quality of life [6–10], and on the influence of the partner in preferences for prostate cancer treatments [11], but the concordance of patients’ and their partners’ preferences for adjuvant chemotherapy in early breast has not yet been evaluated. The average benefits conferred by adjuvant chemotherapy to groups of women with early breast cancer are well established and beyond reasonable doubt, but the actual benefit for any particular woman cannot be known [23]. For some women, relapse and death will occur despite chemotherapy, while for others, chemotherapy is not required, because disease has already been cured by surgery. However, chemotherapy causes side effects in all who get it. Our systematic review of patients’ preferences for adjuvant chemotherapy for early breast cancer showed that many women judged small benefits sufficient to make it worthwhile despite a wide variety of study types and methods [12]. Additional survival time of 6 months was judged sufficient by half the women, and additional an 12 months was judged sufficient by three quarters; a 1% improvement in the chance of survival or cure was judged sufficient by half the women, and a 5% improvement was judged sufficient by three quarters [12]. The purpose of this study was to compare women and their partners’ preferences for adjuvant chemotherapy in early breast cancer. We were also interested in determining levels and predictors of agreement within couples. 2. Methods 2.1. Participants and recruitment This study is part of a larger, continuing study of patients’ preferences for adjuvant chemotherapy in early breast cancer in an international randomized trial [13]. The study described in this paper started after recruitment for the larger study had already begun. In the larger study, 126 consecutive women who had adjuvant chemotherapy for early breast cancer were recruited from four tertiary referral cancer centres in metropolitan Sydney. We interviewed 46 of the last 54 women recruited who had partners (that is, a spouse or de facto spouse; response rate of 85%, 46 of 54). Patients were asked directly if they had a spouse or partner. Consent to approach and interview partners was first obtained from each woman and then separately from her partner. Women were interviewed 3–34 months after finishing their treatment. Women and their partners were interviewed separately. The interviewer did not disclose the patient’s preferences to her partner or vice versa. The project was approved by the human
research ethics committees of all participating institutions. All participants gave written, informed consent. 2.2. Preferences interview Preferences were elicited with a structured and scripted interview using the trade-off method. Participants were asked a series of questions based on hypothetical clinical scenarios. Their development and validation are detailed elsewhere [13]. The interview took about 45 min to complete. All participants were interviewed by the same clinical psychologist (VMD). There were two different types of question: one about adding time to a given life expectancy (‘survival time’) and the other about adding to the percent chance of surviving a fixed length of time (‘survival rate’). There were two scenarios for each type of question, with different baselines of survival chosen to reflect better and worse prognoses. The baselines were 5 years or 15 years for the survival time questions, and a 65% or 85% chance of 5-year survival for the survival rate questions. Participants were asked a series of questions to determine the minimum amount of extra time (beyond 5 years or 15 years) and the minimum extra chance of surviving 5 years (beyond 65% or 85%) that would be sufficient to make adjuvant chemotherapy worthwhile. The script emphasized that the choices were hypothetical; the numbers did not apply to the individual women personally; there were no right or wrong answers; and that the answers would not affect the women’s medical care in any way. 2.3. Statistical methods Agreement within couples is described with proportions of exact agreement (when the woman’s preference was identical to her partner’s) and approximate agreement (when the woman and her partner’s preferences were within 1 year or 3% of one another). The criteria for approximate agreement were chosen to reflect differences that would not materially affect decisions about treatment and were based on conservative estimates of the benefits of adjuvant chemotherapy [23]. Agreement within couples was assessed with the intraclass correlation coefficient (ICC) [14]. The ICC was calculated using a two-way mixed effects model with absolute agreement, where couples were regarded as random effects, and raters (woman versus partner) were regarded as fixed effects [15,16]. Guidelines for interpreting the strength of agreement measured by the ICC are from Landis and Koch: 0.2–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 good, and 0.81–1.00 excellent [17,18]. The preferences data were highly skewed so the ICCs were also calculated using ranks (equivalent to a quadratically weighted kappa) [19]. Associations between baseline characteristics and preferences were assessed with linear regression after normal score transformation and corroborated by logistic regression as in our previous study [14]. Associations between baseline characteristics and agreement within couples were similarly tested with linear regression. The outcome variable for agreement within
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
241
couples for each scenario was the absolute difference between the smallest benefit judged sufficient by each woman and her partner for each scenario. Multivariable models were developed by using both forward selection and backward elimination, with entry and removal criteria of P < 0.05. Only variables selected by both methods were considered independently significant. The effect of having dependants on the preferences of women versus their partners was assessed by testing the interaction between rater (woman or partner) and having dependents (yes or no). Our systematic review of patients’ preferences for adjuvant chemotherapy for early breast cancer showed that many women judged small benefits sufficient to make it worthwhile despite a wide variety of study types and methods. Women who had dependants, good support, involved lymph nodes, and fewer side-effects considered smaller benefits sufficient than their counterparts; women who had adjuvant radiation therapy considered larger benefits necessary [12]. However, different groups of people have remarkably different preferences for chemotherapy and are inherently difficult to predict [13]. Because of the paucity of previous data we made and tested only three a priori hypotheses: Hypothesis (1) partners (like patients) would judge small benefits sufficient; Hypothesis (2) the benefits required by partners would be larger than those required by patients because previous studies have found that those closest to the experience of treatment require the smallest benefits (5); and, Hypothesis (3) agreement would be better within couples having dependents. The intent of the analysis was therefore more descriptive and exploratory rather than hypothesis-testing. All P-values are two-sided, and used to indicate levels of support for individual hypotheses, rather than as strict criteria for their acceptance or rejection.
Table 1 Baseline demographic characteristics of 46 couples
3. Results
3.2. Women’s and their partners’ preferences
3.1. Baseline characteristics
The benefits judged sufficient to make adjuvant chemotherapy worthwhile are summarized in Figs. 1 and 2. Plausible benefits (an extra 1 year in life expectancy or 3% in survival rate) were judged sufficient to make adjuvant chemotherapy worthwhile by most women (83–87%) and their partners (63– 89%). However, 59–72% of the women and 54–59% of the partners judged benefits of even 1 day or 0.1% sufficient to make adjuvant chemotherapy worthwhile, regardless of the baseline survival. At the other extreme, 2–7% women judged the maximum benefit offered insufficient to make their adjuvant chemotherapy worthwhile for the better-prognosis scenarios; and 4% of partners judged the maximum benefits offered insufficient to make adjuvant chemotherapy worthwhile for any but the worst scenario (5-year life expectancy).
