Caregiver satisfaction with out-patient oncology services: utility of the FAMCARE instrument and development of the FAMCARE-6

Transcription

1 DOI /s ORIGINAL ARTICLE Caregiver satisfaction with out-patient oncology services: utility of the FAMCARE instrument and development of the FAMCARE-6 Gregory Leigh Carter & Terry J. Lewin & Louisa Gianacas & Kerrie Clover & Catherine Adams Received: 18 August 2009 / Accepted: 25 February 2010 # Springer-Verlag 2010 Abstract Purpose To evaluate caregivers experience of oncology services for ambulatory patients and to develop a short instrument (FAMCARE-6) suitable for computerised administration in the clinical setting. Methods A sample of 234 caregivers recruited from 388 ambulatory oncology patients completed a computerised version of the 20-item family satisfaction with advanced cancer care (FAMCARE) instrument, which was originally developed for use in palliative care settings. Results Caregivers reported generally high satisfaction with all aspects of ambulatory oncology services: overall score; mean, 3.96 (SD, 0.67); information giving, 3.88 (0.78); physical patient care, 4.00 (0.71); availability of care, 3.89 (0.77); and psychosocial care, 4.05 (0.72), from a possible score of 5. Factor analyses identified a single factor structure; the items were reduced to six (FAMCARE-6), which yielded a scale with adequate psychometric properties (completion G. L. Carter : T. J. Lewin : L. Gianacas : K. Clover : C. Adams Centre for Brain and Mental Health Research (CBMHR), University of Newcastle, Newcastle, NSW, Australia T. J. Lewin Hunter New England Mental Health, Newcastle, NSW, Australia G. L. Carter : L. Gianacas : K. Clover : C. Adams Psycho-Oncology Service, Calvary Mater Newcastle, Newcastle, NSW, Australia G. L. Carter (*) Department of Consultation-Liaison Psychiatry, Locked Bag #7 Hunter Region Mail Centre NSW, Newcastle 2310, Australia gregory.carter@newcastle.edu.au rates over 90% for every item, correlation of 0.7 or above with the factor identified in the individual item factor analysis, and internal reliability of α=0.85). The overall mean score was 3.91 (SD, 0.73) for the FAMCARE-6. Conclusions FAMCARE-6 can be used to assess caregiver satisfaction with ambulatory oncology services and may be suitable to be included as part of a computerised screening system for the psychological care of oncology patients. Keywords Psycho-oncology. Caregivers. Service evaluation. FAMCARE Background Caregivers provide an important component of non-medical and medical care for oncology patients. In a survey of terminally ill patients in the USA, family members, usually women, provided the majority of assistance with nonmedical care, 15% had assistance from paid caregivers and only 3% had assistance from volunteers [1]. Moreover, caregivers are increasingly replacing skilled health workers in the delivery of unfamiliar, complex cancer care at home [2]. Caregivers contributions are not restricted to palliative care or terminal care phases of the illness. Qualitative studies have suggested that the involvement of the caregiver with medical systems may change over the course of the cancer journey. In the diagnosis phase, carers were more unsure of their role and required active inclusion by patients and health professionals, whilst increasing experience of the medical setting led to greater confidence in their role as they found ways of being involved and sharing aspects of the experience with the patient [3]. Assessment of family-based satisfaction with the care of their loved one is not only useful in the exercise of quality

2 assurance and identification of unmet needs, but can also help flag any aspects of services that influence both patient and familial adjustment and distress [4]. Conversely, it is known that insufficient medical care can reflect increased burden and stress within the family unit [5]. A review of research on caregivers in cancer contexts has highlighted the existence of two contrasting approaches addressing different aspects of the caregiver experience. Psycho-oncology has dominated the field, primarily using a quantitative approach where caregivers become legitimate cousers of cancer services, whilst research drawing on sociological perspectives sees caregivers as co-carers of cancer patients in the overall care