Psychosocial Consequences of Cancer Cachexia: The Development of an Item Bank

Transcription

1 Vol. 46 No. 6 December 2013 Journal of Pain and Symptom Management 795 Original Article Psychosocial Consequences of Cancer Cachexia: The Development of an Item Bank Hanspeter H ane, MD, Rolf Oberholzer, MD, Jochen Walker, Jane B. Hopkinson, PhD, RN, Susanne de Wolf-Linder, and Florian Strasser, MD, ABHPM Oncological Palliative Medicine (H.H., R.O., J.W., S.d.W.-L., F.S.), Section Oncology, Department of Internal Medicine and Palliative Care Centre, Cantonal Hospital, St. Gallen, Switzerland; and School of Nursing and Midwifery Studies (J.B.H.), Cardiff University, Cardiff, United Kingdom Abstract Context. Cancer cachexia syndrome (CCS) is often accompanied by psychosocial consequences (PSC). To alleviate PSC, a systematic assessment method is required. Currently, few assessment tools are available (e.g., Functional Assessment of Anorexia/Cachexia Therapy). There is no systematic assessment tool that captures the PSC of CCS. Objectives. To develop a pilot item bank to assess the PSC of CCS. Methods. A total of 132 questions, generated from patient answers in a previous study, were reduced to 121 items by content analysis and evaluation by multidisciplinary experts (doctor, nutritionists, and nurses). In our two-step, crosssectional study, patients, judged by staff to have PSC of CCS, were included, and the questions were randomly allocated to the patients. Questions were evaluated for understandability and triggering emotions, and patients were asked to provide a response using a four-point Likert scale. Subsequently, problematic questions were revised, reformulated, and retested. Results. A total of 20 patients with a variety of tumor types participated. Of the 121 questions, 31 had to be reformulated after Step 1 and were retested in Step 2, after which seven were again evaluated as not being perfectly comprehensible. In Step 1, 22 questions were found to trigger emotions, but no item required remodeling. Item performance using the Likert scale revealed no consistent floor or ceiling effects. Our final pilot question bank comprised 117 questions. Conclusion. The final item bank contains questions that are understood and accepted by the patients. This item bank now needs to be developed into a measurement tool that groups items into domains and can be used in future research studies. J Pain Symptom Manage 2013;46:795e806. Ó 2013 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Address correspondence to: Florian Strasser, MD, ABHPM, Oncological Palliative Medicine, Division of Oncology, Department Internal Medicine and Palliative Care Centre, Cantonal Hospital, Ó 2013 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Rorschacherstrasse 95, 9007 St. Gallen, Switzerland. Florian.Strasser@kssg.ch Accepted for publication: January 2, /$ - see front matter

2 796 H ane et al. Vol. 46 No. 6 December 2013 Key Words Cancer cachexia syndrome, psychosocial consequences, item bank, assessment tool Introduction Cancer cachexia is defined as a frequent, distressing, and debilitating multifactorial syndrome, 1 which affects most patients with advanced cancer. 2 It is accompanied by symptoms such as loss of muscle, loss of fat mass, 1,3 inability to eat, early satiety, taste changes, and food aversion. 4 Cancer cachexia syndrome (CCS) cannot be reversed by conventional nutritional support and leads to progressive functional impairment. 1 Malnutrition is often a component of CCS, affects 20% of patients with cancer, and increases morbidity and mortality. 5 The prevalence of weight loss amounts to 39e82%, and 30e80% of the patients are affected by loss of appetite. 6,7 In addition, patients (and their relatives) are concerned about poor food intake, weight loss, and change in eating habits, 8,9 and they also are troubled by associated emotions, 10 social challenges, 11 and existential questions. 7 These emotional aspects of CCS are captured by the terms eating-related distress 12 and weight losserelated distress 13 and have been conceptualized as psychosocial consequences (PSC) of CCS. 14,15 In 2011, Oberholzer et al. 16 reviewed a total of 15 studies about the PSC of CCS. The data were categorized into causes, presentations, and mechanisms leading to PSC and responses increasing or minimizing/alleviating PSC. This systematic literature review demonstrated, together with all the mentioned studies, how important the PSC of CCS are and that these consequences need to be more widely recognized and addressed by health care providers. Also, the work of Reid et al. 17 shows that weight loss in patients with cancer should be acknowledged more readily by the health care professionals; both information about it and interventions to deal with it should be made available. Furthermore, loss of appetite can be a source of anxiety and conflict within the family 11 because patients do not contemplate anorexia as much as their relatives do. 18 Although patients are concerned about their own weight loss and/or their loss of appetite, 19 they are more concerned about other problems they are experiencing, such as pain. They would prefer that their families spend less energy on encouraging them to eat. 11,12 Moreover, patients feel more pressure to eat than their partners estimate. 