Demographic characteristics are summarized in Table 1. Women’s mean age at the time of the interview was 57 years (range 37–71) and their partners’ was 60 (range 40–74) (Table 1). Most of the couples were married and had children. About half the couples had dependent children. About half the women were educated at a postsecondary level and about half were employed at the time of the interview. Most had a friend or relative who had died from cancer. Women’s baseline disease characteristics, menopausal status, and treatment details are summarized in Table 2. Two-thirds had involved lymph nodes, about half were premenopausal when diagnosed, but only 11% were premenopausal when interviewed. The average number of months between finishing chemotherapy and being interviewed was 12 months (range 3–34 months). Most women had 3–6 months of chemotherapy including an anthracycline. Most also had adjuvant endocrine therapy and adjuvant radiation therapy. Breast cancer had recurred before the interview in 6 of the 46 women (13%).
Characteristic
%
Age of woman (years) 37–40 41–50 51–60 61–69 70–71
2 20 40 36 2
Age of partner (years) 40 41–50 51–60 61–69 70–74
3 8 37 42 10
Marital status of couple Married De facto married Other
91 8 1
Have children Have dependent children Woman at home during chemotherapy Friend or relative died of cancer
83 44 96 93
Women’s highest level of completed education Junior high school Senior high school University or college degree Trade or technical qualification
7 33 48 12
Women’s employment status at interview Full-time employment Part-time or casual employment Unemployed Retired Home duties
37 20 7 17 19
3.3. Baseline characteristics associated with preferences Baseline characteristics associated with women’s and their partners’ preferences are summarised in Table 3. Tumor size was the only baseline characteristic associated with women’s preferences for adjuvant chemotherapy, but this association was
242
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
Table 2 Women’s disease, menopausal, and treatment characteristics at baseline (n = 46 women) Characteristic
%
Primary tumor size 2.0 cm 2.1–5.0 cm >5.0 cm
50 38 12
Axillary lymph nodes involved
71
Tumor grade 1 2 3
11 30 59
Premenopausal at time of surgery Premenopausal at time of interview Disease recurrence since adjuvant chemotherapy
57 11 13
Chemotherapy regimens AC alone AC then CMF AC then paclitaxel CMF alone
50 17 17 16
Adjuvant endocrine therapy Adjuvant radiation therapy
70 70
AC = anthracycline-based therapy, CMF = cyclophosphamide, methotrexate, and 5-fluorouracil.
Fig. 1. Cumulative proportions of women and partners judging whether adjuvant chemotherapy would be worthwhile for varying improvements in 5- and 15-year life expectancy (n = 46).
weak (P = 0.04) and may be spurious given the small sample size and large number of associations tested. Partners judged smaller benefits sufficient if they (P = 0.002) were younger, if the women were not employed at the time of the
Fig. 2. Cumulative proportions of women and partners judging whether adjuvant chemotherapy would be worthwhile for varying improvements in 65% and 85% chance of 5-year survival (n = 46).
interview (P = 0.002), and if the couple did not have dependent children (P = 0.001). The factors independently associated with partners judging smaller benefits sufficient in multivariable analysis were the woman was not employed at the time of the interview (P = 0.01) and the couple did not have any dependent children (P = 0.003). Having dependents was associated with partners judging larger benefits necessary to make adjuvant chemotherapy worthwhile, in the opposite direction to that hypothesized. Having dependent children was not significantly associated with women’s preferences, but the trend was in the same direction as in previous studies (women with dependent children judging smaller benefits sufficient), and in the opposite direction to that of their partners. Having dependent children had a significantly different effect on women than on their partners: having dependent children was associated with partners requiring larger benefits but patients requiring smaller benefits to make adjuvant chemotherapy worthwhile (interaction P = 0.001). Participants were asked their reasons for judging benefits of 1 day or 0.1% sufficient to make adjuvant chemotherapy worthwhile. We envisaged this to be a rhetorical question that would illustrate the scenarios [13]. We were surprised to find that many participants responded that chemotherapy would be worthwhile in this situation, and so we systematically recorded responses to this question. Responses were recorded verbatim and classified later into the following common themes or concepts: having adjuvant chemotherapy would reduce possible future regret at not having pursued all the available options; doubt about the accuracy of information provided and the reliability of doctors’ recommendations; and not feeling like
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
243
Table 3 Characteristics associated with women and partners judging smaller benefits sufficient to make adjuvant chemotherapy worthwhile Factor
Women
Partners
Linear univariable models P*
Best linear multivariable model P*
Linear univariable models P*
Younger age of women Younger age of partners Differences between ages of women and partners
0.9 0.6 0.2
– – –
0.03 0.002 0.6
If patient Had tertiary education Was not employed at time of interview Had a friend or relative who died from cancer
0.5 1.0 0.2
– – –
0.6 0.002 0.9
If the couple Had any children Did not have dependent children
0.5 0.2 a
– –
1.0 0.001a
Characteristics of disease and treatment Larger tumor size Lymph nodes involved Adjuvant radiotherapy given Adjuvant endocrine therapy given Disease recurrence before interview Shorter interval since chemotherapy
0.04 0.5 0.9 0.6 0.2 0.6
0.04 – – – – –
0.4 0.5 0.7 0.9 0.2 0.6
a *
Best linear multivariable model P*
0.01
0.003
Having dependent children is associated with partners judging larger benefits necessary, but patients judging smaller benefits sufficient. P-value <0.05 are indicated in bold.
they had a choice. Women and partners who did not feel like they had a choice seemed to dichotomize the choice into doing something versus doing nothing, and felt that they had to do something. A few women, but none of the partners, said they were motivated to extend their lives as much as possible for their children and families. 3.4. Agreement between women and their partners Agreement between women and their partners is summarised in Table 4. Depending on the scenario, agreement was exact for 35–41% of couples and approximate for 59–83%. Agreement was better for scenarios with a worse prognosis (ICC = 0.67 and 0.35) than scenarios with a better prognosis (ICC = 0.05 and 0.13). Women’s and their partners’ preferences differed significantly on the survival rate questions (P = 0.01 and 0.02) with partners requiring an average of 2–3% more, but not the survival time questions (P = 0.6 and 0.3). There were no significant associations between baseline characteristics and agreement (all P > 0.05).