package [6]. Using either conceptual approach, it seems worthwhile to consider the question of how caregivers rate the clinical care given by treatment services at different phases of treatment of the illness. Although there have been a number of validated instruments developed to measure quality of life in caregivers for oncology patients at different phases of the illness [7], there are fewer available instruments for measuring caregiver satisfaction with services for oncology patients, and these tend to be restricted to the terminal phase of illness. A study of the perceived quality of palliative care by 181 family members of cancer patients with terminal disease explored the underlying factor structure and dimensionality of the 20 items of the FAMCARE Scale. The results from a factor analysis, a Mokken Scaling Program analysis and a reliability analysis, showed that 19 out of the 20 items formed a strong one-dimensional scale. The authors suggested that since the scale is one-dimensional, the possibility of reducing the number of items should be explored in future research [5]. In the clinical setting, we have developed a two-stage computerised screening programme for pain, psychological distress and depression or anxiety in ambulatory oncology patients, and we sought to extend this computerised screening to the caregivers to include a measure of satisfaction with delivery of clinical care for the patient [8-10]. In order to do this, we required a validated instrument that was quick to use and therefore short in terms of number of items, available in the public domain and practical to use in the ambulatory care setting of a busy oncology out-patient clinic. Although we could not find such an instrument, the FAMCARE instrument, developed in the palliative care setting [11, 12], looked suitable for potential modification for this clinical setting. Aims 1. Measure the level of service satisfaction by caregivers of oncology out-patients using the FAMCARE instrument 2. Assess the internal reliability and factor structure of the FAMCARE instrument in this population 3. Develop a shorter version of the FAMCARE instrument with comparable utility to the full instrument 4. Identify the relationships between service satisfaction and caregiver demographic factors and quality of life. Methods Participants Full details of the recruitment of caregivers via oncology patient attendances at oncology out-patients have been previously described [8]. Participants were recruited from all clinic attendees at an ambulatory out-patient clinic in Oncology Haematology, based at the Calvary Mater Newcastle Hospital, Australia. Recruitment occurred over a 5- week period using a two-step procedure approved by the relevant Ethics Committee. The initial approach was made to the oncology patients by reception staff at the registration desk with subsequent contact from the research staff. Any accompanying caregiver was also initially approached by the reception staff and if interested was formally invited, recruited and consented by the research staff. A sample of 234 caregivers recruited from 388 oncology patients became the participants for this study and undertook the computer-based questionnaire. Four carer participants did not complete the FAMCARE section of the research assessment, leaving a sample of 230 participants included in this study. Instruments and procedures All participants completed data collection using a computerised version of the instruments. A bank of five personal computers was placed in a partially screened area in the waiting room of the usual clinic area. All questions were presented one per screen, and any missed questions were summarised after the last question was answered; the participants were then returned to answer each missing question and offered the option of completing them. Assistance from the research staff was available if required. Participants were not required to have had any previous computer operation experience. Demographic information was collected for the independent variables: age, gender, employment status, marital status and area of residence based on local government area. The two instruments used were the family satisfaction with advanced cancer care (FAMCARE) [12] and the World Health Organisation Quality of Life (BREF version) (WHOQOL-BREF) [13].

