Read the article and answer the questions
Program Evaluation Report Critique for Breakout Group Assignment Session 13 PURPOSE: The purpose of this assignment is to provide you with the opportunity to put your skills to the test by critiquing a program evaluation report. You should have already read the article and answered questions about the background and sampling strategy. DIRECTIONS: Now, we will divide into breakout groups. As a group answer the questions in the outline below fully and completely. DO NOT DIVIDE THE QUESTIONS UP AMONG YOU. EACH PERSON SHOULD BE PART OF THE PROCESS OF RESPONDING TO EACH QUESTION. Explain the rationale for each of your answers. Create a narrative that follows the outline below. Please use headings and subheadings in your paper. (Doing so helps you stay organized, and it helps your instructor find the information he/she is searching for quickly.) If the information to answer the question is not available in the article, please state this and comment on the author’s decision not to include (Does it make sense to leave out? Would including this have helped the reader in understanding the evaluation?) Describe the data collection method. This could include interviews/surveys/questionnaires, focus groups, observations, record reviews, etc. Consider the following questions: Where, when, by whom, and under what conditions were data collected? How might data be biased in any way because of these factors? What is one strength and one limitation of the data collection strategy? What are the strengths and limitations of the measurement strategy? Consider the following questions: Are operationalizations presented for each process goal/objective and each outcome goal/objective for the program? How clearly are they described? Select one variable and provide its operational definition. Is measurement purely quantitative, purely qualitative or mixed methods? Explain. For quantitative measurements, are standardized instruments with demonstrated reliability and validity used? Are any self-constructed measures used? In either case, what have researchers included to describe and/or improve reliability and validity of the measures? For qualitative measurements, are measurements flexible, but still require some specific data elements; that is, systematic, verifiable, consistently measured? What is said about the qualitative measurement reliability and validity? Do the data come from multiple perspectives (e.g., client, family members, service providers, staff, administration, etc.) or rely on one group of key informants? Explain. Describe the data analysis strategy . Consider the following questions: For quantitative data analysis, what types of statistics were run? Univariates (e.g., frequencies, means, medians, and modes), bi-variates (e.g., chi-squares, t-tests, ANOVAs, correlations), multivariate models (e.g., multiple linear regression or multiple logistical regression) For qualitative data, were themes developed by two or more coders? Was inter-rate reliability calculated? What steps are discussed to enhance qualitative rigor. If this was a summative evaluation, what are the key threats to internal validity (i.e., confidence that the intervention, as opposed to other factors, is responsible for any changes). Think about history, maturation, testing, instrumentation, statistical regression, experimental mortality, etc. If threats to internal validity are not relevant explain why. If this is a formative evaluation, name one research limitations. For both formative and summative how well does the sample represent the population? Think about sample size, selection biases, etc. What are other limitations of the evaluation? Findings/Results Consider the following questions: Provide a brief summary of what was found about one goal/objective of the study.. How are tables, charts, figures, and graphs, used to help the reader understand the overall findings? Be specific. If there are qualitative findings, how are they presented? Provide an example of how this is done for one qualitative finding. Conclusions/Implications What conclusions can you draw from this study? Identify one way for each group member to incorporate the findings, inferences, or conclusions from this evaluation into your current or future clinical social work practice. 3
Read the article and answer the questions