4. Discussion and conclusion 4.1. Discussion Only about a third of the couples in our study were in exact agreement. However, agreement was better on scenarios with a worse prognosis. Women required smaller benefits than their partners if prognosis was expressed in survival rates, but not in survival times. More than half of the women and their partners judged an extra day or an extra 0.1% sufficient to make adjuvant chemotherapy worthwhile, regardless of the baseline prognosis. Nevertheless, there was much variability between women’s and their partners’ preferences. Partners judged smaller benefits sufficient if they were younger, if they did not have dependent children and if the woman was not employed when interviewed. Having dependents was associated with divergent preferences for women and their partners. These results are striking because they highlight the differences between patients’ and their partners’ preferences and the factors that motivate them. Other studies have also
Table 4 Agreement between women and partners (n = 46) Scenario
Exact agreement (%)
Approximate agreement (%)
Intraclass correlation coefficienta
Average difference (woman partner)
P-value for difference between women and partners
5-year life expectancy 15-year life expectancy 65% chance of 5-year survival 85% chance of 5-year survival
37 35 41 37
83 65 61 59
0.67 0.05 0.35 0.13
1 year 1 year 3% 2%
0.6 0.3 0.01 0.02
a This ICC was calculated on raw scores. ICC calculated on ranks was qualitatively similar but the coefficients were smaller because this method does not take into account the different intervals between levels (ICC ranks = 0.01–0.36).
244
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
found that concerns about parenting are important in decisionmaking about cancer chemotherapy [20]. Our results confirm the importance of parenting concerns, but suggest that they may motivate patients’ and their partners differently. Spouse’s proxy ratings in other studies were more accurate for aspects of quality of life that are easier to observe, like physical function and role function, than for aspects that are harder to observe, like fatigue and satisfaction [6–10]. Patients’ preferences are likely to be influenced by their individual situations, meanings and the interpretations they assign to their illness [21,21,22]. The effects of cancer and its treatment on a person’s psychological, social and physical resources are also uniquely personal, as are their hopes, fears, ethics, and values. These may be shared with life partners, but are abstract, intangible and difficult to observe. This may partly explain the variability we observed between women and their partners preferences. Our data raise doubts about the suitability of partners as surrogates for decision-making about adjuvant chemotherapy in early breast cancer. This may not be a primary issue for women from some western cultures, but it may be important for women from other cultures and settings where husbands have more influence on medical decisions. Patients and partners preferences differed systematically when benefits were expressed as percentage improvements in survival rates but not as additional days, months, or years in survival time. This may reflect greater familiarity with benefits expressed as improvements in survival rates, notwithstanding the difficulties of understanding such complex information. If women had been given a numerical prognosis by their oncologist, then it is likely that this would have been expressed as an absolute improvement in 5- or 10-year survival rates, because these figures are readily available from published metaanalyses [23] and prognostic tools Adjuvant! (www.adjuvantonline.com). The lack of a systematic difference between patients and partners when benefits were expressed as improvements in survival times suggests greater random error (poorer understanding), probably due to the unfamiliarity of considering benefits this way. An alternative explanation is that survival times are more concrete and easier to understand than survival rates. The main limitation of our study is its small sample size and homogeneity: some real associations may have been missed. For example, the lack of an association between preferences and either interval since completing chemotherapy or recurrence of disease might reflect the lack of variation in the former and the small number of the latter. Baseline characteristics did not have significant effects on women’s preferences, but the sample was younger, more educated and had more dependent children than our previous samples [13,24]. Our study was descriptive and exploratory. We did not try to determine if or how partners’ preferences influence patients’ preferences, or on how differences in opinion are resolved. We did not try to determine whether agreement was better among couples who had been together longer, communicated better, or shared decision-making more. Australians have access to government-subsidised healthcare which covers the vast majority of medical costs, including
drugs. Our findings may be less applicable to health-care settings where individuals must pay for treatment themselves. Other limitations of this kind of research have been discussed in detail previously [13]. Preferences were based on women’s and their partners’ recollections of what chemotherapy was like, and were likely to be influenced by psychological mechanisms, including adaptation, coping, and cognitive dissonance reduction. However, the interval of time since completing chemotherapy was not associated with either preferences or agreement, as has been the case in our previous larger studies of preferences [13]. Furthermore, because only women who had adjuvant chemotherapy were recruited, they are not representative of women (or their partners) considering whether to have adjuvant chemotherapy, some of whom declined. Women who chose to have adjuvant chemotherapy would probably judge smaller benefits sufficient than women who declined it. Lastly, subjects’ responses to generic, hypothetical scenarios may differ from their responses to specific, real situations. Our study’s value is in its demonstration of the variability, disagreement and unpredictability of patients’ and their partners’ preferences. Our results highlight the importance of exploring preferences in shared decision-making, and of testing assumptions about them because individuals may differ in the importance they give to benefits and harms, even if they share similar circumstances and values. 4.2. Conclusion The factors influencing people’s judgements about the benefits needed to make adjuvant chemotherapy worthwhile can be conceptualised along four dimensions: length of life, quality of life, circumstances and attitudes. The first two are familiar, obvious and often measured in clinical trials. Lengthening life (in the long term) is adjuvant chemotherapy’s main benefit; impairing quality of life (in the short term) is its main harm. The second two are private, individual, and usually ignored in clinical trials, but must explain a substantial part of the variability observed in our current and previous studies [13,24]. Circumstances include demographic, social and environmental factors. Attitudes include intrinsic values, personality, and other factors that influence how people interpret and react to situations. Future research should determine how these factors influence the preferences of patients, doctors, and significant others so that they can be incorporated into better descriptive models. Studies are also needed to help clinicians communicate better about the benefits and harms of treatments. Given that guidelines endorse the involvement of a support person in consultations and decision-making [11], studies are also needed of how clinicians might best elicit and deal with divergent preferences. 4.3. Practice implications The actual benefits of adjuvant chemotherapy in early breast cancer exceed the minimum judged sufficient by most women and their partners. For example, the web-based program Adjuvant! (www.adjuvantonline.com) estimates an absolute
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
survival benefit at 10 years of about 3% for a 55-year-old woman with a 1.5-cm primary, negative hormone receptors, and no involved lymph nodes treated with four cycles of AC; and of about 5% if two lymph nodes are involved. The benefits for such women increase to 5% and 10%, respectively with the addition of four cycles of paclitaxel. It is nonetheless sobering that many said that an extra day or 0.1% would be sufficient to make adjuvant chemotherapy worthwhile. Clinicians have their own thresholds for what improvements warrant a discussion about adjuvant chemotherapy. Previous data suggest that clinicians’ thresholds are higher than those of patients’ [5]. Our data suggest that most patients with early breast cancer (and their partners) would want to know about treatments that improved survival rates by only a few percentage points. However, our data do not help determine the benefits that make adjuvant chemotherapy worthwhile from a community or policy perspective. Recommendations to include partners and support people in decision-making about cancer treatment are widespread. Despite their shared circumstances, patients’ and their partners’ attitudes, motivations and preferences may differ substantially. Clinicians who are aware of these differences can tailor their discussions to ensure that all interested parties understand and agree on the goals and benefits of treatment. Acknowledgments This research was supported by grants from the National Health and Medical Research Council of Australia, the Cancer Council of Australia, and the Australia New Zealand Breast Cancer Trials Group. Vlatka Duric was supported by a PhD Scholarship from the National Health and Medical Research Council of Australia. Our thanks go to the study participants for their time and effort and the Consumer Advisory Panel of the Australia New Zealand Breast Cancer Trials Group for their help in piloting the interview. Thanks also to Haryana Dhillon for thoughtful discussions and ideas, large proportion of data entry, and, the ‘Companion Study’ name. References [1] Charles C, Gafni A, Whelan T. Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango) Social Sci Med 1997;44. Mar-692. [2] Reust CE, Mattingly S. Family involvement in medical decision making. Family Med 1996;28:39–45. [3] Sutherland HJ, da Cunha R, Lockwood GA, Till JE. What attitudes and beliefs underlie patients’ decisions about participating in chemotherapy trials? Med Decis Making 1998;18:61–9. [4] National Health and Medical Research Council of Australia. Psychosocial clinical practice guidelines: information, support and counselling for women with breast cancer; 2004. Available from URL: http://nhmrc. gov.au/publications/.