3 TheFAMCAREscaleisa20-item self-report instrument designed to measure the extent to which a family member (caregiver) is satisfied with the medical care provided to the patient for which they care. This instrument was specifically developed for use in advanced cancer care settings, principally in home-based palliative care services. Family members are asked to rate their satisfaction across the four subscales of physical patient care, information giving, availability of care and psychosocial care [12]. Each of the 20 items can be rated from 1 =low satisfaction to 5=high satisfaction. The FAMCARE scale has reported high internal consistency (α=0.93), test retest reliability (r=0.92) and mean inter-item correlation for the total scale of 0.40 (SD=0.18) [12]. Later studies have substantiated the scale s psychometric rigour when used in a combination of in- and out-patient palliative settings [14, 15]. The WHOQOL-BREF is a standardised 26-item instrument measuring quality of life over the previous 2 weeks [13]. The WHOQOL-BREF was designed to be primarily used as a brief assessment tool in routine clinical work, large scale epidemiological studies and in clinical trials. It has been shown to be a valid and reliable alternative to the assessment of four domain profiles using the longer WHOQOL-100 [13]. Six variables were derived from the WHOQOL-BREF, based on the two preliminary questions and the four domain scores (physical, psychological, social and environment). These two preliminary questions were: 1. How would you rate your quality of life? (overall quality of life); and 2. How satisfied are you with your health? (overall health satisfaction). For each of the four domains, all items were rated on a five-point scale with response sets assessing either intensity, frequency or satisfaction. Scores on negative items were reversed, and domain scores were summed and transformed into percentile scores (range, 0 100) to aid comparison with other studies. Higher scores indicate better quality of life or functioning. Analyses Descriptive statistics were used to characterise the sample, while traditional psychometric techniques (internal consistency coefficients or Cronbach s alpha and principal components factor analyses) were used to examine the performance of the FAMCARE scale, subscales and individual items. Relationships among the FAMCARE measures with quality of life measures were assessed using Pearson s product-moment correlations, while oneway analysis of variance (ANOVA) was used to examine the relationship between caregiver demographics, quality of life, satisfaction with health and mean FAMCARE scores. Data analyses were conducted using SPSS statistical software (version 14.0 SPSS, Chicago, Ill, USA). Ethics clearance The Hunter Area Research Ethics Committee approved the Psycho-Oncology Caregivers Study (02/12/11/3.26), which is reported in this paper. Results Demographic characteristics of the caregivers The demographic characteristics of the caregivers can be seen in Table 1. The caregivers were predominately female (69%), married or de facto (89%), receiving a pension or other government benefits (56%) and lived in geographical areas of close proximity to the hospital clinics (85%). The mean age of the caregivers was 58.4 years (SD=13.35). Table 1 Demographic characteristics of caregivers Characteristic n=230 % Gender Male Female Marital status Not currently married Married or de facto Age Under and over Employment status Full time or part time employed Pension or other benefits Other Home location (distance from CMN) a Newcastle or Lake Macquarie Port Stephens or Lower Hunter Other areas a CMN Calvary Mater Newcastle Residents of Newcastle or Lake Macquarie 0 30 km distant Residents of Port Stephens or Lower Hunter km distant Residents of other areas 80+km distant

4 Caregiver satisfaction with oncology services measured by the FAMCARE The pattern of individual item scores and non-response rates can be seen in Table 2. Six items, (1, 6, 7, 12, 13 & 16) had substantially lower response rates than the other items (>20% non-response rate). The FAMCARE has four subscales, which showed the following scores from caregivers, for a maximum possible score of five: physical patient care, mean, 4.00 (SD, 0.71); information giving, mean, 3.88 (SD, 0.78); availability of care, mean, 3.89 (SD, 0.77); and psychosocial care, mean, 4.05 (SD, 0.72); together with an overall score, mean, 3.96 (SD, 0.67). As evidence of the uniformity of responses, all individual items, the four subscales and the overall score, had a median of four. Psychometric properties of the FAMCARE The correlation coefficients among the four FAMCARE subscales can be seen in Table 3, which ranged from 0.63 to The internal consistency (or reliability) of the scale as measured by the Cronbach alpha was high (α=0.95) for the 92 carers who completed all 20 items on the scale. For 222 carers who completed 10 or more items on the scale, the Cronbach alpha was similar (α=0.96). An exploratory factor analysis of the four subscales in the FAMCARE instrument showed only one component with an eigenvalue greater than