Psycho-Oncology Psycho-Oncology 20: 387–393 (2011) Published online 5 April 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pon.1740 The supportive care needs survey for partners and caregivers of cancer survivors: development and psychometric evaluation Afaf Girgis, Sylvie Lambert and Christophe Lecathelinais Centre for Health Research & Psycho-oncology (CHeRP), Cancer Council NSW, University of Newcastle & Hunter Medical Research Institute, Newcastle, Australia * Correspondence to: Centre for Health Research & Psycho-oncology, Cancer Council NSW & University of Newcastle, Room 230A, Level 2, David Maddison Building, Callaghan NSW 2308, Australia. E-mail: afaf.girgis@newcastle.edu.au Received: 19 May 2009 Revised: 7 February 2010 Accepted: 8 February 2010 Abstract Objective: Begin to test the psychometric properties of a measure designed to capture the multi-dimensional supportive care needs of cancer caregivers: the Supportive Care Needs Survey—Partners and Caregivers (SCNS-P&C). Methods: Draft SCNS-P&C items were developed with reference to the literature and existing instruments and reviewed for face and content validity. The final SCNS-P&C was then completed by 547 cancer caregivers. Psychometric analyses conducted included principal factor analysis, internal consistency, and construct validity through the known-group approach. Results: Factor analysis revealed four domains of needs: Health Care Service Needs, Psychological and Emotional Needs, Work and Social Needs, and Information Needs, with Cronbach’s a coefficients ranging from 0.88 to 0.94. Construct validity of the SCNS-P&C was partially supported. Across all domains, individuals with anxiety or depression were more likely to report at least one unmet moderate or high need in comparison to non-anxious or nondepressed participants. A greater proportion of younger participants experienced at least one unmet moderate or high need within the Psychological and Emotional Needs and Work and Social Needs domains. Proportion of reported unmet needs varied across cancer types for the Health Care Service Needs and Information Needs domains. Conclusions: The SCNS-P&C has the potential to comprehensively assess the range of caregivers’ supportive care needs, across the illness trajectory. Analyses supported the tool’s internal consistency and construct validity. The SCNS-P&C can be used by researchers and clinicians to determine caregivers’ unmet needs, prioritise health-care resources, and tailor supportive cancer care services accordingly. Copyright r 2010 John Wiley & Sons, Ltd. Keywords: caregivers; needs assessment; psycho-oncology; psychometrics; validity; cancer supportive care needs Introduction Cancer affects not only the life of the individual diagnosed but also the well-being of those close to them by imposing countless physical, psychosocial, and financial demands [1–4]. Recognition of the impact of a cancer diagnosis on caregivers has led contemporary approaches to psychosocial care to integrate in-depth documentation of their supportive care needs [5,6]. Such a comprehensive approach to cancer care offers caregivers an opportunity to call attention to their own needs and express the issues they want recognised [3]. Current evidence suggests that caregivers’ supportive care needs cluster in the domains of information [2,5–9], emotional/psychological [2,8,9], physical health [8,9], health-care professionals and cancer care [6,10], relationship with the patient [9], Copyright r 2010 John Wiley & Sons, Ltd. practical support [8,9], and legal and financial [9]. Despite researchers and clinicians efforts toward addressing caregivers’ needs, many of these reportedly remain unmet [1,2,6,9,11]. In a study by Soothill et al. [12], 28% of cancer caregivers identified three or more significant supportive care unmet needs. Though there are several supportive care needs assessments available, they have the following limitations in terms of large-scale use with a generic population of cancer caregivers. First, some of the measures focus on a specific