245
[5] Slevin ML, Stubbs L, Plant HJ, Wilson P, Gregory WM, Armes PJ, Downer SM. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses, and general public. Br Med J 1990;300:1458–60. [6] Sneeuw KC, Aaronson NK, Osoba D, Muller MJ, Hsu MA, Yung WK, Brada M, Newlands ES. The use of significant others as proxy raters of the quality of life of patients with brain cancer. Med Care 1997;35:490–506. [7] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Value of caregiver ratings in evaluating the quality of life of patients with cancer. J Clin Oncol 1997;15:1206–17. [8] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Comparison of patient and proxy EORTC QLQ-C30 ratings in assessing the quality of life of cancer patients. J Clin Epidemiol 1998;51:617–31. [9] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Evaluating the quality of life of cancer patients: assessments by patients, significant others, physicians and nurses. Br J Cancer 1999;81:87–94. [10] Sneeuw KC, Albertsen PC, Aaronson NK. Comparison of patient and spouse assessments of health related quality of life in men with metastatic prostate cancer. J Urol 2001;165:478–82. [11] Srirangam SJ, Pearson E, Grose C, Brown SC, Collins GN, O’Reilly PH. Partner’s influence on patient preference for treatment in early prostate cancer. BJU Int 2003;92:365–9. [12] Duric V, Stockler M. Patients’ preferences for adjuvant chemotherapy in early breast cancer: a review of what makes it worthwhile. Lancet Oncol 2001;2:691–7. [13] Duric VM, Stockler MR, Heritier S, Boyle F, Beith J, Sullivan A, Wilcken N, Coates AS, Simes RJ. Patients’ preferences for adjuvant chemotherapy in early breast cancer: what makes AC and CMF worthwhile now? Ann Oncol 2005;16:1786–94. [14] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420–8. [15] McGraw K, Wong S. Forming inferences about some intraclass correlations coefficients: correction. Psychol Methods 1996;1:390. [16] McGraw K, Wong S. Forming inferences about some intraclass correlation coefficients. Psychol Methods 1996;1:30–46. [17] Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–74. [18] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [19] Streiner DL, Norman GR. Health measurement scales: a practical guide to their development and use, 3rd ed., Oxford: Oxford University Press; 2003. [20] Kiebert GM, Stiggelbout AM, Kievit J, Leer JW, van de Velde CJ, de Haes HJ. Choices in oncology: factors that influence patients’ treatment preference. Qual Life Res 1994;3:175–82. [21] Llewellyn-Thomas HA. Investigating patients’ preferences for different treatment options. Can J Nurs Res 1997;29:45–64. [22] Llewellyn-Thomas HA, Thiel EC, McGreal MJ. Cancer patients’ evaluations of their current health states: the influences of expectations, comparisons, actual health status, and mood. Med Decis Making 1992;12: 115–22. [23] Early Breast Cancer Trialists’ Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15year survival: an overview of the randomised trials. Lancet 2005;365: 1687–717. [24] Simes RJ, Coates AS. Patient preferences for adjuvant chemotherapy of early breast cancer: how much benefit is needed? J Natl Cancer Inst 2001;Monographs:146–52.