Evaluation of a Peer Support Program for Women with Breast CancerÐLessons for Practitioners JEFFREY DUNN, 1* SUZANNE K. STEGINGA, 1,2 STEFANO OCCHIPINTI 2 and KEITHIA WILSON 3 1Community Services, Queensland Cancer Fund, 553 Gregory Terrace, Spring Hill, Queensland 4004, Australia 2Australian Catholic University, McAuley Campus, 53 Prospect Rd, Mitchelton, Queensland 4053, Australia3Grith University, Mt Gravatt Campus, Messines Ridge Road, Mt Gravatt, Queensland 4122, Australia ABSTRACT The present study aimed to describe how a volunteer peer support service assists women with breast cancer, and provides guidelines for practitioners in the development and implementa- tion of such programmes. A two-phase evaluation of a breast cancer peer support program was undertaken to describe important attributes of the peer support intervention, the impact of the volunteer visit on women’s self-reports of anxiety, and key indicators of a successful volunteer visit. Phase 1 included focus groups with 57 women previously treated for breast cancer. Phase 2 included a survey of 245 women also treated previously for breast cancer and visited by a Breast Cancer Support Volunteer. The key aspect of the peer support process was the bond of common experience leading to a decrease in social isolation, an increase in optimism about the future and reassurance about personal reactions and femininity. It is recommended that peer support programmes should aim to time support visits to coincide with the time when patient support needs are highest, that volunteers need to be recruited from a range of backgrounds and matched to patients most similar to them in way of life, and that peer support services should be embedded in a broad network of community support services. Copyright# 1999 John Wiley & Sons, Ltd. Key words:breast cancer; peer support; social support; social comparison INTRODUCTION The diagnosis and treatment of cancer is a major life stress, which is followed by well-described psychological and physical sequelae (Cohen, 1982; Moorey and Greer, 1989). At diagnosis, patients may be o€ered a range of support services to assist them to cope e€ectively with treatment. One type of support that they may be o€ered is CCC 1052±9284/99/010013±10$17.50Received 5 November 1997 Copyright# 1999 John Wiley & Sons, Ltd.Accepted 14 April 1998 Journal of Community & Applied Social Psychology J. Community Appl. Soc. Psychol.9: 13±22 (1999) * Correspondence to: Je€ Dunn, Director of Community Service, Queensland Cancer Fund, PO Box 201, Spring Hill, Queensland 4004, Australia. peer support, that is, support from a person who has herself previously experienced cancer. A number of models have been advanced to explain how peer support programs help individuals cope with cancer. These models include social support (Mastrovito, Moynihan and Parsonnet, 1990) and social comparison (Wood, Taylor and Lichtman, 1985; van den Borne, Pruyn and van den Heuvel, 1987). Social support may facilitate adjustment to cancer by providing patients with the resources needed to engage in active coping strategies or problem solving (Lazarus and Folkman, 1984). Social support has been de®ned as encompassing emotional, tangible and informational support (Schaefer, Coyne and Lazarus, 1981). Emotional support is described as intimacy, reassurance, being able to con®de in another and feeling cared for. Tangible support involves practical and physical assistance. Informational support includes information and advice about the problem or stresses the individual is facing. By contrast, the social comparison model suggests that patients seek to evaluate their emotional and physical responses to their treatment by comparing themselves with others. This self-evaluation and comparison process may then lead to a decrease in uncertainty and anxiety. The Breast Cancer Support Service (BCSS) provides an example of this type of programme, where women who have themselves been treated for breast cancer o€er peer support to women currently being treated for breast cancer. Women are eligible to become BCSS volunteers after a period of one year has elapsed since their breast cancer treatment and following the completion of a training programme. Training includes medical content about breast cancer and available treatments, as well as workshops on the psychosocial e€ects of breast cancer and e€ective communication and helping skills. Following training, volunteers visit women recently diagnosed with breast cancer 3±5 days after surgery and this is followed by a telephone support call three weeks after the initial visit. The visit most often occurs in the hospital or treatment setting. In addition to peer support, the volunteer provides the patient with an information kit of education material and a temporary breast prosthesis or bra ®ller. This type of service has been developed throughout the world under the name of Reach to Recovery International, and indeed continues to extend to countries where breast cancer support services are in a developmental stage. First, the present