domain of needs and provide a limited account of the broad range of caregivers’ supportive care needs [7]. Second, they focus on caregivers’ needs at a particular stage along the illness trajectory [9,10,13], restricting the tool’s suitability across the cancer continuum. Third, some measures appear 388 out-dated [14] and/or have no reported psychometric properties [12]. To date, there is no psychometrically robust measure that provides a comprehensive assessment of the multi-dimensional supportive care needs of cancer caregivers, across the illness trajectory. This article reports on the development and psychometric testing of a questionnaire designed to address this gap in the literature: the Supportive Care Needs Survey— Partners and Caregivers (SCNS-P&C). To our knowledge, this is the first study evaluating such a measure among a large population-based sample of caregivers of individuals diagnosed with the eight most incident cancers in Australia. A. Girgis et al. Procedure Cancer survivors were recruited into the Cancer Survival Study through the New South Wales and Victorian cancer registries. At 6–8 months postdiagnosis, participants were sent along with their own survey, a separate Partners and Caregivers Study information package, to pass on to their caregiver. Ethics approval was obtained from the University of Newcastle Human Research Ethics Committee. Interested caregivers returned their Partners and Caregivers Study consent form to the research team. Data collection Methods Item generation The SCNS-P&C items were based on: (1) a literature review to identify the main supportive care needs of caregivers of individuals diagnosed with cancer, (2) an examination of the existing tools assessing caregivers’ unmet needs, and (3) the adaptation of the items from the Supportive Care Needs Survey (SCNS) [15,16]. Initial items were reviewed for face and content validity by experts in psycho-oncology, members of the general public, and cancer caregivers of cancer survivors. Suggestion were integrated, which resulted in a 44-item tool that measures caregivers’ unmet needs across the information, health-care services, daily living, and psychological domains. Items are rated on a 5-point response scale (1 ‘No Need—Not applicable’ to 5 ‘Some Need—High’) which was modelled on the scale of the SCNS [15,16]. Construct validity and internal consistency Participants Caregivers were referred to the Partners and Caregivers Study by cancer survivors participating in the Cancer Survival Study conducted by the research team [17]. A caregiver was defined as a person nominated by the survivor as most involved in supporting them through the illness [6]. The study was named the ‘Partners and Caregivers’ Study to emphasise that a caregiver can include someone who performs hands-on care and/or provides emotional support. For the purpose of this article, the term ‘caregiver’ is used to encompass both partners and caregivers. All Englishspeaking caregivers, caring for or living with someone who had pathologically confirmed colorectal, female breast, prostate, lung, or head and neck (HN) cancer, or leukaemia, non-Hodgkin’s lymphoma (NHL), or melanoma, and able to provide written consent were eligible to participate in the Partners and Caregivers Study. Copyright r 2010 John Wiley & Sons, Ltd. Consenting caregivers were mailed a self-administered, scannable survey measuring the demographic characteristics and the health, financial, and psychosocial variables of interest, including unmet needs (SCNS-P&C) and anxiety and depression (Hospital Anxiety and Depression Scale). Participants returned their survey in the reply-paid envelope provided. Data analysis The psychometric analyses of the SCNS-P&C were undertaken using SAS Version 9. The responses 1 ‘No need—satisfied’ and 2 ‘No need—not applicable’ were re-coded as 1 ‘No need’ and subsequent response categories were re-scored accordingly (1–4) to ensure a linear response format. Descriptive item statistics were examined to identify flooring or ceiling effects. Factor analysis