Comparing patients’ and their partners’ preferences for adjuvant chemotherapy in early breast cancer Vlatka M. Duric a,*, Phyllis N. Butow b, Louise Sharpe b, Stephane Heritier a, Frances Boyle c, Jane Beith d, Nicholas R.C. Wilcken e, Alan S. Coates f,g, R. John Simes a, Martin R. Stockler a,d,* a
NHMRC Clinical Trials Centre, University of Sydney, NSW, Australia b School of Psychology, University of Sydney, NSW, Australia c Mater Hospital, North Sydney, NSW, Australia d Sydney Cancer Centre, NSW, Australia e Nepean Hospital, NSW, Australia f The Cancer Council, Australia g School of Public Health, University of Sydney, NSW, Australia
Received 18 July 2007; received in revised form 4 February 2008; accepted 23 February 2008
Abstract Objective: Preferences of women who had completed adjuvant chemotherapy for early breast cancer were compared with those of their partners by determining the smallest benefits they judged sufficient to make adjuvant chemotherapy worthwhile. Methods: Forty-six women and their partners were interviewed separately, 3–34 months after completing adjuvant chemotherapy. Preferences were elicited using four sets of validated, hypothetical trade-off scenarios about the possible benefits of adjuvant chemotherapy on survival times (5 and 15 years) and survival rates (65% and 85% at 5 years). Agreement within couples was assessed with the intraclass correlation coefficient (ICC). Associations between baseline characteristics, preferences, and agreement within couples were assessed with linear regression after normal score transformation. Results: The mean age of the women was 57 years and of their partners’ was 60. Most couples were married (91%). Benefits of an extra 1 day or 0.1% were judged sufficient to make adjuvant chemotherapy worthwhile by 59–72% of women and 54–59% of partners. Agreement was exact in 35–41% of couples and approximate in 59–83%. Agreement was better for scenarios with a worse prognosis (ICC 0.67 and 0.35) than for scenarios with a better prognosis (ICC 0.13 and 0.05). Having dependent children was associated with partners requiring larger benefits but patients requiring smaller benefits to make adjuvant chemotherapy worthwhile (interaction P = 0.001). Conclusion: Patients’ and partners’ preferences for adjuvant chemotherapy differed and were influenced by considerations other than length and quality of life. Practice implications: Clinicians who are aware of these differences can tailor discussions to ensure that all interested parties understand and agree on the goals and benefits of treatment. # 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Adjuvant chemotherapy; Early breast cancer; Patient preferences; Partner preferences
1. Introduction
* Corresponding authors at: NHMRC Clinical Trials Centre, University of Sydney, Locked Bag 77, Camperdown, NSW 2050, Australia. Tel.: +61 2 9562 5000; fax: +61 2 9565 1863. E-mail addresses: [email protected] (V.M. Duric), [email protected] (M.R. Stockler). 0738-3991/$ – see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2008.02.021
Medical decision-making is often depicted as a dialogue between doctor and patient. In reality, however, others are often involved in decision-making [1,2]. For example, the decision to participate in a clinical trial was found to be influenced by approximately 10 different people, including the family doctor, the surgeon, the spouse or life partner, and family members [3]. Patients’ families may actively participate in decision-making,
240
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
or at least influence it. Partners probably have the most intimate knowledge of a patients’ circumstances and also share concerns such as parenting, work aspirations and financial issues. Clinical practice guidelines recommend that clinicians invite and endorse the involvement of partners and other supporters in medical decision-making [4]. Previous studies of preferences for treatments have compared patients’ preferences with those of health professionals and lay people and have found that they differ substantially [5]. Some research has focused on partners as proxies for assessing patients’ quality of life [6–10], and on the influence of the partner in preferences for prostate cancer treatments [11], but the concordance of patients’ and their partners’ preferences for adjuvant chemotherapy in early breast has not yet been evaluated. The average benefits conferred by adjuvant chemotherapy to groups of women with early breast cancer are well established and beyond reasonable doubt, but the actual benefit for any particular woman cannot be known [23]. For some women, relapse and death will occur despite chemotherapy, while for others, chemotherapy is not required, because disease has already been cured by surgery. However, chemotherapy causes side effects in all who get it. Our systematic review of patients’ preferences for adjuvant chemotherapy for early breast cancer showed that many women judged small benefits sufficient to make it worthwhile despite a wide variety of study types and methods [12]. Additional survival time of 6 months was judged sufficient by half the women, and additional an 12 months was judged sufficient by three quarters; a 1% improvement in the chance of survival or cure was judged sufficient by half the women, and a 5% improvement was judged sufficient by three quarters [12]. The purpose of this study was to compare women and their partners’ preferences for adjuvant chemotherapy in early breast cancer. We were also interested in determining levels and predictors of agreement within couples. 2. Methods 2.1. Participants and recruitment This study is part of a larger, continuing study of patients’ preferences for adjuvant chemotherapy in early breast cancer in an international randomized trial [13]. The study described in this paper started after recruitment for the larger study had already begun. In the larger study, 126 consecutive women who had adjuvant chemotherapy for early breast cancer were recruited from four tertiary referral cancer centres in metropolitan Sydney. We interviewed 46 of the last 54 women recruited who had partners (that is, a spouse or de facto spouse; response rate of 85%, 46 of 54). Patients were asked directly if they had a spouse or partner. Consent to approach and interview partners was first obtained from each woman and then separately from her partner. Women were interviewed 3–34 months after finishing their treatment. Women and their partners were interviewed separately. The interviewer did not disclose the patient’s preferences to her partner or vice versa. The project was approved by the human
research ethics committees of all participating institutions. All participants gave written, informed consent. 2.2. Preferences interview Preferences were elicited with a structured and scripted interview using the trade-off method. Participants were asked a series of questions based on hypothetical clinical scenarios. Their development and validation are detailed elsewhere [13]. The interview took about 45 min to complete. All participants were interviewed by the same clinical psychologist (VMD). There were two different types of question: one about adding time to a given life expectancy (‘survival time’) and the other about adding to the percent chance of surviving a fixed length of time (‘survival rate’). There were two scenarios for each type of question, with different baselines of survival chosen to reflect better and worse prognoses. The baselines were 5 years or 15 years for the survival time questions, and a 65% or 85% chance of 5-year survival for the survival rate questions. Participants were asked a series of questions to determine the minimum amount of extra time (beyond 5 years or 15 years) and the minimum extra chance of surviving 5 years (beyond 65% or 85%) that would be sufficient to make adjuvant chemotherapy worthwhile. The script emphasized that the choices were hypothetical; the numbers did not apply to the individual women personally; there were no right or wrong answers; and that the answers would not affect the women’s medical care in any way. 2.3. Statistical methods Agreement within couples is described with proportions of exact agreement (when the woman’s preference was identical to her partner’s) and approximate agreement (when the woman and her partner’s preferences were within 1 year or 3% of one another). The criteria for approximate agreement were chosen to reflect differences that would not materially affect decisions about treatment and were based on conservative estimates of the benefits of adjuvant chemotherapy [23]. Agreement within couples was assessed with the intraclass correlation coefficient (ICC) [14]. The ICC was calculated using a two-way mixed effects model with absolute agreement, where couples were regarded as random effects, and raters (woman versus partner) were regarded as fixed effects [15,16]. Guidelines for interpreting the strength of agreement measured by the ICC are from Landis and Koch: 0.2–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 good, and 0.81–1.00 excellent [17,18]. The preferences data were highly skewed so the ICCs were also calculated using ranks (equivalent to a quadratically weighted kappa) [19]. Associations between baseline characteristics and preferences were assessed with linear regression after normal score transformation and corroborated by logistic regression as in our previous study [14]. Associations between baseline characteristics and agreement within couples were similarly tested with linear regression. The outcome variable for agreement within
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
241
couples for each scenario was the absolute difference between the smallest benefit judged sufficient by each woman and her partner for each scenario. Multivariable models were developed by using both forward selection and backward elimination, with entry and removal criteria of P < 0.05. Only variables selected by both methods were considered independently significant. The effect of having dependants on the preferences of women versus their partners was assessed by testing the interaction between rater (woman or partner) and having dependents (yes or no). Our systematic review of patients’ preferences for adjuvant chemotherapy for early breast cancer showed that many women judged small benefits sufficient to make it worthwhile despite a wide variety of study types and methods. Women who had dependants, good support, involved lymph nodes, and fewer side-effects considered smaller benefits sufficient than their counterparts; women who had adjuvant radiation therapy considered larger benefits necessary [12]. However, different groups of people have remarkably different preferences for chemotherapy and are inherently difficult to predict [13]. Because of the paucity of previous data we made and tested only three a priori hypotheses: Hypothesis (1) partners (like patients) would judge small benefits sufficient; Hypothesis (2) the benefits required by partners would be larger than those required by patients because previous studies have found that those closest to the experience of treatment require the smallest benefits (5); and, Hypothesis (3) agreement would be better within couples having dependents. The intent of the analysis was therefore more descriptive and exploratory rather than hypothesis-testing. All P-values are two-sided, and used to indicate levels of support for individual hypotheses, rather than as strict criteria for their acceptance or rejection.