one (eigenvalue, 3.21), which accounted for 80% of the variance. This suggested that in this particular sample, the FAMCARE instrument is probably measuring a unitary construct. A further exploratory factor analysis using the 20 individual items in the FAMCARE instrument (using pairwise correlations, n=151 to 227, with an oblique rotation and inspection of the scree plot), yielded two possible components with eigenvalues greater than one, with the first factor accounting for 52% of the variance and the second factor accounting for 7% of the variance, attributable to a small number of items (e.g., #1 pain relief). However, as shown in Table 2, all of these items also had substantial loadings on the first principal component. This again suggested one primary underlying dimension or construct of satisfaction in the FAMCARE scale. Developing the FAMCARE-6 These results raised an important question. If measurement of caregiver satisfaction is unitary and individual items are highly representative of one another, can we reduce the number of items to create a shorter questionnaire that Table 2 FAMCARE individual item scores and factor loadings for caregivers Item (range=1 to 5) 1=low satisfaction, 5=high satisfaction a Mean SD Non-response % (n) Factor loading 1. Pain Relief (46) Information about prognosis (8) Answers from health professionals (4) Information about side effects (7) Referrals to specialists (3) Availability of hospital bed (56) Family conferences (79) Speed symptoms are treated (17) Doctor s attention to symptoms (9) Tests and treatments (9) Availability of doctors to family (20) Availability of nurses to family (79) Coordination of care (59) Time to make diagnosis (15) Family inclusion in decisions (36) Pain Management information (63) Test information (11) Assessment of patient s symptoms (14) Follow up by doctor (13) Availability of doctor to patient (11) 0.74 a Item wording has been abbreviated for inclusion in the table

5 Table 3 Correlations among FAMCARE subscales All correlations are significant at the p<0.01 level (two-tailed) Total score Information giving Physical care Availability of care Information giving 0.90 Physical care Availability of care Psychosocial care adequately taps the construct of satisfaction and covers the subscales of the FAMCARE scale? The questionnaire was shortened by including items only if: they correlated strongly with the one factor solution presented in Table 2 (r>0.70); they had a high response rate (>90%) implying applicability to the ambulatory setting, and they provided a reasonable cross representation of factors identified in previous studies by Kristjanson [12] and Ringdal et al. [5]. Specifically, these were Kristjanson s conceptual structure, information giving (items 3 and 4), availability of care (items 11 and 20), physical care (items 8 and 14) and psychosocial care (none); Kristjanson s empirical structure, information giving (items 3 and 4), availability of care (items 11 and 20), physical care (items 8 and 14) and pain control (none); and Ringdal s empirical three factor structure, general care (items 11 and 20), information (items 8 and 14) and physical care (items 3and4). We chose six items, which fulfilled the required criteria: 3, 4, 8, 11, 14 and 20. We called this scale the FAMCARE- 6, which showed sound internal consistency (or reliability) as measured by the Cronbach alpha (α=0.85). For the 212 caregivers who had a maximum of one non-response item on the FAMCARE-6, the overall mean score was 3.91 (SD 0.73). Relationship between caregiver demographics, quality of life, satisfaction with health and FAMCARE scores There were no significant associations between demographic variables and FAMCARE overall scores: gender F (1,220)=3.10, p=0.08, marital status F (1,220)= 3.00, p=0.08, age F (1,219)=1.42, p=0.25, employment status F (2,219)=0.33, p=0.72 and geographical location F (2,219)=0.19, p=0.83. A similar pattern of nonsignificant associations was found for the demographic variables using FAMCARE-6. The correlation matrix of caregiver satisfaction as determined by FAMCARE and FAMCARE-6 with WHOQOL-BREF can be seen in Table 4. Although there are a number of statistically significant correlations, these are mostly weak relationships (r<0.35). The results for FAMCARE overall and FAMCARE-6 (not reported in detail) were very similar. Discussion Strengths and limitations of the study design The study had a primary aim to examine service satisfaction in caregivers of ambulatory oncology patients. The sample used a broad sampling frame, drawn from a regional service for oncology and haematology patients, and all patients and carers were eligible so there was no filtering of subjects (e. g., by clinicians deciding who would be suitable). A standardised instrument with well-established psychometric properties and potential applicability to caregivers in the ambulatory care setting was used. Lastly, the sample size was moderately large, data collection was prospective, and caregiver participants included both genders. We used a cross-sectional design with a convenience sample of ambulatory patients attending