using principal factor analysis with oblique rotation [18,19] was performed to identify underlying needs domains (no firm a priori expectations of these). The eigenvalueo1 rule, scree plot, and parallel analysis were used to determine the number of factors to retain [18,19]. Items were primarily included in the factor where their loading was the highest (minimum 0.30) [18]. A factor’s final composition was also determined by its internal consistency (minimum a 5 0.70) [20]. Missing values were managed using pairwise deletion. Construct validity was examined via the knowngroup approach [20]. Construct validity of the SCNS-P&C is supported if the scale can discriminate between groups of individuals hypothesised to experience different levels of unmet needs. Most compelling in the literature is that higher levels of unmet supportive care needs are reported by individuals with high levels of distress, including anxiety or depression [10,11]. For instance, Janda et al. [11] found that clinical anxiety and depression significantly predicted higher than average supportive care needs in caregivers of patients with a brain tumour (ORanxiety 5 2.20, ORdepression 5 5.75). Although the evidence remains equivocal, some studies also tend to indicate that several types of unmet supportive care needs are experienced more Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon Supportive care needs survey—P&C frequently among women caregivers than men caregivers [9] and among younger caregivers (less than 60 years of age) than older caregivers [9,10]. Based on this evidence, the primary hypothesis put forward to establish the construct validity of the SCNS-P&C was that a greater proportion of individuals with anxiety or depression (HADS score of eight or more) would report at least one unmet moderate or high need across domains. The secondary hypotheses tested were that a greater proportion of participants less than 60 years of age and women would report at least one unmet moderate or high need across domains. It was also hypothesised that the proportion of participants identifying at least one unmet moderate or high need for each domain will vary across cancer types. Although there is no other study documenting caregivers’ level of unmet needs across the eight most incident cancers in Australia; it can be expected that participants caring for patients with a poorer prognosis (e.g. lung cancer) will identify more unmet needs across domains in comparison with caregivers of other patients. Despite limited research in this area, the above hypotheses are further supported by findings from the patient literature. Where higher levels of distress (including anxiety and depression) [21,22], female gender [23], younger age [22,24,25] and more advanced disease at diagnosis [21,25,26] were identified as predictors of level and type of unmet needs in cancer patients. For this last analysis, the small proportion of individuals diagnosed with leukaemia was excluded (o10%). As the endorsement of unmet needs was skewed towards no unmet needs and based on the recommendations of others [10,11], it was chosen to dichotomise this variable as ‘no unmet need’ or ‘at least one unmet moderate or high unmet need’. These hypotheses were tested using w2analysis and a p-value of less than 0.05 was considered significant. Post-hoc analyses were conducted using 95% confidence intervals (CI). 389 Item distribution Item raw means ranged from 1.06 to 1.61 (Table 1). Items 18, 19, 24, and 25 showed a significant flooring effect (‘no need’ responses endorsed by more than 90% of participants) and were deleted from subsequent analyses. Factor analysis Four factors were retained and provided a clear factor solution that is conceptually and practically relevant (Table 1). With the exception of items 2 and 6, all items loaded more on the chosen factor. Factor 1 relates to receiving optimal health-care services and/or appropriate support from healthcare professionals [14,15] and is labelled ‘Health Care Service Needs’. Factor 2, labelled ‘Psychological and