|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Symptom Experiences: Perceptual Accuracy Between Advanced-Stage Cancer Patients and Family Caregivers in the Home Care SettingByFrom the Faculty of Nursing, University of Manitoba, Winnipeg, Manitoba, Canada. Address reprint requests to Michelle Lobchuk, PhD, Faculty of Nursing, University of Manitoba, Rm 312, Helen Glass Centre for Nursing, Winnipeg, Manitoba, Canada R3T 2N2; email: michelle_lobchuk@ umanitoba.ca.
PURPOSE: This study used a comparative descriptive design to compare family caregivers’ and advanced-stage cancer patients’ perceptions of patients’ multidimensional symptom experiences on presence, frequency, severity, and distress. PATIENTS AND METHODS: A convenience sample of 98 dyads, composed of advanced-stage heterogeneous cancer patients and their caregivers, completed the Memorial Symptom Assessment Scale in the home care setting on a one-time basis. This scale is a 32-item Likert-type scale for assessing the presence, frequency, severity, and distress arising from symptoms in cancer patients. RESULTS: There was confirmation of trends previously described in related studies where, for example, caregivers tend to overreport on symptom experiences. However, the degree of absolute difference between patient and caregiver responses was normally around 1 unit (on a theoretical range of 0 to 4 units). Levels of patient-caregiver agreement were better on more concrete questions related to symptom frequency, severity, and distress than on broad questions related to the presence of a symptom. Patients and caregivers achieved better levels of agreement on physical versus psychological symptoms. CONCLUSION: The findings indicated that family caregivers can provide reasonable proxy or complementary reports on patient symptom experiences of frequency, severity, and distress. However, family caregivers have greater difficulty in achieving high levels of accuracy on psychological versus physical symptoms.
INCREASINGLY, FAMILY members are being expected to provide complex home health care in light of aging patient populations and health care cost restraints in the area of cancer care worldwide. This expectation is often based on the assumption that they already possess such requisite skills and knowledge, despite not having formal training.1,2 However, the provision of more complex caregiving by family members requires a level of knowledge and skill that is of growing concern to clinicians and researchers. Although the health care literature documents skills needed by family caregivers, little to date has been developed in terms of "family caregiving skill" as a concept.3 In an attempt to conceptualize "family caregiving skill," Schumacher et al3 identified nine family caregiving skill processes in a cancer chemotherapy illness situation. Two of these skill processes are "monitoring" and "interpreting" that indicate or signify dimensions of the caregiving role. These authors suggested that clinicians should conduct clinical assessments on the basis of these processes to determine the caregiver’s ability to detect changes in the patient’s condition. It has been suggested that the cancer experience is best understood as an interpersonal one that is shared between patients and their loved ones.4,5 Accumulated evidence suggests that in comparison to patients’ self-reports on symptom events, family caregivers’ ratings seem to be reasonably accurate representations in the terminal phase of the cancer experience.6-9 On the basis of a review (Lobchuk and Degner, manuscript submitted for publication), evidence across health-related quality of life studies describes a consistent response bias of family caregivers to report more negatively than patients on patient symptom experiences. In general, however, investigators have concluded that these differences are of relatively small magnitude, such that only a modest amount of bias would be introduced when substituting patients’ reports for family caregivers’ responses (Lobchuk and Degner, manuscript submitted for publication). Despite supportive evidence on the reliability of family caregivers as reasonable proxy respondents, one major limitation currently exists in this body of literature. There is a growing consensus in health care literature that symptom experiences are not unifocal, but rather multifaceted events comprising underlying components of symptom presence, frequency, severity or intensity, and distress. Nonetheless, symptom tools commonly used in a majority of extant response comparability studies are flawed in that they tend to provide a unifocal measurement of symptoms (eg, severity alone). Moreover, a smaller number of related studies7,9 have used symptom tools (eg, the Symptom Distress Scale10) that tend to confound symptom occurrence with symptom distress items. Therefore, it remains difficult to discern the exact nature of symptom characteristics that caregivers target in their estimates of patient symptom experiences. Given this major shortcoming, this is the first known study to comprehensively compare patient and family caregiver perceptual congruence on the multifaceted nature of advanced-stage cancer patient symptom experiences as captured on the Memorial Symptom Assessment Scale (MSAS).11 To facilitate targeted and timely health care interventions, we made an attempt to identify areas where caregivers are having difficulty in achieving accurate symptom assessments.