a haematology or oncology clinic, which was unselected for illness type or stage. Caregivers were recruited via patients, needed to be attending with the patient and were not paired in design or analysis to the patients. The characteristics of the patients who declined to participate in the study and for carers who therefore could not be approached or who did not attend the clinic are not known. The magnitude and direction of any such ascertainment biases cannot be determined. Hence, the generalisability of these results to other oncology patient and carer groups is unknown. Females as caregivers The majority of caregivers in this study were female and care giving by women is common. A survey of 742 women in midlife in New Zealand found that 12% were involved in the care of another person with a chronic health disorder, and about a third of the carers were supporting more than one person [16]. Wives are only one third as likely as husbands to select their spouses as caregivers [17]. In a US study of terminally ill populations, 72.1% of the care givers were women, and

6 Table 4 Correlation between FAMCARE measures and WHOQOL-BREF measures FAMCARE measures Information giving Physical Availability of care Psychosocial care Overall FAMCARE-6 WHOQOL-BREF measures Quality of life 0.20** 0.26** 0.19** 0.21** 0.26** 0.20** Satisfaction with health 0.20** 0.24** 0.18* ** 0.23** Physical domain * 0.15* 0.15* 0.17* 0.10 Psychological domain 0.26** 0.24** 0.28** 0.23** 0.27** 0.26** Social domain 0.23** 0.14* 0.23** 0.23** 0.21** 0.19** Environment domain 0.23* 0.29** 0.35** 0.31** 0.32** 0.28** **p<0.01 level, two-tailed *p<0.05 level, two-tailed primary care givers were family members in 96.0% of cases [1]. Comparison to reports of service satisfaction in other studies using FAMCARE Despite the variation in the provision of palliative care services, satisfaction amongst family members tends to be consistently high across studies, even when caregivers report higher levels of psychological distress and elevated illness burden [18]. A cluster randomised trial showed nearuniversal approval of both of the two types of intervention in a palliative care setting [19], whilst a longitudinal study showed very high approval of care 1 month after the death of the patient [5]. In a Canadian longitudinal study of advanced breast cancer patients and their caregivers, the caregivers showed a substantial satisfaction with care, at baseline mean 79.3 (SD, 14.9) and at the end of palliative period of treatment 83.4 (SD, 11.2) for the full scale FAMCARE [18]. Development of FAMCARE and FAMCARE-6 The original FAMCARE Scale was developed in two phases; the first aimed to identify the specific factors that families perceive as helpful or unhelpful in the care of a terminally ill family member, while the second utilised Q- Sort methodology to rank these factors according to the level of importance assigned by the participant group. The combined outcome of these studies was a greater understanding of the features of the palliative care setting salient to family satisfaction, thus providing the basis for individual scale items [11, 20]. Later psychometric evaluation of the 20-item scale determined a high level of internal reliability (α=0.93), test retest reliability (r=0.92) and inter-item correlation amongst the four subscales of physical patient care, information giving, availability of care and psychosocial care (over 50% r= ). Criterion validity of the instrument was substantiated through subscale comparison with the McClusker Scale (r= ). Despite study limitations imposed by a small sample size (n=30), it was concluded that the FAMCARE scale had the potential to be a valid and reliable measurement of family satisfaction in advanced cancer settings [12]. In the current study, items 1, 6, 7, 12, 13 and 16 had substantially lower response rates (or higher number of non-responses), and this is probably because these items are not commonly applicable to patient issues found in ambulatory settings. This reflects the idea that the FAM- CARE instrument was originally designed to capture service satisfaction for patients in Palliative care settings. We omitted these high non-response items, the single pain item and included items representative of the sub scales. Omission of pain item In a study of 86 family caregivers of patients with cancerrelated pain and 42 caregivers of pain-free patients, the presence of pain in a cancer patient adversely affected the mood states of family caregivers, particularly their level of depression and anxiety. They concluded that health care professionals needed to assess the psychological needs of family members caring for cancer patients experiencing pain [12]. We decided to omit the pain relief item (#1), since it had a low response rate, and it did not load on the unitary factor, although we recognise that pain is an important component of ambulatory oncology care. In our clinical computerised screening programme we measure the patients experience of pain on a visual analogue scale in order to alert the clinician caring for the patient to the presence of pain [10].