Table 1 Baseline demographic characteristics of 46 couples
3. Results
3.2. Women’s and their partners’ preferences
3.1. Baseline characteristics
The benefits judged sufficient to make adjuvant chemotherapy worthwhile are summarized in Figs. 1 and 2. Plausible benefits (an extra 1 year in life expectancy or 3% in survival rate) were judged sufficient to make adjuvant chemotherapy worthwhile by most women (83–87%) and their partners (63– 89%). However, 59–72% of the women and 54–59% of the partners judged benefits of even 1 day or 0.1% sufficient to make adjuvant chemotherapy worthwhile, regardless of the baseline survival. At the other extreme, 2–7% women judged the maximum benefit offered insufficient to make their adjuvant chemotherapy worthwhile for the better-prognosis scenarios; and 4% of partners judged the maximum benefits offered insufficient to make adjuvant chemotherapy worthwhile for any but the worst scenario (5-year life expectancy).
Demographic characteristics are summarized in Table 1. Women’s mean age at the time of the interview was 57 years (range 37–71) and their partners’ was 60 (range 40–74) (Table 1). Most of the couples were married and had children. About half the couples had dependent children. About half the women were educated at a postsecondary level and about half were employed at the time of the interview. Most had a friend or relative who had died from cancer. Women’s baseline disease characteristics, menopausal status, and treatment details are summarized in Table 2. Two-thirds had involved lymph nodes, about half were premenopausal when diagnosed, but only 11% were premenopausal when interviewed. The average number of months between finishing chemotherapy and being interviewed was 12 months (range 3–34 months). Most women had 3–6 months of chemotherapy including an anthracycline. Most also had adjuvant endocrine therapy and adjuvant radiation therapy. Breast cancer had recurred before the interview in 6 of the 46 women (13%).
Characteristic
%
Age of woman (years) 37–40 41–50 51–60 61–69 70–71
2 20 40 36 2
Age of partner (years) 40 41–50 51–60 61–69 70–74
3 8 37 42 10
Marital status of couple Married De facto married Other
91 8 1
Have children Have dependent children Woman at home during chemotherapy Friend or relative died of cancer
83 44 96 93
Women’s highest level of completed education Junior high school Senior high school University or college degree Trade or technical qualification
7 33 48 12
Women’s employment status at interview Full-time employment Part-time or casual employment Unemployed Retired Home duties
37 20 7 17 19
3.3. Baseline characteristics associated with preferences Baseline characteristics associated with women’s and their partners’ preferences are summarised in Table 3. Tumor size was the only baseline characteristic associated with women’s preferences for adjuvant chemotherapy, but this association was
242
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
Table 2 Women’s disease, menopausal, and treatment characteristics at baseline (n = 46 women) Characteristic
%
Primary tumor size 2.0 cm 2.1–5.0 cm >5.0 cm
50 38 12
Axillary lymph nodes involved
71
Tumor grade 1 2 3
11 30 59
Premenopausal at time of surgery Premenopausal at time of interview Disease recurrence since adjuvant chemotherapy
57 11 13
Chemotherapy regimens AC alone AC then CMF AC then paclitaxel CMF alone
50 17 17 16
Adjuvant endocrine therapy Adjuvant radiation therapy
70 70
AC = anthracycline-based therapy, CMF = cyclophosphamide, methotrexate, and 5-fluorouracil.
Fig. 1. Cumulative proportions of women and partners judging whether adjuvant chemotherapy would be worthwhile for varying improvements in 5- and 15-year life expectancy (n = 46).
weak (P = 0.04) and may be spurious given the small sample size and large number of associations tested. Partners judged smaller benefits sufficient if they (P = 0.002) were younger, if the women were not employed at the time of the
Fig. 2. Cumulative proportions of women and partners judging whether adjuvant chemotherapy would be worthwhile for varying improvements in 65% and 85% chance of 5-year survival (n = 46).
interview (P = 0.002), and if the couple did not have dependent children (P = 0.001). The factors independently associated with partners judging smaller benefits sufficient in multivariable analysis were the woman was not employed at the time of the interview (P = 0.01) and the couple did not have any dependent children (P = 0.003). Having dependents was associated with partners judging larger benefits necessary to make adjuvant chemotherapy worthwhile, in the opposite direction to that hypothesized. Having dependent children was not significantly associated with women’s preferences, but the trend was in the same direction as in previous studies (women with dependent children judging smaller benefits sufficient), and in the opposite direction to that of their partners. Having dependent children had a significantly different effect on women than on their partners: having dependent children was associated with partners requiring larger benefits but patients requiring smaller benefits to make adjuvant chemotherapy worthwhile (interaction P = 0.001). Participants were asked their reasons for judging benefits of 1 day or 0.1% sufficient to make adjuvant chemotherapy worthwhile. We envisaged this to be a rhetorical question that would illustrate the scenarios [13]. We were surprised to find that many participants responded that chemotherapy would be worthwhile in this situation, and so we systematically recorded responses to this question. Responses were recorded verbatim and classified later into the following common themes or concepts: having adjuvant chemotherapy would reduce possible future regret at not having pursued all the available options; doubt about the accuracy of information provided and the reliability of doctors’ recommendations; and not feeling like
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
243
Table 3 Characteristics associated with women and partners judging smaller benefits sufficient to make adjuvant chemotherapy worthwhile Factor
Women
Partners
Linear univariable models P*
Best linear multivariable model P*
Linear univariable models P*
Younger age of women Younger age of partners Differences between ages of women and partners
0.9 0.6 0.2
– – –
0.03 0.002 0.6
If patient Had tertiary education Was not employed at time of interview Had a friend or relative who died from cancer
0.5 1.0 0.2
– – –
0.6 0.002 0.9
If the couple Had any children Did not have dependent children
0.5 0.2 a
– –
1.0 0.001a
Characteristics of disease and treatment Larger tumor size Lymph nodes involved Adjuvant radiotherapy given Adjuvant endocrine therapy given Disease recurrence before interview Shorter interval since chemotherapy
0.04 0.5 0.9 0.6 0.2 0.6
0.04 – – – – –
0.4 0.5 0.7 0.9 0.2 0.6
a *
Best linear multivariable model P*
0.01
0.003
Having dependent children is associated with partners judging larger benefits necessary, but patients judging smaller benefits sufficient. P-value <0.05 are indicated in bold.