Over a 7-month period between January and August 2000, patient and family caregiver dyads were recruited from three palliative home care programs, two inpatient palliative care units, and three outpatient radiation, chemotherapy, and symptom management oncology departments in Winnipeg, Canada. To be included in the study, patients met the following criteria: (1) medically diagnosed with stage III or IV cancer, (2) 18 years of age or older, (3) able to speak, read, and write English, and (4) well enough to participate and give no evidence of mental confusion (that is, patients who met the preset criterion of  24 of 30 on the Folstein’s Mini-Mental Examination12; this tool is commonly used in studies involving patients capable of responding reliably on self-reports of symptom experiences13,14). The criteria for participation of family caregivers included the following: (1) identified by the patient as being primarily involved in the care of the patient in the home setting, and (2) able to speak, read, and write English. Demographic, disease-related, and treatment-related information were obtained from the patient’s chart and through patient and caregiver simultaneous interviews that were conducted in the home setting by the first investigator. Each patient and caregiver completed the MSAS in separate areas of the patient’s residence where their verbal responses could not be overheard by the other dyad member. The MSAS is a four- and five-point Likert-type, 32-unit physical and psychological symptom scale characterized in terms of frequency, severity, and distress. A response item is included that allowed for the absence of a symptom. The MSAS was developed in a heterogeneous population of cancer patients who were diagnosed primarily with metastatic disease, with Cronbach’s alpha reliability levels assessed as 0.88 in the high-prevalence physical symptom, 0.58 in the low prevalence physical symptom, and 0.83 in the psychological symptom subscale groups.11 In the present sample, the reliability estimates on the high-prevalence physical symptom subscale were 0.82 and 0.84 for patients and caregivers, respectively, 0.59 for patients and 0.61 for caregivers on the low prevalence physical symptom subscale, and 0.80 for patients and 0.82 for caregivers on the psychological symptom subscale.
Descriptive statistics such as frequency distributions, means, SDs, and percentages were used to describe the overall sample characteristics in terms of demographic, disease-related, and treatment-related variables, as well as patient symptom experiences as reported by patients and caregivers. Several recommended statistical techniques were used to evaluate broad patterns of patient-caregiver response comparability at the individual and group levels of analysis.15-19 To determine differences between categorical data, the paired
Wilcoxon matched-pairs signed rank tests examined the degree of difference in group mean responses to ordinal MSAS items. The mean directional difference scores (patient minus caregiver ratings) as determined by paired Student’s t tests were standardized by relating these scores to their SDs to interpret the size of observed differences.23 For effect size (d) calculations on paired observations, the following guidelines were used: 0.2, small amount of bias to overestimate; 0.5, moderate; and 0.8, large.23 Discrepancy variables were created by calculating the absolute difference between group ratings on symptom experience item scores (theoretical range of differences: 0 to 4). Statistical analyses were carried out by SPSS Base 9.0.24 Statistical significance was set at P
Of the 159 eligible patients, 98 agreed to participate (62% response). The majority of the patients who agreed to participate (87 of 98, 89%) were recruited from three clinical outpatient settings. Of the 61 nonrespondents, the most frequently cited reasons for refusal were "it’s not the right time for me to do a study," "I don’t feel appropriate for the study," and "I am not interested." No formal comparisons of the differences between responders and nonresponders were undertaken. Because of recruitment difficulty, a total of 27 patients were not surveyed because of the following: recent hospital admission or acute illness (n = 15), wrong diagnosis for the study (n = 10), language barrier (n = 1), and cognitive inability (n = 1). The demographic and medical characteristics of the patients are presented in Table 1. The mean MMSE score for the entire group was 29 (SD 1.30), indicating satisfactory cognitive performance. On the basis of 1995 Canadian Cancer Statistics,25 this patient sample was representative with respect to age and primary diagnosis, except that prostate cancer was somewhat underrepresented and gynecologic and lung cancers were overrepresented. Female patients were overrepresented and male patients were underrepresented in this sample.