20 There is a fine line between offering food to a patient and forcing him/her to eat. 21 Throughout the course of the illness, eating can change into an uncertain, unpredictable, and distressing experience. 20 With a consensus of expert opinion that the development of a tool to assess PCS of CCS is needed, a tentative classification of an item bank into domains to assess PSC in CCS was undertaken by Strasser et al. 20 This classification assumed: 1. Patient distress to be caused by missing the capacity to eat and/or the loss of weight, which can cause existential stress when not eating is equated with dying; 2. Partner distress to arise from changes in cooking and eating habits; and 3. Couple distress to result from the sudden predominance of food and eating in the relationship. 20 Similarly, Reid et al. 9 categorized the experience of CCS into six themes (psychological changes in appetite, visuality of cachexia, weight loss interpreted as a bad sign, conflict over food, response from health care professionals, and coping responses) to reflect the complex dynamics of CCS. However, a study that systematically explores items for an assessment of patients with PSC in CCS is still missing from the literature. Only two scales have been developed for the assessment of patients with cachexia. Focusing on patients with human immunodeficiency virus infection and cancer, Cella et al. 22 generated the Functional Assessment of Anorexia/Cachexia Therapy (FAACT) questionnaire to measure patients quality of life in general and to assess their concerns about symptoms associated with anorexia/cachexia during megestrol acetate therapy. Ribaudo et al. 23

3 Vol. 46 No. 6 December 2013 Psychosocial Consequences of Cancer Cachexia 797 shortened the FAACT questionnaire by reducing the number of anorexia/cachexia subscale items to 12 questions. Our study focuses on patients suffering from cancer cachexia only, and we used a systematic approach to develop a basic item pool to assess PSC in CCS. Methods The pilot item bank was developed using a multistep process including item bank generation, revision, and clinical experts consensus, and a two-step, cross-sectional validity study with two face validities with independent patient cohorts. A study protocol for the crosssectional validity study was created in 2008/ 2009 and approved by the local ethics committee of St. Gallen, Switzerland in early The study design is summarized in Fig. 1. Item Bank Generation and Clinical Experts Consensus on Pilot Item Bank Based on the findings from the 2007 study by Strasser et al., 20 an item pool with 132 questions was created by F. S. Adopting the procedure of Sudman and Bradburn, 24 the third author (J. W.) checked the item bank for double (questions that asked about more than one topic), similar, overly long, and questions with complex wording. Then, an internal expert group (three nutritionists, five nurses, and three doctors) judged the questions, marking them as to leave unchanged, to modify, and to delete. Each question had to be judged by one expert from each disciplinary group, namely nutritionists, nurses, and doctors. The experts were not members of the research group and had at least one year s experience in the field of palliative cancer care. Patient Characterization and Eligibility for the Cross-Sectional Face Validity Study Eligible patients were adults (older than 18 years) with incurable advanced cancer, cognitively able to complete a semistructured interview (tested with the short Mini-Mental State Examination 25 ), and meeting the cachexia definition characteristics (weight loss of 2% or more in two months or of 5% or more in six months). 1 The responsible clinical physician and nurse had to consider eating issues and weight loss as a care priority and as distressing for the patients and their families. Patients were characterized by tumor type, age, sex, body weight and height, living situation, nationality, education, religion, and symptom burden (Edmonton Symptom Assessment Scale 26 and Hospital Anxiety and Depression Scale [HADS] 27 ). The subjectively perceived importance of eating was assessed by the patient s own words and prompted with the ad hoc questions: How important is food to you?, How important is it for your partner to eat meals with you?, and What percentage of the food that you typically used to eat do you eat now? The subjectively perceived eatingrelated suffering of each patient (and the perceived strain on the family/partner) was explored by another three questions, adopted from Strasser et al., 20 which were answered by the patient using a zero to 10 visual analogue scale (0 ¼ no problem and 10 ¼ maximum problem). Common Data Collection Procedures for the Cross-Sectional Face Validity Study A sample size of 20 patients (15 for Step 1 and five for Step 2) was considered sufficient to explore the face validity of items because the item bank had been generated and already checked by specialists/experts. All interviews were held in hospital and conducted by H. H., who was trained in communications skills and directly supervised by senior palliative care specialists (F. S., R. O., and S. d. W.