they had a choice. Women and partners who did not feel like they had a choice seemed to dichotomize the choice into doing something versus doing nothing, and felt that they had to do something. A few women, but none of the partners, said they were motivated to extend their lives as much as possible for their children and families. 3.4. Agreement between women and their partners Agreement between women and their partners is summarised in Table 4. Depending on the scenario, agreement was exact for 35–41% of couples and approximate for 59–83%. Agreement was better for scenarios with a worse prognosis (ICC = 0.67 and 0.35) than scenarios with a better prognosis (ICC = 0.05 and 0.13). Women’s and their partners’ preferences differed significantly on the survival rate questions (P = 0.01 and 0.02) with partners requiring an average of 2–3% more, but not the survival time questions (P = 0.6 and 0.3). There were no significant associations between baseline characteristics and agreement (all P > 0.05).
4. Discussion and conclusion 4.1. Discussion Only about a third of the couples in our study were in exact agreement. However, agreement was better on scenarios with a worse prognosis. Women required smaller benefits than their partners if prognosis was expressed in survival rates, but not in survival times. More than half of the women and their partners judged an extra day or an extra 0.1% sufficient to make adjuvant chemotherapy worthwhile, regardless of the baseline prognosis. Nevertheless, there was much variability between women’s and their partners’ preferences. Partners judged smaller benefits sufficient if they were younger, if they did not have dependent children and if the woman was not employed when interviewed. Having dependents was associated with divergent preferences for women and their partners. These results are striking because they highlight the differences between patients’ and their partners’ preferences and the factors that motivate them. Other studies have also
Table 4 Agreement between women and partners (n = 46) Scenario
Exact agreement (%)
Approximate agreement (%)
Intraclass correlation coefficienta
Average difference (woman partner)
P-value for difference between women and partners
5-year life expectancy 15-year life expectancy 65% chance of 5-year survival 85% chance of 5-year survival
37 35 41 37
83 65 61 59
0.67 0.05 0.35 0.13
1 year 1 year 3% 2%
0.6 0.3 0.01 0.02
a This ICC was calculated on raw scores. ICC calculated on ranks was qualitatively similar but the coefficients were smaller because this method does not take into account the different intervals between levels (ICC ranks = 0.01–0.36).
244
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
found that concerns about parenting are important in decisionmaking about cancer chemotherapy [20]. Our results confirm the importance of parenting concerns, but suggest that they may motivate patients’ and their partners differently. Spouse’s proxy ratings in other studies were more accurate for aspects of quality of life that are easier to observe, like physical function and role function, than for aspects that are harder to observe, like fatigue and satisfaction [6–10]. Patients’ preferences are likely to be influenced by their individual situations, meanings and the interpretations they assign to their illness [21,21,22]. The effects of cancer and its treatment on a person’s psychological, social and physical resources are also uniquely personal, as are their hopes, fears, ethics, and values. These may be shared with life partners, but are abstract, intangible and difficult to observe. This may partly explain the variability we observed between women and their partners preferences. Our data raise doubts about the suitability of partners as surrogates for decision-making about adjuvant chemotherapy in early breast cancer. This may not be a primary issue for women from some western cultures, but it may be important for women from other cultures and settings where husbands have more influence on medical decisions. Patients and partners preferences differed systematically when benefits were expressed as percentage improvements in survival rates but not as additional days, months, or years in survival time. This may reflect greater familiarity with benefits expressed as improvements in survival rates, notwithstanding the difficulties of understanding such complex information. If women had been given a numerical prognosis by their oncologist, then it is likely that this would have been expressed as an absolute improvement in 5- or 10-year survival rates, because these figures are readily available from published metaanalyses [23] and prognostic tools Adjuvant! (www.adjuvantonline.com). The lack of a systematic difference between patients and partners when benefits were expressed as improvements in survival times suggests greater random error (poorer understanding), probably due to the unfamiliarity of considering benefits this way. An alternative explanation is that survival times are more concrete and easier to understand than survival rates. The main limitation of our study is its small sample size and homogeneity: some real associations may have been missed. For example, the lack of an association between preferences and either interval since completing chemotherapy or recurrence of disease might reflect the lack of variation in the former and the small number of the latter. Baseline characteristics did not have significant effects on women’s preferences, but the sample was younger, more educated and had more dependent children than our previous samples [13,24]. Our study was descriptive and exploratory. We did not try to determine if or how partners’ preferences influence patients’ preferences, or on how differences in opinion are resolved. We did not try to determine whether agreement was better among couples who had been together longer, communicated better, or shared decision-making more. Australians have access to government-subsidised healthcare which covers the vast majority of medical costs, including
drugs. Our findings may be less applicable to health-care settings where individuals must pay for treatment themselves. Other limitations of this kind of research have been discussed in detail previously [13]. Preferences were based on women’s and their partners’ recollections of what chemotherapy was like, and were likely to be influenced by psychological mechanisms, including adaptation, coping, and cognitive dissonance reduction. However, the interval of time since completing chemotherapy was not associated with either preferences or agreement, as has been the case in our previous larger studies of preferences [13]. Furthermore, because only women who had adjuvant chemotherapy were recruited, they are not representative of women (or their partners) considering whether to have adjuvant chemotherapy, some of whom declined. Women who chose to have adjuvant chemotherapy would probably judge smaller benefits sufficient than women who declined it. Lastly, subjects’ responses to generic, hypothetical scenarios may differ from their responses to specific, real situations. Our study’s value is in its demonstration of the variability, disagreement and unpredictability of patients’ and their partners’ preferences. Our results highlight the importance of exploring preferences in shared decision-making, and of testing assumptions about them because individuals may differ in the importance they give to benefits and harms, even if they share similar circumstances and values. 