The demographic and medical characteristics of the family caregivers are also presented in Table 1. Most of the caregivers stated the following: that they (1) either "frequently" or "usually" talked openly with the patient about the patient’s thoughts and feelings in regard to his or her symptoms, (2) had an "adequate" or a "very good" degree of knowledge on the patient’s thoughts and feelings about his or her symptoms, (3) lived with the patient and were the patient’s spouse, and (4) "frequently" or "always" assisted the patient in coping with his or her medical condition and symptoms. The MSAS was generally acceptable to patients and caregivers, taking 15 to 20 minutes to complete with minimal assistance by the first investigator. However, missing data by caregivers existed on several MSAS items where they expressed that either it was "not an appropriate item" for them to assess (eg, sexual interest/activity) or they simply did not wish to "guess" on the patient’s particular symptom experience (eg, problems with urination, constipation, or feeling sad). This subgroup of family caregivers identified themselves as either the patient’s spouse, parent, daughter, son, or no relation to the patient. It seems that some of the symptoms may have been regarded as either taboo or threatening topics for family caregivers to discuss with patients. Figure 1 summarizes the prevalence of MSAS symptoms in cancer patients as reported by patients and caregivers. The mean ± SD number of patient symptoms as reported by patients was 11.4 ± 6.06, and the range was zero to 26. The mean ± number of patient symptoms as reported by caregivers was 14.12 ± 6.86, and the range was zero to 31, which was significantly higher than patient reports (t = -4.539, df = 97, P = .00). Both patients and caregivers reported the most highly prevalent physical symptoms as: lack of energy, pain, and feeling drowsy. On low-prevalence physical symptoms, both patients and caregivers reported the highest number of observations on shortness of breath and cough. The patients and caregivers reported worrying as the most prevalent psychological symptom and difficulty concentrating as the least prevalent psychological symptom.
Figure 1 also summarizes kappa values, which ranged between 0.22 to 0.70 on high-prevalence physical symptoms, 0.19 to 0.61 on low-prevalence physical symptoms, and 0.16 to 0.48 on psychological symptoms. In terms of differences between categorical data, a disparity in dyadic perceptions was revealed on the presence of 12 of 32 symptom items. Significant differences occurred on all but one (feeling irritable) of six psychological symptoms, where caregivers tended to report the presence of symptoms when the patients did not. Where incongruence existed, caregivers tended to over- rather than underestimate the presence of 11 of 12 discrepant symptoms. The one exception was on dry mouth, where caregivers tended to provide lower reports than patients themselves.
Table 2 presents a comparison of patients’ and family caregivers’ percentage ratings for frequency (ratings
Tables 3 to 5 present levels of agreement (ie, mean scores and ICCs) at the individual symptom and symptom dimension level for high-prevalence physical, low prevalence physical, and psychological symptom subscale groups. First, Table 3 reveals that for high-prevalence physical symptoms, the ICC results ranged between fair to substantial across symptom frequency, symptom severity, symptom distress, and total symptom experience measures. The discrepancy scores (not reported in the tables) ranged between 0.20 and 1.15 units on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. Significant differences in mean scores (accounting for the direction of difference) ranged from -0.55 to 0.47 units on symptom frequency, symptom severity, symptom distress, and total symptom experience. Where significant differences existed, the statistical magnitude (not reported in the tables) was small (d = -0.21 to -0.39) for caregivers to overestimate across the three underlying symptom dimensions and on total symptom experience. Dry mouth was the only symptom where there was a small statistical bias for caregivers to underestimate across dimensions and on total symptom experience.
Table 4 reveals that at the individual level of agreement for low prevalence physical symptoms, the ICC results varied from fair to substantial on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. The discrepancy scores on ranged between 0.22 and 0.85 units on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. Significant differences in mean scores ranged from -0.38 to -0.14 units on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. Where significant differences existed, the statistical magnitude was small (d = -0.19 to -0.39) for caregivers to overestimate across the three underlying symptom dimensions and on total symptom experience for low prevalence physical symptoms. Table 5 indicates that at the individual level of agreement for psychological symptoms, the ICC results for symptom frequency varied between fair and moderate on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. The discrepancy scores ranged between 0.70 and 1.27 units on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. Significant differences in mean scores were obtained across all symptoms and symptom dimensions, with the exception of feeling irritable where no significant differences were found. Statistical differences in mean scores ranged from -0.72 to -0.30 units on symptom frequency, symptom severity, symptom distress, and total symptom experience measures. As compared with patient responses, the statistical magnitude was small to moderate (d = -0.25 to -0.58) for caregivers to overestimate on symptom frequency, symptom severity, symptom distress, and on total symptom experience measures.