-L.). Before the interviews, written informed consent, patient characteristic data, and information about eating and cachexiarelated issues were obtained. Each patient was then asked the questions from the pilot item bank during a semistructured interview. Patients were encouraged to interrupt the interview at any time, that is, in case of fatigue, misunderstanding or to make comments on the question. The interviews concluded with an open discussion about possible adverse consequences of the interview and an invitation to talk about possible missing questions. Specific Data Collection Procedures for Step 1 Taking into account the burden on patients, no more than 50 questions were asked of each patient. Two sets with 21e25 questions each

4 798 H ane et al. Vol. 46 No. 6 December 2013 Strasser et al. 19 : "Fighting a losing battle" Compilation of pilot item bank from patients answers Item bank generation, revision, and clinical experts consensus Cross-sectional face validity study First step face validity Internal expert round concerning similarity Assessment for grading Assessment for comprehensibility Assessment for emotional effects Evaluation of the results concerning age, education, and HADS score Evaluation of the reasons for reduced comprehensibility (foreign words, levels, age, education, and HADS score) Evaluation of the results concerning age, education, and HADS score Compare with results from internal expert rounds Discussion in the palliative research group Reformulation of critical questions Second step face validity (reassessment for comprehensibility) Compare with results from internal expert rounds Discussion in the palliative research group Final item bank Fig. 1. Flowchart of the study design. HADS ¼ Hospital Anxiety and Depression Scale. had to be presented to 15 patients. Each item was evaluated by six patients regarding the ease of comprehension and the potential for triggering negative emotions ( Does this question affect you in any way? ). The patients also completed a quantitative grading on a Likert scale ( not at all, a little bit, moderately, and much ) concerning the extent to which they felt personally affected by each question. Data Analysis for Step 1 H. H. performed a content analysis of the patients answers with regard to comprehensibility, triggering emotions, and quantitative

5 Vol. 46 No. 6 December 2013 Psychosocial Consequences of Cancer Cachexia 799 grading. In addition, each item was examined for similarity and structure of the question (subordinate clauses/levels, diction, vocabulary, and formulation). Problematic questions were identified and reformulated by H. H. The findings were presented to the palliative research group in cancer cachexia (H. H., F. S., R. O., and S. d. W.-L.); the group decided to delete all questions rated by all the internal experts (three nutritionists, five nurses, and three doctors) as to delete and to definitely keep all questions assessed by all internal experts as to leave unchanged. The pairs of questions that had been found to be similar to each other by H. H. were presented to a new independent, internal expert group (three nutritionists, three doctors, and two nurses). The group was asked if any information would be lost by deleting one question in each pair. The possible reasons for stating not comprehensible, triggering emotions, and grading were further studied, taking into consideration demographics (age and education) and HADS score. 27 Specific Data Collection Procedures for Step 2 Using the same data collection procedures, the comprehensibility of the reformulated questions was tested once more. Every reformulated question was assessed as comprehensible or not comprehensible five times. To assure consistency, nine already-approved items from the first step were added to the reformulated questions. Data Analysis for Step 2 A consensus decision was made by the palliative care research group in cancer cachexia (H.H.,F.S.,R.O.,andS.d.W.-L.)todelete the questions that were assessed as not comprehensible by both two patients and two experts (during the first internal expert round). The possible reasons for stating not comprehensible were further studied, taking into consideration demographics (age and education) and HADS score. 