study aimed to describe in lay terms the ways in which women perceive the Breast Cancer Support Service as helpful or unhelpful. This in-depth analysis of the helping relationship between the woman and her peer support volunteer aims to provide a description of the process that underlies peer support. Second, this descriptive evaluation is used as a basis for further identifying key aspects of a peer support programme that need to be considered by practitioners who develop and implement such programmes. METHOD The study evaluated the BCSS using a two-phase procedure, where the measurement instrument was ®rst developed from the women who receive the service. In this approach, the perspective of the patient is seen as expert and valid in evaluating the usefulness of this service. 14J. Dunnet al. Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) Phase 1 In the ®rst phase, a series of focus groups was held with a total of 57 women previously diagnosed and treated for primary breast cancer, who were visited by a BCSS volunteer. Focus group participants were recruited by a mailed invitation to the total population group of women visited in two distinct urban geographic areas and within three time-frames. It is estimated from breast cancer incidence ®gures that in the geographical areas chosen, 90% of all women diagnosed with breast cancer received the Breast Cancer Support Service and thus the population was predicted to be a representative group of breast cancer patients (Queensland Cancer Registry, 1990). The selected time-frames were 0±3 months post-surgery, 9±12 months post- surgery and 24±27 months post-surgery. Participation rates were 23% (0±3 months), 33% (9±12 months) and 14% (24±27 months). Classi®cation data was not collected for non-responders. Of the total group of 57 women, 25 were 0±3 months post- surgery, 26 women were 9±12 months post-surgery, and 6 women were 24±27 months post-surgery. Focus group participants were selected from these di€erent time-frames to provide a broad perspective on women’s experiences after breast cancer surgery (Morgan, 1989). In this sense, it was expected that women who were further from the time of treatment may provide di€erent insights to those women closer to the time of surgery. Table 1 describes treatment types and ages of participants. In all, six focus groups were conducted using a group process method, where women were ®rst asked to list privately how the volunteer visit was helpful or unhelpful. Next, this information was discussed in the group, to allow for clari®cation and expansion of individual issues. The focus group data was collated by two coders with experience in the develop- ment and evaluation of patient support programmes and themes that described helpful and unhelpful attributes of the volunteer visit were identi®ed. Coders com- pleted this task independently before consulting to reach ®nal agreement on the list of attributes. The items identi®ed in this procedure served as the categories of interest for the second phase of the study. In all, 16 items were identi®ed to describe the volunteer visits. Table 2 lists the items under the relevant conceptual headings, i.e. helpful and unhelpful social support and social comparison. Of all items, 12 items described the provision of social support. In addition, four items were included that described ways in which the volunteer was perceived as similar to the women visited. These were included on the basis of their value to women identi®ed from the focus group data, Table 1. Description of focus group participantsÐPhase 1 Description 0±3 Months post-surgery9±12 Months post-surgery24±27 Months post-surgery Mastectomy 19 16 4 Breast conservation 6 10 2 Mean age 52.3 years 46.8 years 60.7 years Age range 41±73 years 32±68 years 50±76 years Of all focus group participants, 13 women had received chemotherapy, 26 had received radiation therapy and 34 women were receiving tamoxifen therapy. For Phase 1 analysis,nˆ57. Evaluation of peer support in breast cancer15 Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) where women found it helpful to talk to a volunteer who was similar to them in age or life-style. Phase 2 For Phase 2, the items listed in Table 2 were used to develop a measure to evaluate ways in which the volunteer’s visit was helpful or unhelpful. For each item, respond- ents, were asked how true that item was for the volunteer visit, using a 7-point Likert Scale, with scale of 1 (not at all true), 4 (moderately true) to 7 (very true). As well, two questions were included to provide global scores on how the women perceived the volunteer visit. The ®rst question asked how helpful the volunteer visit was overall in assisting the woman