At the symptom dimension subscale group level, the ICC results between patient and caregiver reports on symptom frequency, severity, and distress were consistently substantial (range, 0.63 to 0.69), and no significant differences based on z values (P For the high-prevalence physical (n = 12) and psychological (n = 8) symptom subscale groups, the ICC results between patient and caregiver reports on physical symptoms were more highly correlated (ICC = 0.68) in comparison to their reports on psychological symptoms (ICC = 0.32). There is a statistical difference between patient-caregiver ICC ratings on physical and psychological symptoms (z = 5.71; P = .00). The absolute difference between dyad member reports was significantly less on the physical symptom subgroup (0.39 units) versus the psychological symptom subgroup (0.77 units) (z = -4.88; P = .00). The differences in mean ratings were not significant for the physical symptom subgroup between patients (0.94 ± = 0.64) and caregivers (0.96 ± 0.67). In comparison to patients’ reports (1.09 ± 1.47) for the psychological subgroup, caregivers (1.42 ± 0.83) significantly overreported on patient symptom experience. However, the amount of systematic bias for caregivers to overreport was small (d = -0.25) on psychological symptoms. These results tend to suggest that caregivers might achieve better accuracy (ie, less discrepancy) and report a scant amount of bias (d = -0.1) on high-prevalence physical symptoms in comparison to psychological symptom items at the group level.
The overarching construct of "family caregiving skill" by Schumacher et al3 guided this study’s aim to make more explicit the status of family caregiver skill in "monitoring" and "interpreting" the cancer patient’s multidimensional symptom experience. To our knowledge, this is the first study to demonstrate the reliable use of the MSAS in a sample of family caregivers who cared for advanced-stage cancer patients in the home care setting. The MSAS demonstrated strong reliability on the symptom subscales, in both samples of patients and family caregivers. Despite some missing data by caregivers, the length of the MSAS and the appropriateness of the symptom items seemed reasonable to patients and caregivers. There are some possible generalizability limitations that caution the reader to carefully interpret the current study’s results. First, because the results are based on a study sample of patients with different types of cancer and different types of treatment, it is difficult to separate the effects of treatment from disease processes on patients’ and caregivers’ perceptions. However, the central issue underlying this study was not the source of symptom experience, but the measurement of symptom experience arising from whatever source and the levels of agreement between patients’ and family caregivers’ perceptions of symptom experience. Second, others have suggested that family caregivers who care for patients not receiving active treatment likely do not receive the vigilant support of health care professionals from outpatient departments.9 These informal caregivers would probably experience greater difficulty in managing patient symptoms and thus perceive patients to experience higher symptom distress. In this study, however, the majority of the patients were recruited from active treatment outpatient settings where their family caregivers had the advantage of receiving the same type of support and access to symptom management information. Third, this study’s findings were based on the responses of family caregivers who were older and in long-term marital relationships, which may make it difficult to generalize conclusions to unmarried and younger caregivers. Fourth, these findings are generalizable only to patients who are cognitively stable. There are three striking findings of this study. First, there is confirmation of previously described trends on patient and family caregiver perceptions on patient symptoms.4,7-9,20 For instance, this study provided further evidence on a trend that depicts family caregivers as having a tendency to overreport on patient symptom experiences at both the group and individual symptom levels. However, the absolute differences were generally within one of four theoretical units of discrepancy. In addition, agreement levels were better at the group versus individual symptom level where there was a wide range of agreement levels on total symptom experience, for example, for high-prevalence physical symptoms. As noted by Nekolaichuk et al,26 an analysis of individual symptom profiles helps to detect differences at the individual symptom level that might otherwise be masked at the group level of symptoms. Second, this is the first study we know of that systematically compared patient and family caregiver congruency on the