27 The remaining questions formed the definitive item bank and were discussed by the research group with respect to whether they could be categorized into the domains proposed in Strasser et al. 20 (patient distress, partner distress, and couple distress). Results Clinical Experts Consensus on Pilot Item Bank In this study, an item bank comprising 132 questions was created by Strasser et al. 20 and revised by J. W. The analysis revealed an average length of 13.7 words per question. The research process included the deletion of 10 duplicate questions, deletion of one question because it was considered to upset patients, and simplification of 68 questions. Cross-Sectional Face Validity Study Patient Characteristics, Step 1 and Step 2. A total of 20 patients participated in the study (18 inpatients and two outpatients). Ten patients were male and 10 patients were female. On average, one interview took about 40 minutes. Patient characteristics are summarized in Table 1. Comprehensibility, Step 1 and Step 2. In Step 1, 36 questions were assessed as not comprehensible (three on three occasions, five on two occasions, and 28 on one occasion). Looking at the structure, 33 of the 36 questions contained subordinate clauses/levels, and the remaining three questions contained foreign words like aversion, ambivalence, and alimentary supplements. Comparison of the patients assessments with the results from the internal expert rounds revealed that one of these 36 questions had to be deleted because it was assessed as to delete by all experts. However, four questions were assessed as to leave unchanged by all experts and, therefore, could be used for the item bank without having to be retested. The remaining 31 questions were reformulated and reevaluated in Step 2 by five patients; five questions were again assessed once and two questions twice as not comprehensible. The added nine questions from the first step were again never assessed as not comprehensible. Comparison of the patient reevaluations with the results from the internal expert round showed that one of the reformulated questions could be deleted. The analysis of the patients assessments concerning comprehensibility showed no correlation to age, education, or HADS score. Thus, there was no indication for a lower comprehensibility if the patient had an elevated HADS

6 800 H ane et al. Vol. 46 No. 6 December 2013 Table 1 Patient Characteristics Age, median (range) (n ¼ 20) 64 (48e75) Marital status: married/single/divorced (n ¼ 20) 16/1/3 Female 10 Male 10 Religion (Catholic-Protestant/nondenominational) (n ¼ 20) 17/3 Nationality (Swiss/other) (n ¼ 20)? 18/2 Education (oblig. school/professional school/technical college) (n ¼ 20) 5/13/2 Tumor type (n ¼ 20) Urogenital 5 Upper gastrointestinal and pancreas 5 Colorectal 2 Breast 1 Lung and mesothelial 6 Head/neck 1 Weight loss Last two months, (mean) % (n ¼ 20) 9 (2.4e20) Last six months, (mean) % (n ¼ 11) 6.6 (2.5e17) Body mass index, kg/m 2, median (range) (n ¼ 20) 21 (14.6e41) Hospital Anxiety and Depression Scale (0e21 points, 0 ¼ minimum, 21 ¼ maximum) Anxiety (range) (n ¼ 20) 5.6 (1e9) Depression (range) (n ¼ 20) 9.5 (4e15) Subjective eating-related suffering (VAS 0e10, 0 ¼ minimum, 10 ¼ maximum) Problems in daily life in the context of cancer (range) (n ¼ 20) 7.8 (0e10) Burden through feeding problems (range) (n ¼ 20) 7.3 (4e10) Burden for partner through feeding problems (range) (n ¼ 20) 7.4 (0.5e10) Amount of eating in comparison with time before cancer, % (n ¼ 20) Edmonton Symptom Assessment System (VAS 0e10), n ¼ 15 Pain 3.1 Fatigue 6.9 Nausea 1.9 Depression 1.5 Anxiety 1.3 Dizziness 1 Appetite 6.1 Well-being 5.2 Dyspnea 1.6 VAS ¼ visual analogue scale. score, lower education, or was of an advanced age. The working flowchart concerning comprehensibility is summarized in Fig. 2. At the end, our question bank comprised 117 questions. Subjective Importance of Eating (Step 1). Nearly all patients considered eating as an important factor of their social life and as an important occasion for communication and conviviality. Also, pleasure, sensualism, and enjoyment of eating were frequent comments. Moreover, many patients stated that mealtimes provide an occasion for them to communicate with their partners. Similarity (Step 1). Eight questions that were problematic regarding similarity were presented to the independent, internal expert group. These experts agreed that another two questions could be excluded because of the concordant finding that no information would be lost by their deletion. Grading (Step 1). With regard to the extent to which they felt personally affected by each question, patients graded six questions as much four to five times and 25 questions as not at all five to six times. There was no correlation between level/subordinate clauses and grading. All 15 patients from Step 1 used both highest and lowest gradings ( not at all and much ). The patients who were found to use either the highest or the lowest grading most frequently were checked for any correlations with their age, education, and HADS score. However, patients with a higher HADS score (depression subscore), higher age, and lower education did not grade questions more often as not at all / much than patients with a lower HADS score. Emotional Effects (Step 1). In the first step of the face validity study, 22 questions were assessed as possibly triggering negative emotions.