4.2. Conclusion The factors influencing people’s judgements about the benefits needed to make adjuvant chemotherapy worthwhile can be conceptualised along four dimensions: length of life, quality of life, circumstances and attitudes. The first two are familiar, obvious and often measured in clinical trials. Lengthening life (in the long term) is adjuvant chemotherapy’s main benefit; impairing quality of life (in the short term) is its main harm. The second two are private, individual, and usually ignored in clinical trials, but must explain a substantial part of the variability observed in our current and previous studies [13,24]. Circumstances include demographic, social and environmental factors. Attitudes include intrinsic values, personality, and other factors that influence how people interpret and react to situations. Future research should determine how these factors influence the preferences of patients, doctors, and significant others so that they can be incorporated into better descriptive models. Studies are also needed to help clinicians communicate better about the benefits and harms of treatments. Given that guidelines endorse the involvement of a support person in consultations and decision-making [11], studies are also needed of how clinicians might best elicit and deal with divergent preferences. 4.3. Practice implications The actual benefits of adjuvant chemotherapy in early breast cancer exceed the minimum judged sufficient by most women and their partners. For example, the web-based program Adjuvant! (www.adjuvantonline.com) estimates an absolute
V.M. Duric et al. / Patient Education and Counseling 72 (2008) 239–245
survival benefit at 10 years of about 3% for a 55-year-old woman with a 1.5-cm primary, negative hormone receptors, and no involved lymph nodes treated with four cycles of AC; and of about 5% if two lymph nodes are involved. The benefits for such women increase to 5% and 10%, respectively with the addition of four cycles of paclitaxel. It is nonetheless sobering that many said that an extra day or 0.1% would be sufficient to make adjuvant chemotherapy worthwhile. Clinicians have their own thresholds for what improvements warrant a discussion about adjuvant chemotherapy. Previous data suggest that clinicians’ thresholds are higher than those of patients’ [5]. Our data suggest that most patients with early breast cancer (and their partners) would want to know about treatments that improved survival rates by only a few percentage points. However, our data do not help determine the benefits that make adjuvant chemotherapy worthwhile from a community or policy perspective. Recommendations to include partners and support people in decision-making about cancer treatment are widespread. Despite their shared circumstances, patients’ and their partners’ attitudes, motivations and preferences may differ substantially. Clinicians who are aware of these differences can tailor their discussions to ensure that all interested parties understand and agree on the goals and benefits of treatment. Acknowledgments This research was supported by grants from the National Health and Medical Research Council of Australia, the Cancer Council of Australia, and the Australia New Zealand Breast Cancer Trials Group. Vlatka Duric was supported by a PhD Scholarship from the National Health and Medical Research Council of Australia. Our thanks go to the study participants for their time and effort and the Consumer Advisory Panel of the Australia New Zealand Breast Cancer Trials Group for their help in piloting the interview. Thanks also to Haryana Dhillon for thoughtful discussions and ideas, large proportion of data entry, and, the ‘Companion Study’ name. References [1] Charles C, Gafni A, Whelan T. Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango) Social Sci Med 1997;44. Mar-692. [2] Reust CE, Mattingly S. Family involvement in medical decision making. Family Med 1996;28:39–45. [3] Sutherland HJ, da Cunha R, Lockwood GA, Till JE. What attitudes and beliefs underlie patients’ decisions about participating in chemotherapy trials? Med Decis Making 1998;18:61–9. [4] National Health and Medical Research Council of Australia. Psychosocial clinical practice guidelines: information, support and counselling for women with breast cancer; 2004. Available from URL: http://nhmrc. gov.au/publications/.
245
[5] Slevin ML, Stubbs L, Plant HJ, Wilson P, Gregory WM, Armes PJ, Downer SM. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses, and general public. Br Med J 1990;300:1458–60. [6] Sneeuw KC, Aaronson NK, Osoba D, Muller MJ, Hsu MA, Yung WK, Brada M, Newlands ES. The use of significant others as proxy raters of the quality of life of patients with brain cancer. Med Care 1997;35:490–506. [7] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Value of caregiver ratings in evaluating the quality of life of patients with cancer. J Clin Oncol 1997;15:1206–17. [8] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Comparison of patient and proxy EORTC QLQ-C30 ratings in assessing the quality of life of cancer patients. J Clin Epidemiol 1998;51:617–31. [9] Sneeuw KC, Aaronson NK, Sprangers MA, Detmar SB, Wever LD, Schornagel JH. Evaluating the quality of life of cancer patients: assessments by patients, significant others, physicians and nurses. Br J Cancer 1999;81:87–94. [10] Sneeuw KC, Albertsen PC, Aaronson NK. Comparison of patient and spouse assessments of health related quality of life in men with metastatic prostate cancer. J Urol 2001;165:478–82. [11] Srirangam SJ, Pearson E, Grose C, Brown SC, Collins GN, O’Reilly PH. Partner’s influence on patient preference for treatment in early prostate cancer. BJU Int 2003;92:365–9. [12] Duric V, Stockler M. Patients’ preferences for adjuvant chemotherapy in early breast cancer: a review of what makes it worthwhile. Lancet Oncol 2001;2:691–7. [13] Duric VM, Stockler MR, Heritier S, Boyle F, Beith J, Sullivan A, Wilcken N, Coates AS, Simes RJ. Patients’ preferences for adjuvant chemotherapy in early breast cancer: what makes AC and CMF worthwhile now? Ann Oncol 2005;16:1786–94. [14] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420–8. [15] McGraw K, Wong S. Forming inferences about some intraclass correlations coefficients: correction. Psychol Methods 1996;1:390. [16] McGraw K, Wong S. Forming inferences about some intraclass correlation coefficients. Psychol Methods 1996;1:30–46. [17] Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–74. [18] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [19] Streiner DL, Norman GR. Health measurement scales: a practical guide to their development and use, 3rd ed., Oxford: Oxford University Press; 2003. [20] Kiebert GM, Stiggelbout AM, Kievit J, Leer JW, van de Velde CJ, de Haes HJ. Choices in oncology: factors that influence patients’ treatment preference. Qual Life Res 1994;3:175–82. [21] Llewellyn-Thomas HA. Investigating patients’ preferences for different treatment options. Can J Nurs Res 1997;29:45–64. [22] Llewellyn-Thomas HA, Thiel EC, McGreal MJ. Cancer patients’ evaluations of their current health states: the influences of expectations, comparisons, actual health status, and mood. Med Decis Making 1992;12: 115–22. [23] Early Breast Cancer Trialists’ Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15year survival: an overview of the randomised trials. Lancet 2005;365: 1687–717. [24] Simes RJ, Coates AS. Patient preferences for adjuvant chemotherapy of early breast cancer: how much benefit is needed? J Natl Cancer Inst 2001;Monographs:146–52.