to adjust to or cope with breast cancer surgery, on a scale of 1 (not at all helpful) to 7 (very helpful). The second question asked the women how anxious she felt after the volunteer visited her, on a scale of 1 (much less anxious) to 7 (much more anxious). The questionnaire rated 7.7 on the Flesch±Kincaid index for readability and so women who have completed primary school education have been able to comprehend the questionnaire. A population group of 350 women who had been treated for primary breast cancer not more than 16 weeks previously, and were visited by a BCSS volunteer in Brisbane and regional Queensland over a designated three-month period, were mailed the evaluation questionnaire. All women contacted were not more than four months post- surgery and so were expected to be in the recovery phase from surgery and able to complete the questionnaire. As previously, the population group was expected to be representative of women treated for breast cancer in Queensland. In all, 245 women responded, providing a 70% response rate. Uncompleted questionnaires were returned Table 2. Ratings for how true each item was for the volunteer who visited in Phase 2 survey Items Mean SD Median Helpful social support Made me feel I was not alone 6.42 1.40 7.00 Gave practical advice about bras, etc. 6.42 1.41 7.00 Reassured me that I was reacting normally 6.28 1.44 7.00 Made me feel hopeful about the future 6.22 1.38 7.00 Reassured me that I would look feminine again 6.17 1.59 7.00 Visited me at the time I most needed her 5.82 1.84 7.00 Explained to me about my treatment 5.39 2.13 7.00 Showed me how to do my exercises 4.54 2.54 5.00 Unhelpful social support Did not understand how I was feeling 2.64 2.37 1.00 Was not realistic about how I would cope 2.24 2.09 1.00 Was reluctant to talk about her personal experience 2.07 2.18 1.00 Left me feeling uncertain about the future 1.59 1.48 1.00 Was dicult to contact when I needed her 1.51 1.46 1.00 Social comparison Had an experience with cancer like me 6.20 1.57 7.00 Had a similar way of life to me 4.09 2.07 4.00 Was a di€erent age to me 3.19 2.43 2.00 Had di€erent beliefs to me 2.17 2.01 1.00 Scale of 1, not at all true; 4, moderately true; to 7, very true/range of 1±7. 16J. Dunnet al. Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) by 13 women. Response rates were equivalent for women who had a mastectomy and women who had breast conservation. Of all responses, 223 evaluation measures were suciently completed to be entered into the data analysis for the evaluation of the volunteer visit. Table 3 describes the women who participated in the second- phase survey. A diagnostic speci®c quality of life measure was also included with this questionnaire and results from this are reported elsewhere (Stegingaet al., 1998). Participation in the study was voluntary and responses were anonymous. RESULTS The ratings for each item describing the volunteer visit are listed by category in Table 3 in descending order from most true of the visit to least true. The items describing helpful attributes of the visit were most commonly endorsed by women as moderately true to very true. By contrast, unhelpful attributes of the visit were generally seen as not true by respondents. The mean score for the global helpfulness of the visit was 6.24 (SDˆ1.39), indicating that the visit was perceived as very helpful by most of the women visited. Using the criterion of a rating of less than the midpoint of 4, only 12 women rated the volunteer visit as unhelpful. For how anxious the woman felt after the volunteer visit, the mean score was 2.21 (SDˆ1.51), indicating that most women felt less anxious after the volunteer had visited them. Using the criterion of a rating of greater than the midpoint of 4, only 13 women reported feeling more anxious after the volunteer visit. A series of hierarchial multiple regressions were performed to assess which of the items describing the volunteer visit predicted the overall helpfulness rating for the visit, and the woman’s rating of her level of anxiety after the visit. For these analyses the global helpfulness and global anxiety items were the criterion variables, and the helpful social support items, the social comparison items and the unhelpful social support items were the predictor variables. For helpful social support, the strongest relationship with global helpfulness was found for the item, `visited me at the time I most needed her’ (p50.0001; see Table 4). Also, the items, `explained to me about my treatment’ and `reassured me I was acting normally’ were signi®cant (p50.05, p50.001, respectively). For social comparison the items, `had a similar way of life to me’ and `had an experience with