7 Vol. 46 No. 6 December 2013 Psychosocial Consequences of Cancer Cachexia 801 Strasser et al. 19 : "Fighting a losing battle" Compilation of 132 questions Review (similarity, wording, and exclusion of double questions) -10 questions deleted- Interdisciplinary expert judgment for content -1 question deleted- First step face validity with 121 questions Internal expert round concerning similarity Delete 2 questions 33 Questions with levels 3 Questions with foreign words 36 Questions rated not comprehensible 3 Questions 3 times rated not comprehensible Containing 4 levels 16 Questions with 3 or 4 levels Compare with results out of internal expert rounds 5 Questions rated twice with not comprehensible 17 Questions with 2 levels Delete 1 question leave 4 questions reformulate Questions rated once with not comprehensible 5 Questions rated once, 2 Questions twice with not comprehensible Second step face validity with 40 (31) questions Compare with results out of expert internal rounds Delete 1 question leave 30 questions 117 Questions remain Fig. 2. Flowchart of working process. During the interviews, four patients became emotional in response to one or more of the questions, but no patient showed strong emotion, such as anger, despair, or tearfulness. A comparison of patients who thought that questions could trigger emotions with those who thought vice versa found that there was no difference in HADS score level. Examining the

8 802 H ane et al. Vol. 46 No. 6 December 2013 questions that triggered emotions, it became evident that they concerned potential problems with the patient s partner or family (e.g., Does it cause you distress that your partner can no longer express his/her affection through food and cooking? ), which corresponds with the findings of Holden 11 and Addington-Hall and McCarthy. 12 Spontaneous Comments and Missing Questions (Step 1 and Step 2). All patients answered all the questions presented to them, and all commented that the importance of eating becomes manifested when it becomes difficult, and that the quality of life of patients with cancer cachexia is greatly affected when it is no longer possible to eat normally. Possible Domains (Step 2). In discussion, the research group of this project agreed that it would be useful to have different domains for PCS of CCS as done in previous studies. 9,20 The proposed domains are presented in Table 2, along with examples of questions from the item pool within each domain. However, domains and structure of the question pool require further research, given the limitations of the study (see Discussion section). A brief overview and a summary of all the results are given in Fig. 3. Discussion Focusing on patients suffering from cancer cachexia, we used a systematic approach to make important steps toward developing an assessment tool for the identification of PCS of CCS. By using a systematic, multistep approach, we developed an item pool of PSC with questions that were found comprehensible and not to trigger unnecessary negative emotions. The FAACT screening tool 22 is considered to be very helpful for screening purposes in this patient group. However, it is worthwhile exploring the problems of PSC in CCS in more detail. In comparison with the Table 2 Possible Domains of Psychological Consequences of Cancer Cachexia Syndrome and Corresponding Items From the Item Bank Possible Domains Examples Patient distress Partner distress Couple distress/ Family distress Social distress Existential distress Physical distress Emotional distress Does your weight loss trigger deeper fears in you? Do you worry about the reasons for your weight loss? Are you fed up with your constant struggle against changes in your appetite? Does it distress you that your partner s feelings might be hurt when, due to your lack of appetite, you cannot eat the food he/she prepared? Does it cause you distress that your partner can no longer express his/her affection through food and cooking? Does it cause you distress to see your partner gaining weight involuntarily because he/she eats your leftovers? Were you able to establish a different/new quality of life regarding your relationship with your partner by focusing on things other than food? In your relationship, have you been experiencing feelings that alternate between the pressure to eat and the acceptance that it is no longer possible? Does it cause you distress that, due to your