7 FAMCARE as a one-dimensional scale Since its development, the FAMCARE scale has been utilised in a number of studies investigating the care perceptions of family members in both in- and outpatient settings. Psychometrically, it has shown good internal consistency (α= ) and inter-item correlation [12, 14, 15]. In perhaps the largest sample size analysis of the psychometric properties of FAMCARE since its development, Ringdal et al. [5] reported strong evidence of the scale s likely uni-dimensionality. This factor analysis of 181 family members identified three subscales referred to as general care, information and physical care as opposed to Kristjanson s original four subscales [12]. The eigenvalue for the general care factor was much higher (10.95) than for information (1.52) or physical care (1.22). This suggested that the scale s measurement is based on a single underlying construct of care satisfaction, providing justification for the reduction of scale items. Ringdal et al. [5] also identified the likely social desirability response bias associated with self-report satisfaction questionnaires, evidenced by the high average score total score (mean=62.53) from a possible scaled score of 100. In the current study, for individual items, for each of four subscales and for the overall FAMCARE score, median scores were 4.0, and mean scores were generally high indicating high caregiver satisfaction, with very little variation in scores. We would suggest that there are four possible reasons for these uniformly high scores in this study of caregivers: 1. Almost all caregivers were genuinely satisfied 2. Only satisfied caregivers consented to completing the scale 3. Only satisfied caregivers accompanied the patient to appointments and were therefore available in the waiting room as study participants to complete the scale 4. The scale has ceiling effects and little sensitivity to detect dissatisfaction in this sample. Relationship of FAMCARE measures with quality of life Although there were numerous significant associations for FAMCARE measures and quality of life measures, the magnitude of these correlations was only modest. In a previous related study, we have shown that these caregivers had lower quality of life on three of the WHOQOL-BREF domains (physical, psychological and environmental) than community subjects and argued for inclusion of quality of life measures in any screening system for caregivers [8]. Since caregiver satisfaction with services and caregiver quality of life are both important and only modestly correlated, we would now recommend inclusion of the FAMCARE-6 instrument in addition to quality of life measures for caregivers in any computerised screening of caregivers. Conclusion The New South Wales Cancer Patient Satisfaction Survey 2007 measured patient ratings, for February 2007, using two extensive questionnaire instruments to determine service delivery strengths and weaknesses [21]. Overall, the results showed a high level of satisfaction by cancer inpatients (over 90%) and cancer out-patients (over 97%), rating services as excellent, very good or good. In order to measure ongoing performance, the Cancer Institute New South Wales plans to repeat the survey annually. This is a valuable, albeit an expensive and time intensive way to measure service satisfaction at the Area Health Service and New South Wales state level. The current study has complementary aims, to measure service satisfaction as assessed by caregivers in a brief, repeatable fashion that can be reported to clinicians on an individual basis for immediate action when required. Although, caregiver evaluation of service provision is probably important there are no suitable instruments available for use. This study has demonstrated that FAMCARE-6 would yield comparable results to the longer parent instrument in an ambulatory oncology clinical setting. Given the high rates of satisfaction expressed, it would be clinically relevant to pay attention to any level of dissatisfaction: a single item on the FAMCARE-6 that was endorsed at a score of two or lower or any total score on the FAMCARE-6of18orless(mean,<3.0).Thefinalversionof the FAMCARE-6 can be seen in Appendix 1. Scoring follows the original FAMCARE instrument with each of the six items rated from 1=very dissatisfied (low satisfaction) to 5= very satisfied (high satisfaction), for a total score of 30. FAMCARE-6 can be used to assess caregiver satisfaction with ambulatory oncology services and may be suitable to be included as part of a computerised screening system for the psychological care of oncology patients. Because it is a short scale, it may be practical to use on repeated administrations covering different periods in the cancer journey. Future studies could examine test retest reliability and whether FAMCARE-6 scores of service satisfaction were predictive of patients illness related help-seeking behaviours e.g., compliance with treatment or attendance at out-patient consultations. We plan to incorporate FAMCARE-6 into the QUICA-TOUCH computerised screening system for clinical implementation in the near future [9, 10].