cancer like me’ were signi®cantly correlated to global helpfulness (p50.0001,p50.001; see Table 5), while a signi®cant negative Table 3. Description of respondents in Phase 2 survey Description Value Time since surgery (weeks) 7.3 (SDˆ3.43) Age (years) 55 (SDˆ12.94) Mastectomy (%) 71 Breast conservation (%) 29 Radiation therapy (%) 29 Chemotherapy (%) 22 Tamoxifen (%) 42 For this analysis,nˆ245. Evaluation of peer support in breast cancer17 Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) relationship was found for the item, `had di€erent beliefs to me’ (p50.05). For the unhelpful social support items, the item `left me feeling uncertain about my future’ was negatively correlated to global helpfulness (p50.0001; see Table 6), as was the item, `was dicult to contact when I most needed her’ (p50.05). The helpful item, `visited me at the time I most needed her’ was predictive of feeling less anxious (p50.01; see Table 7). From the social comparison items, `had di€erent beliefs to me’ and `had a similar way of life to me’ were correlated to global anxiety Table 4. Multiple regression results for the helpful items on the Global Helpfulness Scale Item B Betasr 2 Showed me how to do my exercises 0.03 0.06 0.04 Made me feel not aloneÿ0.06ÿ0.06 0.04 Visited me at the time I most needed her 0.33 0.47*** 0.35 Made me feel hopeful about the future 0.16 0.17* 0.11 Gave me practical advice about brasÿ0.09ÿ0.11 0.06 Explained to me about my treatments 0.07 0.12* 0.09 Reassured me I was acting normally 0.22 0.22** 0.16 Reassured me that I could be feminine 0.07 0.10 0.05 Rˆ0.82*** R 2ˆ0.67 * p50.05;** p50.001;*** p50.0001. Table 5. Multiple regression results for the social comparison items on the Global Helpfulness Scale Item B Betasr 2 Had di€erent beliefs to meÿ0.12ÿ0.17* 0.16 Had a similar way of life to me 0.25 0.36*** 0.35 Had an experience with cancer like me 0.24 0.25** 0.24 Was a di€erent age to me 0.05 0.09 0.08 Rˆ0.53*** R 2ˆ0.28 * p50.05;** p50.001;*** p50.0001. Table 6. Multiple regression results for the unhelpful items on the Global Helpfulness Scale Item B Betasr 2 Was dicult to contact when I needed her 0.17 0.19* 0.17 Did not understand how I was feeling 0.01 0.02 0.02 Left me feeling uncertain about the futureÿ0.44ÿ0.50** 0.41 Was reluctant to talk about her experience 0.02 0.04 0.04 Was not realistic about how I would cope 0.01 0.02 0.01 Rˆ0.43** R 2ˆ0.19 * p50.05;** p50.0001. 18J. Dunnet al. Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) (p50.05; see Table 8). For the unhelpful social support items, `left me feeling uncertain about the future’ was predictive of feeling more anxious (p50.0001; see Table 9). DISCUSSION The results of the present evaluation suggest that women who are visited by the BCSS peer support service ®nd the service helpful and report feeling less anxious Table 7. Multiple regression results for the helpful items on the Global Anxiety Scale Item B Betasr 2 Showed me how to do my exercisesÿ0.04ÿ0.07 0.07 Made me feel not alone 0.03 0.03 0.02 Visited me at the time I most needed herÿ0.21ÿ0.25* 0.20 Made me feel hopeful about the futureÿ0.20ÿ0.18 0.12 Gave me practical advice about bras 0.03 0.03 0.02 Explained to me about my treatments 0.06 0.09 0.07 Reassured me I was acting normallyÿ0.16ÿ0.13 0.10 Reassured me that I could be feminineÿ0.02ÿ0.03 0.01 Rˆ0.46*** R 2ˆ0.21 * p50.01;** p50.0001. Table 8. Multiple regression results for the social comparison items on the Global Anxiety Scale Item B Betasr 2 Had di€erent beliefs to me 0.15 0.19* 0.18 Had a similar way of life to meÿ0.12ÿ0.16* 0.15 Had an experience with cancer like meÿ0.16ÿ0.15 0.14 Was a di€erent age to meÿ0.05ÿ0.07 0.06 Rˆ0.31** R 2ˆ0.1 * p50.05;** p50.005. Table 9. Multiple regression results for the unhelpful items on the Global Anxiety Scale Item B Betasr 2 Was dicult to contact when I needed herÿ0.12ÿ0.12 0.10 Did not understand how I was feelingÿ0.06ÿ0.09 0.07 Left me feeling uncertain about the future 0.35 0.34** 0.28 Was reluctant to talk about her experience 0.02 0.03 0.02 Was not realistic about how I could cope 0.02 0.02 0.02 Rˆ0.31* R 2ˆ0.1 * p50.005;** p50.0001. Evaluation of peer support in breast cancer19 Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) after the visit. These results are consistent with the ®ndings of recent evaluation of similar programmes in the UK and Canada (McCrum, 1996; National Cancer Institute of Canada and Canadian Cancer Society, 1995). Further, the BCSS is equally well evaluated by patients who receive breast-conserving surgery compared to patients who receive mastectomy. This ®nding is important when considering that the BCSS was ®rst initiated as a service speci®cally for women who had undergone a