Emotional Needs’, assesses needs pertaining to preserving or managing emotions, thoughts, and/or relationships with the patient and others [14,15]. Despite crossloading on factor 2, items 32, 33 and 34 were included in factor 3, as their loading was greater on this factor and it is thought that conceptually these items are more fitting within this grouping of items. Factor 3 addresses needs regarding the caregiver’s or patient’s work and those pertaining to interpersonal exchanges and/or talking about cancer [14,15]. This domain was termed ‘Work and Social Needs’. Factor 4 relates to the caregiver’s information needs or understanding of the illness and is termed ‘Information Needs’. After much consideration, items 2 and 6 were retained in the Information needs domain, despite greater loading on the Health Care Service Needs, as conceptually these were developed to assess information needs. Item 15 did not load on any factor and was excluded from subsequent domain analyses. Scale reliability Results Participants Analyses were conducted with 547 participants (survey return rate 5 82.6%). Participants’ age ranged from 16 to 85 years (mean 5 60.6, SD 5 11.1) and more than two-thirds of them were women (69.8%). Most participants were born in Australia (81.0%), English-speaking (96.5%), with partner (95.8%), living with (92.4%) or caring for their spouse/partner with cancer (90.3%). The average age of the cancer survivor at diagnosis was 60.5 years (range 5 24–80, SD 5 10.5). Cancer survivors had a diagnosis of prostate cancer (32.0%), NHL (13.9%), melanoma (11.5%), breast (13.2%), colorectal (11.3%), HN (8.6%), or lung cancer (7.1%) or leukaemia (2.4%). Copyright r 2010 John Wiley & Sons, Ltd. As listed in Table 1, the reliability coefficients for all four factor-based domains were found to be appropriate and range from a 5 0.88–0.94. Construct validity of the needs domains Across all domains significantly more participants with baseline/clinical anxiety or depression identified experiencing at least one unmet moderate or high need than non-anxious or non-depressed participants (po0.01) (Table 2). Although across domains a greater proportion of the younger participants identified at least one unmet moderate or high need in comparison to older participants, these differences were only significant for the Psychological and Emotional (po0.01) and Work and Social Needs (po0.01) (Table 2). Analyses did Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon 390 Copyright r 2010 John Wiley & Sons, Ltd. Table 1. Item mean, frequency of unmet needs, and underlying factor structure of the SCNS-P&C (n 5 547) SCNS-P&C items ‘In the last month, what was your level of need for help withy’ 1.21 1.44 1.27 1.26 1.36 1.47 1.46 1.28 1.37 1.43 1.40 1.40 1.32 1.61 1.41 1.28 1.48 1.13 1.11 1.48 1.38 1.41 1.14 1.14 1.06 1.29 1.22 1.23 1.26 1.17 1.61 1.43 1.51 1.46 1.42 1.42 1.41 1.34 1.42 1.43 1.33 1.40 1.18 1.33 (0.59) (0.87) (0.68) (0.67) (0.75) (0.89) (0.99) (0.77) (0.87) (0.90) (0.92) (0.91) (0.82) (1.01) (0.81) (0.79) (0.91) (0.53) (0.46) (0.98) (0.80) (0.81) (0.84) (0.55) (0.31) (0.75) (0.64) (0.66) (0.67) (0.51) (0.99) (0.83) (0.91) (0.86) (0.83) (0.85) (0.79) (0.75) (0.83) (0.82) (0.74) (0.81) (0.58) (0.77) Factor loadings in bold indicate to which factor items were assigned. Means calculated with response format 1–4. Some need Factor 1 (a 5 0.94) Low (%) Moderate–high (%) 8.0 10.8 10.0 8.4 10.9 10.8 5.6 4.8 4.3 9.1 5.2 6.1 5.5 14.2 11.5 4.4 12.2 2.8 3.9 5.7 11.7 13.0 6.1 4.3 3.5 6.9 6.3 7.2 10.4 8.5 14.7 12.6 12.8 10.9 10.9 9.5 12.8 11.7 12.8 14.5 10.5 13.1 6.1 8.9 5.8 13.5 7.1 7.5 11.4 15.1 14.7 9.3 13.0 13.4 13.3 13.0 9.9 18.1 12.8 8.9 14.3 4.2 3.0 15.9 10.7 11.9 12.0 3.7 1.1 8.5 6.0 6.3 6.5 3.5 18.6 12.6 16.1 15.0 13.7 13.6 12.4 9.5 12.3 11.7 9.7 11.1 5.0 9.5 Factor 2(a 5 0.94) Factor 3(a 5 0.90) 0.50 0.33 0.60 0.30 0.53 0.46 0.50 0.49 0.72 0.50 0.66 0.73 0.95 0.83 0.82 0.52 0.70 0.51 — — 0.42 Factor 4(a 5 0.88) — — — — — — 0.45 0.38 — — — — 0.47 0.53 0.53 0.53 0.59 0.53 0.68 0.64 0.88 0.65 0.67 0.89 0.48 0.63 — — 0.82 0.73 0.53 0.39 0.50 0.32 — — 0.33 0.33 0.38 0.31 A. Girgis et al. Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon 1. Information carer needs 2. Information prognosis 3. Information support services 4. Information alternative therapies 5. Information patient physical needs 6. Information for decision making 7. Best medical care patient 8. Access local health services 9. Involved in patient care 10. Discuss concern with doctor 11. Doctor coordinated care 12. Case manager coordinated services 13. Complaints regarding care addressed 14. Reduce stress for patient 15. Look after own health 16. Pain control for patient 17. Fears about patient deterioration 18. Fertility problems in patient 19. Practical caring tasks 20. Accessible hospital parking 21. Changes to patient’s life/work 22. Life/work changes for carer 23. Financial/government support 24. Insurance for patient 25. Access legal services 26. Communicate with patient 27. Communicate with family 28. Support from family 29. Talk to other cancer carers 30. Discuss cancer at work/socially 31. Concerns about recurrence 32. Impact on relationship with patient 33. Understand patient experience 34. Balancing own and patient’s needs 35. Changes in patient’s body 36. Problems with sex life 37. Emotional support for self 38. Emotional support for loved ones 39. Feelings about death 40. Not acknowledging impact of caring 41. Recovery not as expected 42. Decision making in uncertainty 43. Own spiritual beliefs 44. Meaning in patient’s illness Mean (7SD) o0.01 0.23 0.08 0.04 45.7 37.1 28.6 34.3 40.4 42.1 31.6 38.6 55.2 48.3 27.6 34.5 21.3 25.5 14.9 14.9 24.9 22.5 11.4 18.0 Health-care service Psychological and emotional Work and social Information X8 (%) o8 (%) 50.2 59.4 37.7 41.1 o0.01 o0.01 o0.01 o0.01 29.6 29.8 16.3 21.6 X8 (%) o8 (%) 59.3 70.3 47.3 52.8 o0.01 o0.01 o0.01 o0.01 37.4 43.3 26.9 31.1 460 (%) p60 (%) 33.1 31.8 16.9 24.0 0.30 o0.01 o0.01 0.07 35.2 35.2 18.8 27.3 Female (%) Male (%) 34.9 37.5 22.3 27.0 0.96 0.60 0.36 0.95 37.0 41.3 17.4 21.7 33.3 31.5 14.8 22.2 23.7 29.8 15.3 20.6 Lung (%) Colorectal (%) Breast (%) Prostate (%) Melanoma (%) NHL (%) HN (%) p Cancer type p Sex p Age p Depression p Anxiety SCNS-P&C domains Table 2. Proportion of participants identifying at least one unmet moderate or high need by anxiety, depression, age, sex and cancer type (Nage/sex 5 546; Nanxiety/depression 5 540; Ncancer type 5 399) Supportive care needs survey—P&C Copyright r 2010 John Wiley & Sons, Ltd. 391 not support the hypothesis that women would experience more unmet needs than men. Finally, the proportion of participants identifying at least one unmet moderate or high need significantly differed across cancer types for Health Care Service (po0.01) and Information Needs (p 5 0.04) (Table 2). Findings were similar whether items 2 and 6 were included in the Health Care Service Needs domain instead of the Information Needs domain and whether items 32, 33, and 34 were included in Work and Social Needs instead of Psychological and Emotional Needs. Discussion The SCNS-P&C is a promising tool to measure the multi-dimensional unmet needs of cancer caregivers across the illness trajectory and to discern among groups experiencing different levels of unmet needs. Factor analysis identified four underlying needs domains—Health Care Service Needs, Psychological and Emotional Needs, Work and Social Needs, and Information Needs. These domains echo the need patterns generally reported in the cancer caregiver literature [1,8,9]. The strength of this factor structure primarily lies in that most items loaded on the factor that was consistent with conceptual expectations, with the exception of items 2 and 6. Item 6 might have been problematic, as it seemingly embeds two related needs: one related to information—‘need for treatment and information’—and the other related to health-care service—‘need to participate in decision making’. Item 6 will be revised in subsequent version of the SCNS-P&C. However, reliability and construct validity analyses support the use of the