eating problems, your family puts much more thought into grocery shopping than in the past? Does it cause you distress to know that your family feels helpless when it comes to offering you specific support. Does it cause you distress that you avoid meeting people because you feel physically unattractive due to your weight loss? Are you afraid that you may soon be a burden to others as a consequence of your eating problems? Do you find that, due to your weight loss, you are no longer able to pursue all of your interests/hobbies? Are you distressed because you fear that weight loss means a shorter life expectancy? Are you worried that your weight loss might lead to death? Are you worried about a possible connection between weight loss and tumor growth? Do you lose your appetite when you are faced with a full plate of food? Does it cause you distress that you do not have much of an appetite? Has the mere smell of food ever caused you to be sick? Do you ever have moments when you feel that you have accepted the fact that you will no longer be able to improve your food intake and weight? Does it cause you distress to weigh yourself? Does it cause you distress that you have little influence on your weight loss?

9 Vol. 46 No. 6 December 2013 Psychosocial Consequences of Cancer Cachexia 803 -Important factor of social life 22 Questions possibly triggering negative emotions -4 Patients became emotional -No difference in HADS level -8 Questions problematic -Presentation to independent, internal experts -Deletion of 2 questions -Occasion for communication and conviviality -Pleasure, sensualism -Occasion for communication -Change from daily pleasure to essential ill See Fig. 2 comprehensibility similarity subjective importance of eating emotional effects Results patient characteristics spontaneous comments and missing questions grading possible domains See Table 1 -Importance of eating becomes manifest when it is no longer possible to eat normally -Quality of life is greatly affected by eating troubles -6 Questions graded with much 4 5 times -25 Questions graded with not at all 5 6 times -All patients used highest and lowest level -No correlation with age, education and HADS score -No correlation between grading and level/subordinate clauses See Table 2 Fig. 3. Summary and overview of the results. HADS ¼ Hospital Anxiety and Depression Scale. FAACT 22 and the shortened version of the FAACT, 23 our items cover physical well-being, social/family well-being, emotional well-being, and additional concerns 22 in more detail. Expanding on the FAACT, more detailed questions, belonging to several possible domains, as proposed in Table 2, are needed to be able to assess this complex problem fundamentally. For example, the question Do you feel distant from your friends? 22 can be replaced by the questions Does it cause you distress that you avoid meeting people because you feel physically unattractive due to your weight loss? and Does it cause you distress that you spend less time with other people because you lack energy

10 804 H ane et al. Vol. 46 No. 6 December 2013 due to your weight loss? from our item pool because it is important to know why someone has problems affecting his/her social wellbeing, if appropriate intervention is to be offered. By generating six superordinated domains, Reid et al. 9 simplified and revealed the complex dynamics of PSC, confirming that cancer cachexia has far-reaching implications for patients and their families. Although Strasser et al. 20 focused on male patients and their female partners and interviewed male and female patients mostly without their partners, some parallels can be observed between the studies. Of course, the item bank was developed based on a study with predominantly male patients, but the initial data set also included a few women, the data from whom were not included in the Strasser et al. 20 study. Preliminary information from these patients, and also subsequent observations with patients, suggest that there are no major gender differences. Based also on the systematic review by Oberholzer et al., 16 we do not expect substantial gender differences. As well, in view of societal deviation from the traditional role of woman as cook legitimizes the inclusion of both genders. We propose our item pool be subdivided into domains (Table 2), which reflect and are consistent with the six themes of Reid et al. 9 and integrate the domains proposed by Strasser et al. 20 For example, conflict over food can be seen as a part of patient distress and existential distress. Of course, domains and structure of the item pool require further research. An assessment tool for PSC in CCS would enable doctors to identify and address specific problems of patients with cancer cachexia. A differentiated and detailed assessment tool responds to the findings that weight loss should be acknowledged more readily by health care professionals that information about CCS and prompt interventions to deal with it should be given, and that patients and their families want this problem to be addressed. 