mastectomy. This suggests that changes in treatment modalities have not a€ected the relevance of this service to women undergoing treatment in the current treatment context. Clinicians have previously proposed that the key aspect of this type of support service is the bond of common experience (Rogers, Bauman and Metzer, 1985). This view is also supported by this evaluation. The items most often endorsed by respondents as describing their volunteer visit include feeling less alone, feeling more hopeful about the future and obtaining reassurance about personal reactions and femininity. Also, the most commonly endorsed comparison item referred to the shared experience of breast cancer. This was described by one participant in the focus group as `seeing someone still alive, well and living normally helped me see there was life after the operation’. Therefore, the way that the volunteer assists the women visited may be best conceptualized as occurring through both the process of social comparison and the provision of social support, in particular emotional support (e.g. `feeling less alone’, `obtaining reassurance’) and informational support (e.g. `explained to me about treatment’). In this study aspects of both social com- parison and social support were signi®cantly related to the overall helpfulness of the visit and to the lessening of women’s anxiety. This suggests that both concepts are important in an analysis of peer support, and this ®nding should be incorporated into the further research in this area. Further investigation of the psychometric properties of the evaluation measure used may provide further insight into the utility of these concepts in the investigation of peer support. An important variable in predicting both the helpfulness of the visit and feeling less anxious was the woman’s report that the volunteer visited at the time she most needed her. For the BCSS, the visit occurs in the early post-operative period (i.e. day 3±5 after surgery) while the woman is still in hospital. Receiving a visit from a volunteer before discharge from hospital has several advantages for women. First, the visit provides a woman with a contact point for social support in her community. Second, it provides the woman with advice and support about coping after discharge. Third, the woman visited is provided with patient education material and a temporary breast prosthesis to assist her in the acute phase of her treatment. This is a time when the woman’s information and support needs may be high and early assistance may be most important in assisting her to cope e€ectively. However, the current trend towards early discharge is making it increasingly dicult for programmes such as the BCSS to access women at this critical early period. As a result, future development of this service may more frequently require volunteers to visit women in their homes and this option brings with it practical diculties with transport issues. Overcoming this diculty will require the assistance of health professionals who utilize the service to ensure early referral. Contrary to expectations, our results suggest that age matching between the volunteer and patient is less important than the woman perceiving the volunteer as having a similar way of life and similar beliefs to her. In the service described, the 20J. Dunnet al. Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) priorities in matching volunteers to patients are principally age and treatment type. There are several reasons for this. First, referrals to the service come from treatment centre nurses who are frequently only able to provide basic medical and demographic information about the patient (certainly nothing describing the patient’s way of life or personal beliefs). Second, in the acute treatment setting, information about treat- ment may be more salient. Third, volunteer matching is limited by the availability of di€erent volunteers. That is, it is not always possible to ®nd a volunteer available to meet the individual requirements of every woman. However, future programme development could usefully address this issue by re®ning referral procedures and recruiting volunteers from as wide a range of backgrounds as possible. Dissatisfaction with the volunteer visit, that is, ®nding the visit unhelpful and feeling more anxious, was related to an increase in fears about the future. The present study does not provide sucient data to explain this result fully. It may be that for some women who are very anxious and fearful about the future, the BCSS is insucient as an intervention to decrease their fears and anxieties. However, this is not necessarily a criticism of the service but, perhaps, an indication of its