presence, frequency, severity, and distress at both the group and individual symptom levels. High levels of agreement occurred when individual symptoms were grouped and analyzed respective to the separate dimensions of severity, frequency, and distress. However, family caregivers tended to overreport more on symptom distress versus frequency and severity. McClement et al27 suggested that it may be more difficult for caregivers to interpret symptom distress that is more subjective in nature than severity or frequency. In addition, when caregivers responded to specific questions on three symptom dimensions, their levels of perceptual agreement were better in comparison to their levels of agreement on the "presence" of the same symptoms. Third, patient and caregiver agreement levels at the group level on high-prevalence physical symptoms were substantial in comparison to psychological symptoms where agreement levels were fair. These findings lend empirical support toward previous work where authors suggested that physical symptoms may in general may be more objective in nature and accompanied by more visual cues that lend themselves toward better agreement.26,28-31 A number of related study findings, including those found in this study, support the generalization that the reliability of family caregivers would be less on symptoms that are more psychological in nature.29,32,33 In summary, this study made more explicit the status of family caregiving skill in regard to the current management of patient symptom experiences within the patient-family caregiver relationship. First, health care professionals need to appreciate that advanced-stage care is a polysymptomatic, multidimensional symptom experience. It is also important to be aware that patients and family caregivers are able to distinguish between each of the dimensions. Furthermore, detailed questioning on the underlying dimensions of severity, frequency, and distress can minimize interpretation and thus promote improved judgment accuracy by family caregivers. Second, health care professionals should not engage in a wholesale rejection of family caregivers’ input on patient symptom experiences. On physical symptoms, they can expect substantial agreement levels and a scant amount of bias for family caregivers to overestimate. On the other hand, a caveat is warranted for psychological symptoms where health care professionals need to carefully interpret the information provided by family caregivers. The bias to overestimate was more pronounced for ratings on psychological symptoms. It is advisable that professional providers verify caregivers’ perceptions on psychological or emotional symptoms by eliciting feedback directly from the patients whenever possible. Future investigation related to the status of the family caregiver’s level of accurate assessment on patient multidimensional symptom experiences should be pursued in order to continue building a conceptual base on "family caregiving skill." Investigators should replicate the current study’s statistical techniques that used various parameters of chance-corrected agreement at both the individual (ie, correlational analysis) and group (ie, mean differences) levels of agreement. For triangulation purposes, investigators might use a multimethod approach (eg, research diaries, observational field notes, or qualitative interviews) where additional context material can provide evidence that further explains patient-family caregiver perceptual differences. In addition, research that examines congruency levels on patient and family caregiver perceptions would be helpful in determining whether caregiver overestimates, underestimates, or accurate inferences impact patient and family caregiver outcomes on psychological or emotive states, and quality of life.
Supported in part by the National Cancer Institute of Canada, the Canadian Cancer Society, a Graduate Scholarship to the first author from the Manitoba Association of Registered Nurses Foundation, and a Canadian Health Services Research Foundation Chair to the second author.
1. Gosselin TK: Family caregivers need nurse support. ONS News 15: 4-5, 2000 2. Pasacreta JV, Barg F, Nuamah I, et al: Participant characteristics before and four months after attendance at a family caregiver cancer education program. Cancer Nurs 23: 295-303, 2000[CrossRef][Medline] 3. Schumacher KL, Stewart BJ, Archbold PG, et al: Family caregiving skill: Development of the concept. Res Nurs Health 23: 191-203, 2000[CrossRef][Medline] 4. Clipp EC, George LK: Patients with cancer and their spouse caregivers. Cancer 69: 1074-1079, 2000 5. Ey S, Compas BE, Epping-Jordan JE, et al: Stress responses and psychological adjustment in patients with cancer and their spouses. Psychooncology 16: 59-77, 1998 6. Hinton J: How reliable are relatives’ retrospective reports of terminal