17 We focused only on patients with weight losse and eating-related cancer cachexia, and we evaluated whether our questions were usable for an assessment before an intervention/therapy. For the development of dietary and psychosocial interventions, a greater understanding of the management of cachexia by patients is essential. 28 In 2010, Hopkinson et al. 29 published the data from their exploratory Phase II trial. Findings showed that psychosocial interventions have an influence on the experience of CCS. Allocating our questions to possible domains (Table 2) can help assess the particular domain in which the patient s distress is present, and thus possible interventions may be delivered according to the identified need. Our results are based on a small yet representative sample. We are aware that our study has limitations. We included more inpatients than outpatients and interviewed them mostly without partners. As we know that partners also are affected by PCS of CCS, 20 we have to consider that assessing patients together with partners could reveal the need for additional questions for the partner. The results concerning the grading with not at all, a bit, moderately, and much have to be viewed in connection with the high number of questions used in the item bank. There are questions that address a very specific problem out of the complexity of PCS of CCS in detail. Such questions would normally be graded with not at all by patients who do not feel concerned by this aspect. But this kind of very specific question can be useful in the context of item response theoryebased computerized adaptive testing 30 and if the aim is to cover all patient subpopulations. Several studies have highlighted the prevalence of weight losse and eating-related concerns in cancer cachexia. 11,18,19 By quantification of the anxiety related to anorexia in patients and relatives, 18 as well as the prevalence of weight losse and eating-related concerns, 19 it has been shown that concern attributed to weight loss and loss of appetite in cancer cachexia is a prevalent problem. The development of an adequate tool to detect PCS of CCS is crucial because emotional aspects of CCS have been emerging as an important part in the management of CCS. 31 Further research should analyze the questions we have developed with regard to possible domains as proposed in Strasser et al. 20 and as shown in Table 2. It is also possible that the need for more domains will emerge from further research, for example, domains concerning

11 Vol. 46 No. 6 December 2013 Psychosocial Consequences of Cancer Cachexia 805 partner distress independent of the patient, in order that all issues are addressed. Conclusion This article reports, for the first time, an item bank that can be used to assess the wide spectrum of PSC of CCS. A systematic method has been used to take an important first step toward developing a questionnaire for the assessment of PSC of CCS for use in trials and clinical practice. Disclosures and Acknowledgments This study was funded by Ostschweizer Stiftung f ur Klinische Krebsforschung. The authors declare no conflicts of interest. The authors thank the patients for taking part in this study, the multidisciplinary team for the enriching discussions, Nicole Schenk for administrative support, Daniel Kauffmann for provision of literature, and Melanie Fanger for language support. References 1. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011;12:489e Holmes S. A difficult clinical problem: diagnosis, impact and clinical management of cachexia in palliative care. Int J Palliat Nurs 2009;15:320e Mantovani G, Madeddu C. Cancer cachexia: medical management. Support Care Cancer 2010; 18:1e9. 