limitations. For many patients, the standard care they receive from their treatment sta€ and the BCSS will meet their psychosocial needs. However, some women will experience psychological distress that will require additional intervention, such as psychological counselling, to assist them to cope with the stress of diagnosis and treatment. This point has already been made by various researchers (Andersen, 1992; Andersen, 1993; Meyer and Mark, 1995; Fallow®eld, 1996) and re¯ects the need for identi®cation of patients who are at greater risk of experiencing psychological morbidity after surgery and will require referral to appropriate services. Principally, however, this ®nding also highlights the importance of volunteer peer support services operating within a broad network of community support services, so that women who require more in-depth assistance can be appropriately referred. In conclusion, peer support programmes can play an important role in the support of patients with cancer. It is also evident that both social comparison and social support are useful concepts in describing how peer support works. In our study, women with breast cancer reported peer support services as helpful, and described feeling less anxious after the support visit. For practitioners, a number of factors should be considered when developing and implementing a peer support programme. First, the timing of the peer support visit is important and the visit may be most helpful if it occurs in the early post-operative period when treatment demands and support needs are greatest. Second, patients ®nd this support more helpful if they perceive the peer support person as similar to them. Therefore, volunteers should be recruited from di€erent backgrounds to cater for a broad range of clients. Finally, peer support programmes need to operate from within a broad network of community support services in order to facilitate more in-depth support when this is needed. REFERENCES Andersen, B. L. (1992) `Psychological interventions for cancer patients to enhance Quality of Life’,Journal of Consulting and Clinical Psychology,60, 552±568. Andersen, B. L. (1993) `Predicting sexual and psychological morbidity and improving the quality of life for women with gynaecological cancer’,Cancer,71, 1678±1690. Evaluation of peer support in breast cancer21 Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999) Cohen, N. (1982) `Psychosocial morbidity in cancer; a clinical perspective ‘, in J. Cohen, J. Cullen and R. M. Martin (eds).Psychosocial Aspects of Cancer, Raven press, New York, pp. 117±127. Fallow®eld, L. (1996) `Has psychosocial oncology helped in the management of women with breast cancer?’,The Breast Journal,2, 107±113. Lazarus, R. S. and Folkman, S. (1984)Stress, Appraisal and Coping, Springer, New York. Mastrovito, R., Moynihan, R. T. and Parsonnet, L. (1990) `Self-help and mutual support programs’, in J. C. Holland and J. H. Rowland (eds),Handbook of Psycho-oncology, Oxford University press, New York, pp. 502±507. McCrum, E. (1996)Evaluation of the Reach to Recovery Program, Ulster Cancer Foundation, Dublin. Meyer, J. G. and Mark, M. M. (1995) `E€ects of psychosocial interventions with adult cancer patients; a meta-analysis of randomised experiments;,Health Psychology,14, 101±108. Moorey, S. and Greer, S. (1989)Psychological Therapy for Patients with Cancer, Heinemann, Oxford. Morgan, D. C. (1989)Focus Groups as Qualitative Research, Sage, Newbury Park, CA. National Cancer Institute of Canada (NCIC) Centre for Behavioural Research and Program Evaluation and Canadian Cancer Society (1995)Reach to Recovery and Cansurmount Program Evaluation, NCIC, Toronto, Ontario. Queensland Cancer Registry (1990)Malignant Neoplasm of the Female Breast: Preliminary Unpublished Data, Queensland Health, Brisbane. Rogers, T. E., Bauman, C. J. and Metzger, L. (1985) `An assessment of the Reach to Recovery Program’,Ca: A Cancer Journal for Clinicians,35, 116±124. Schaefer, C., Coyne, T. and Lazarus, R. S. (1981) `The health-related functions of social support’,Journal of Behavioural Medicine,4, 381±406. Steginga, S. K., Occhipinti, S., Wilson, K. and Dunn, J. (1998) `Domains of distress: the experience of breast cancer in Australia’,Oncology Nursing Forum. van den Borne, H. W., Pruyn, J. E. A. and van den Heuvel (1987) `E€ects of contacts between cancer patients on their psychosocial problems’,Patient education and Counselling, 9, 33±51. Wood, J., Taylor, S. E. and Lichtman, R. R. (1985) `Social comparison in adjustment to breast cancer’,Journal of Personality and Social Psychology,49, 1169±1183. 22J. Dunnet al. Copyright# 1999 John Wiley & Sons, Ltd. J. Community Appl. Soc. Psychol.9: 13±22 (1999)