Information Needs domain in its present form. The identified rank-ordering of unmet needs in this study is similar to the one reported by Janda et al. [11], who administered the SCNS-P&C to 70 caregivers of individuals diagnosed with a brain tumour. However, a greater proportion of caregivers in the study of Janda et al. [11] reported experiencing these unmet needs. In addition, some unmet needs such as ‘decision making in uncertainty’ and ‘involved in patient care with medical team’ were rated higher than in our study. These differences might be explained by differences in illness severity and further supports the SCNS-P&C responsiveness to different care giving contexts. Furthermore, analyses supported that anxious or depressed individuals are more likely to report at least one unmet moderate or high need than nonanxious or non-depressed participants. It is foreseeable, when examining the top five unmet moderate or high needs identified in this study, that experiencing any one of these, even if just one unmet need, can be quite distressing. Until further Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon 392 studies are conducted, the authors would caution against any assumptions made about a ‘clinically significant’ level of unmet need and trivialisation of the level of unmet needs reported by the caregivers in this sample. Also, according to our findings and the analyses conducted by other studies [10,11], future psychometric studies of the SCNS-P&C will examine whether a dichotomous response format is more appropriate. Moreover, our findings are consistent with other studies [9,10] suggesting that younger caregivers are more likely to experience certain types of unmet needs, particularly in the Psychological and Emotional Needs and Work and Social Needs domains. Finally, to our knowledge, this is one of the first studies comparing caregivers’ level of unmet needs across a range of cancers with the highest incidence in Australia. The hypothesis that the proportion of caregivers experiencing unmet moderate or high needs would vary across cancer groups was supported, particularly for the Health Care Service as well as Information Needs. Although the construct validity of the scale was partially supported via the known-group approach, the cross-sectional nature of this analysis did not allow us to examine the predictive validity of the SCNS-P&C and, as no other needs assessment measure was concurrently given to participants, convergent validity of the scale was not assessed. Acknowledgements This project was undertaken by The Cancer Council New South Wales’ Centre for Health Research & Psychooncology (CHeRP), with funding from the Cancer Council NSW and infrastructure support from the Hunter Medical Research Institute. The views expressed are not necessarily those of the Cancer Council. We are grateful to Prof. Phyllis Butow and Dr Katharine Hodgkinson for their generosity and critical input into the development of the SCNS-P&C; and to Fiona Stacey for her support of the Partners and Caregivers Study, which provided the data for this article. The authors also acknowledge the contribution of Dr Harindra Jayasekara in the early stages of the development of this article. There are no known conflicts of interest for any of the authors. References 1. Clavarino AM, Lowe JB, Carmont S-A, Balanda K. The needs of cancer patients and their families from rural and remote areas of Queensland. Aust J Rural Health 2002;10:188–195. 2. Grimm PM, Zawacki KL, Mock V, Krumm S, Frink BB. Caregiver responses and needs: an ambulatory bone marrow transplant model. Cancer Prac 2000;8:120–128. 3. Northouse LL, Mood D, Templin T, Mellon S, George T. Couples’ patterns of adjustment to colon cancer. Soc Sci Med 2000;50:271–284. 4. Wagner CD, Bigatti SM, Storniolo AM. Quality of life of husbands of women with breast cancer. PsychoOncology 2006;15:109–120. Copyright r 2010 John Wiley & Sons, Ltd. A. Girgis et al. 5. Schmid-Buchi S, Halfens RJG, Dassen T, den Borne B. A review of psychosocial needs of breast cancer patients and their relatives. J Clin Nurs 2008;17: 2895–2909. 