4. Hopkinson J, Corner J. Helping patients with advanced cancer live with concerns about eating: a challenge for palliative care professionals. J Pain Symptom Manage 2006;31:293e Demoor-Goldschmidt C, Raynard B. How can we integrate nutritional support in medical oncology? [In French]. Bull Cancer 2009;96:665e Potter J, Hami F, Bryan T, Quigley C. Symptoms in 400 patients referred to palliative care services: prevalence and patterns. Palliat Med 2003;17:310e Poole K, Froggatt K. Loss of weight and loss of appetite in advanced cancer: a problem for the patient, the carer, or the health professional? Palliat Med 2002;16:499e Strasser F. Eating-related disorders in patients with advanced cancer. Support Care Cancer 2003; 11:11e Reid J, McKenna H, Fitzsimons D, McCance T. The experience of cancer cachexia: a qualitative study of advanced cancer patients and their family members. Int J Nurs Stud 2009;46:606e Higginson I, Winget C. Psychological impact of cancer cachexia on the patient and family. In: Bruera E, Higginson I, eds. Cachexia-anorexia in cancer patients. New York: Oxford University Press, 1996:172e Holden CM. Anorexia in the terminally ill cancer patient: the emotional impact on the patient and the family. Hosp J 1991;7:73e Addington-Hall J, McCarthy M. Dying from cancer: results of a national population-based investigation. Palliat Med 1995;9:295e Hopkinson J, Wright D, Corner J. Exploring the experience of weight loss in people with advanced cancer. J Adv Nurs 2006;54:304e Kaasa S, Loge JH, Fayers P, et al. Symptom assessment in palliative care: a need for international collaboration. J Clin Oncol 2008;26:3867e Knudsen AK, Brunelli C, Kaasa S, et al. Which variables are associated with pain intensity and treatment response in advanced cancer patients? Implications for a future classification system for cancer pain. Eur J Pain 2011;15:320e Oberholzer R, Hopkinson JB, Baumann K, et al. Psychosocial effects of cancer cachexia? A systematic literature search and qualitative analysis. J Pain Symptom Manage 2013;46:77e Reid J, McKenna HP, Fitzsimons D, McCance TV. An exploration of the experience of cancer cachexia: what patients and their families want from healthcare professionals. Eur J Cancer Care (Engl) 2010;19:682e Hawkins C. Anorexia and anxiety in advanced malignancy: the relative problem. J Hum Nutr Diet 2000;13:113e Hopkinson JB, Wright DN, McDonald JW, Corner JL. The prevalence of concern about weight loss and change in eating habits in people with advanced cancer. J Pain Symptom Manage 2006;32: 322e Strasser F, Binswanger J, Cerny T, Kesselring A. Fighting a losing battle: eating-related distress of men with advanced cancer and their female partners. A mixed-methods study. Palliat Med 2007;21: 129e Reid J, McKenna H, Fitzsimons D, McCance T. Fighting over food: patient and family understanding of cancer cachexia. Oncol Nurs Forum 2009; 36:439e Cella D, Bonomi A, Leslie W, Von Roenn J, Tchekmedyian N. Quality of life and nutritional well-being: measurement and relationship. Oncology 1993;7:105e111.

12 806 H ane et al. Vol. 46 No. 6 December Ribaudo JM, Cella D, Hahn EA, et al. Re-validation and shortening of the Functional Assessment of Anorexia/Cachexia Therapy (FAACT) questionnaire. Qual Life Res 2000;9:1137e Sudman S, Brudman N. Asking questions: A practical guide to questionnaire design. San Francisco: Jossey-Bass, Folstein MF, Folstein SE, McHugh PR. "Minimental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189e Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care 1991;7:6e Snaith RP, Zigmond AS. The hospital anxiety and depression scale. Br Med J (Clin Res Ed) 1986;292: Shragge JE, Wismer WV, Olson KL, Baracos VE. The management of anorexia by patients with advanced cancer: a critical review of the literature. Palliat Med 2006;20:623e Hopkinson JB, Fenlon DR, Okamoto I, et al. The deliverability, acceptability, and perceived effect of the Macmillan approach to weight loss and eating difficulties: a phase II, cluster-randomized, exploratory trial of a psychosocial intervention for weightand eating-related distress in people with advanced cancer. J Pain Symptom Manage 2010;40:684e Walker J, Bohnke JR, Cerny T, Strasser F. Development of symptom assessments utilising item response theory and computer-adaptive testingda practical method based on a systematic review. Crit Rev Oncol Hematol 2010;73:47e Hopkinson JB. The emotional aspects of cancer anorexia. Curr Opin Support Palliat Care 2010;4: 254e258.