8 Appendix 1 FAMCARE-6 Instructions Think about the care that your family member has received. Please answer the following questions below indicating how satisfied you are with the care received: very satisfied (VS), satisfied (S), undecided (U), dissatisfied (D) or very dissatisfied (VD). Please circle the letters below that best match your experience. How satisfied are you with 1. Answers from health professionals VS S U D VD 2. Information given about side effects VS S U D VD 3. Speed with which symptoms are VS S U D VD treated 4. Availability of doctors to the family VS S U D VD 5. Time required to make a diagnosis VS S U D VD 6. Availability of the doctor to the patient VS S U D VD Scale modified from the original FAMCARE Scale (Kristjanson, 1993) [12]. References 1. Emanuel EJ, Fairclough DL, Slutsman J, Alpert H, Baldwin D, Emanuel LL (1999) Assistance from family members, friends, paid care givers, and volunteers in the care of terminally ill patients. N Engl J Med 341: Girgis A, Johnson C, Aoun S, Currow D (2006) Challenges experienced by informal caregivers in cancer. Cancer Forum 30: Morris SM, Thomas C (2001) The carer s place in the cancer situation: where does the carer stand in the medical setting? Eur J Cancer Care (Engl) 10: Kristjanson LJ, Aoun S (2004) Palliative care for families: remembering the hidden patients. Can J Psychiatry 49: Ringdal GI, Jordhoy MS, Kaasa S (2003) Measuring quality of palliative care: psychometric properties of the FAMCARE Scale. Qual Life Res 12: Thomas C, Morris SM (2002) Informal carers in cancer contexts. Eur J Cancer Care (Engl) 11: Edwards B, Ung L (2002) Quality of life instruments for caregivers of patients with cancer: a review of their psychometric properties. Cancer Nurs 25: Carter G, Lewin T, Rashid G, Adams C, Clover K (2008) Computerised assessment of quality of life in oncology patients and carers. Psychooncology 17: Clover K, Carter GL, Mackinnon A, Adams C (2009) Is my patient suffering clinically significant emotional distress? Demonstration of a probabilities approach to evaluating algorithms for screening for distress. Support Care Cancer -online publication March 2009: doi: /s Clover K, Rogers K, Carter G, Adams C (2008) QUICA-TOUCH: the first 12 months of screening for distress, pain and psychopathology. Asia Pacific J Clin Oncol 4(Suppl 2):A Kristjanson LJ (1989) Quality of terminal care: salient indicators identified by families. J Palliat Care 5: Kristjanson LJ (1993) Validity and reliability testing of the FAMCARE scale measuring family satisfaction with advanced cancer care. Soc Sci Med 36: The WHOQOL Group (1998) Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychol Med 28: Kristjanson LJ, Leis A, Koop PM, Carriere KC, Mueller B (1997) Family members care expectations, care perceptions, and satisfaction with advanced cancer care: results of a multi-site pilot study. J Palliat Care 13: Medigovich K, Porock D, Kristjanson LJ, Smith M (1999) Predictors of family satisfaction with an Australian palliative home care service: a test of discrepancy theory. J Palliat Care 15: Knight RG, Williams S, McGee R, Olaman S (1998) Caregiving and well-being in a sample of women in midlife. Aust N Z J Public Health 22: Allen SM, Goldscheider F, Ciambrone DA (1999) Gender roles, marital intimacy, and nomination of spouse as primary caregiver. Gerontologist 39: Grunfeld E, Coyle D, Whelan T, Clinch J, Reyno L, Earle CC, Willan A, Viola R, Coristine M, Janz T, Glossop R (2004) Family caregiver burden: results of a longitudinal study of breast cancer patients and their principal caregivers. Can Med Assoc J 170: Ringdal GI, Jordhoy MS, Kaasa S (2002) Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 24: Kristjanson LJ (1986) Indicators of quality of palliative care from a family perspective. J Palliat Care 1: Heading G, Mallock N, Sinclair S, Bishop J (2008) New South Wales Cancer Patient Satisfaction Survey 2007, Interim Report. Cancer Institute NSW catalogue number: CR