6. Thomas C, Morris S, Soothill K, McIllmurray M, Francis B, Harman J. What are the Psychosocial Needs of Cancer Patients and Their Main Carers? The Institute for Health Research, Lancaster University: Lancaster, UK. Available from: http://www.lancs.ac.uk/ fass/ihr/publication/carolthomas/the%20psychosocial% 20needs%20report.pdf. Accessed 10 March 2009. 7. Dubenske LL, Wen K-Y, David GH et al. Caregivers’ differing needs across key experiences of the advanced cancer disease trajectory. Palliat Support Care 2008;6: 265–272. 8. Petrie W, Logan J, DeGrasse C. Research review of the supportive care needs of spouses of women with breast cancer. Oncol Nurs Forum 2001;28:1601–1607. 9. Osse BHP, Vernooij-Dassen MJFJ, Schade E, Grol RPTM. Problems experienced by the informal caregivers of cancer patients and their needs for support. Cancer Nurs 2006;29:378–388. 10. Hodgkinson K, Butow P, Hobbs KM, Hunt GE, Lo SK, Wain G. Assessing unmet supportive care needs in partners of cancer survivors: the development and evaluation of the Cancer survivors’ partners unmet needs measure (CaSPUN). Psycho-Oncology 2007;16: 805–813. 11. Janda M, Steginga S, Dunn J, Langbecker D, Walker D, Eakin E. Unmet supportive care needs and interest in services among patients with a brain tumour and their carers. Patient Educ Couns 2008;71:251–258. 12. Soothill K, Morris SM, Harman JC, Francis B, Thomas C, McIllmurray MB. Informal carers of cancer patients: what are their unmet psychosocial needs? Health Soc Care Community 2001;9:464–475. 13. Campbell HS, Sanson-Fisher R, Taylor-Brown J, Hayward L, Wang XS, Turner D. The cancer support person’s unmet needs survey. Cancer 2009;115: 3351–3359. 14. Hileman JW, Lackey NR, Hassanein RS. Identifying the needs of home caregivers of patients with cancer. Oncol Nurs Forum 1992;19:771–777. 15. Bonevski B, Sanson-Fisher R, Girgis A, Burton L, Cook P, Boyes A. Evaluation of an instrument to assess the needs of patients with cancer. Cancer 2000;88:218–225. 16. McElduff P, Boyes A, Zucca A, Girgis A. Supportive Care Needs Survey: A Guide to Administration, Scoring and Analysis. Centre for Health Research & PsychoOncology (CHeRP), University of Newcastle: Newcastle, 2004. 17. Boyes A, Zucca A, Girgis A, Lecathelinais C. Undertaking population-based, longitudinal survivorship research: practicalities, challenges and solutions (abstract). Asia-Pacific J Clin Oncol 2006;2(S2):A93. 18. Costello AB, Osborne JW. Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Pract Assess Res Eval 2005;10(7):1–9. Available from: http://pareonline.net/getvn.asp?v 5 10&n 5 7. 19. DeVellis RF. Scale Development: Theory and Applications. Sage Publications: London, 2003. 20. Nunnally JC, Bernstein IH. Psychometric Theory (3rd edn). McGraw-Hill: New York, NY, 1999. 21. Hodgkinson K, Butow P, Fuchs A et al. Long-term survival from gynaecologic cancer: Psychosocial outcomes, supportive care needs and positive outcomes. Gynecol Oncol 2007;104:381–389. Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon Supportive care needs survey—P&C 22. Armes J, Crowe M, Colbourne L et al. Patients’ supportive care needs beyond the end of cancer treatment: a prospective, longitudinal survey. J Clin Oncol 2009;27:6172–6179. 23. Sanson-Fisher R, Girgis A, Boyes A, Bonevski B, Burton L, Cook P. The unmet supportive care needs of patients with cancer. Cancer 2000;88:226–237. 24. Soothill K, Morris SM, Harman J, Francis B, Thomas C, McIllmurray MB. The significant unmet needs of Copyright r 2010 John Wiley & Sons, Ltd. 393 cancer patients: probing psychosocial concerns. Support Care Cancer 2001;9:597–605. 25. Barg FK, Cronholm PF, Straton JB et al. Unmet psychosocial needs of Pennsylvanians with cancer: 1986–2005. Cancer 2007;110:631–639. 26. Smith DP, Supramaniam R, King MT, Ward J, Berry M, Armstrong BK. Age, health, and education determine supportive care needs of men younger than 70 years with prostate cancer. J Clin Oncol 2007;25:2560–2566. Psycho-Oncology 20: 387–393 (2011) DOI: 10.1002/pon