illness? Patients’ and relatives’ accounts compared. Soc Sci Med 43: 1229-1236, 1996[CrossRef][Medline] 7. Kristjanson L, Nikoletti S, Porock D, et al: Congruence between patients’ and family caregivers’ perceptions of symptom distress in patients with terminal cancer. J Palliat Care 14: 24-32, 1998[Medline] 8. Kurtz ME, Kurtz JC, Given CC, et al: Concordance of cancer patient and caregiver symptom reports. Cancer Pract 4: 185-190, 1996[Medline] 9. Lobchuk M, Kristjanson L, Degner L, et al: Perceptions of symptom distress: I. Congruence between patients and family caregivers. J Pain Symptom Manage 14: 136-146, 1997[CrossRef][Medline] 10. McCorkle R, Young K: Development of a symptom distress scale. Cancer Nurs 1: 373-378, 1978[Medline] 11. Portenoy RK, Thaler HT, Kornblith AB, et al: The Memorial Symptom Assessment Scale: An instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer 30A: 1326-1336, 1994[CrossRef][Medline] 12. Folstein MF, Folstein SE, McHugh PR: "Mini-mental state": A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12: 189-198, 1975[CrossRef][Medline] 13. Bruera E, Fainsinger RL, Miller MJ, et al: The assessment of pain intensity in patients with cognitive failure: A preliminary report. J Pain Symptom Manage 7: 267-270, 1992[CrossRef][Medline] 14. Grossman SA, Sheidler VR, Swedeen K, et al: Correlation of patient and caregiving ratings of cancer pain. J Pain Symptom Manage 6: 53-57, 1991[CrossRef][Medline] 15. Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307-310, 1986[CrossRef][Medline] 16. Chambers CT, Reid GJ, Craig KD, et al: Agreement between child and parent reports of pain. Clin J Pain 14: 336-342, 1998[CrossRef][Medline] 17. Lee J, Koh D, Ong CN: Statistical evaluation of agreement between two methods for measurement a quantitative variable. Comput Biol Med 19: 61-70, 1989[CrossRef][Medline]
18. Nelson LM, Lonstreth WT, Koepsell TD, et al: Proxy respondents in epidemiologic research. Epidemiol Rev 12: 71-86, 1990 19. Sneeuw KC, Aaronson NK, Sprangers MA, et al: Compar-ison of patient and proxy EORTC QLQ-C30 ratings in assessing the quality of life of cancer patients. J Clin Epidemiol 51: 617-631, 1998[CrossRef][Medline]
20. Magaziner J, Bassett SS, Hebel JR, et al: Use of proxies to measure health and functional status in epidemiologic studies of community-dwelling women aged 65 years and older. Am J Epidemiol 143: 283-292, 1996 21. Wyller TB, Sveen U, Bautz-Holder E: The Frenchay Activities Index in stroke patients: Agreement between scores by patients and by relatives. Disabil Rehabil 18: 454-459, 1996[Medline] 22. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33: 159-174, 1977[CrossRef][Medline] 23. Cohen J: Statistical Power Analysis for the Behavioural Sciences, ed 2 . Hillsdale, NJ, Lawrence Erlbaum Associates, 1988 24. Norusis JJ: SPSS for Windows Base System User’s Guide: Release 6.0. Chicago, IL, SPSS, 1994 25. National Cancer Institute of Canada: Canadian Cancer Statistics 2000. Toronto, Canada, National Cancer Institute of Canada, 2000 26. Nekolaichuk CL, Maguire TO, Suarez-Almazor M, et al: Assessing the reliability of patient, nurse, and family caregiver symptom ratings in hospitalized advanced cancer patients. J Clin Oncol 17: 1137-1143, 1997 27. McClement S, Woodgate RL, Degner L: Symptom distress in adult patients with cancer. Cancer Nurs 20: 236-243, 1997[CrossRef][Medline] 28. Spiller JA, Alexander DA: Domiciliary care: A comparison of the views of the terminally ill patients and their family caregivers. Palliat Med 7: 109-115, 1993[Medline] 29. Grassi L, Indelli M, Maltoni M, et al: Quality of life of homebound patients with advanced cancer: Assessments by patients, family members, and oncologists. Psychooncology 14: 31-45, 1996[CrossRef] 30. Rogers J, Ridley S, Chrispin P, et al: Reliability of the next of kins’ estimates of critically ill patients’ quality of life. Anaesthesia 52: 1137-1143, 1997[CrossRef][Medline] 31. Rothman ML, Hedrick SC, Bulcroft KA, et al: The validity of proxy-generated scores as measures of patient health status. Med Care 29: 115-124, 1991[Medline] 32. Brunelli C, Costantini M, Di Giulio P: Quality-of-life evaluation: When do terminal cancer patients and health care providers agree? J Pain Symptom Manage 15: 151-158, 1998[CrossRef][Medline] 33. Carlson LE, Ottenbreit N, St Pierre M, et al: Partner understanding of the breast and prostate cancer experience. Cancer Nurs 24: 231-239, 2001[CrossRef][Medline] Submitted January 31, 2002; accepted May 9, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test or McNemar’s test was conducted. To test agreement between categorical data on the MSAS scale, Cohen’s kappa statistic (K) was calculated as a measure of chance-corrected agreement that allows one to compare the ratings of two observers for the same group of objects.
0.40, fair to no agreement.
