Journal of Neuro-Oncology 14: 243-253, 1992. © 1992 Kluwer Academic Publishers. Printed in the Netherlands.
Clinical Study
Quality of life self-reports from 200 brain tumor patients: comparisons with Karnofsky performance scores Norman Mackworth, 1 Patricia Fobair 2 and Michael D. Prados 1
1Neuro-Oncology Service, Brain Tumor Research Center, Department of Neurological Surgery, School of Medicine, University of California, San Francisco, CA 94143;2Department of Radiation Oncology, Division of Radiation Therapy, Stanford University Medical Center, Palo Alto, CA 94305, USA
Key words: Karnofsky performance scale, quality of life, brain tumors, depression Summary To test a method of assessing quality of life, 200 primary brain tumor patients in an outpatient clinic answered a 20-minute questionnaire covering ten aspects of quality of life. These results were compared with Karnofsky performance scale (KPS) scores, taking age into account. Among patients with KPS 90-100 (two-thirds of the patients), the KPS alone was difficult to interpret. The questionnaire, with its specific questions related to the key dimension of well-being, provided a more definitive assessment of status. The central importance of well-being was supported by its strong statistical relationships with other dimensions. Particularly, well-being was related to freedom from depression (p < 0.0001), active social life (p < 0.0001), energy (p < 0.01), and fewer symptoms (p < 0.05). The KPS was more useful in differentiating the other one-third of the patients (KPS 50--80) but was highly sensitive to age. KPS scores, therefore, may have been unreliable. Depression, reported by half of the patients, was not predicted by the KPS when age was excluded from the regression analysis but was related to the scores in well-being and socializing. Neither depression, well-being, nor socializing was influenced by age. Thus, the questionnaire directly assessed these central, emotionally based variables, particularly among patients with satisfactory KPS. Such an assessment is especially crucial as a supplement to the KPS in evaluating brain tumor survivors, whose emotional well-being is often severely challenged by treatment after surgical excision.
Introduction Quality of life has long been defined as a sense of well-being derived from a person's current experience of life as a whole [1]. This sense depends on the person's ability to satisfy his or her basic needs, including sense of safety, social belonging, selfesteem, the attainment of potential, and, most fundamentally, physiological well-being [2, 3]. Cancer and its treatments can threaten all of these aspects of quality of life. In fact, as many as 35 distinct psychosocial losses have been described in cancer patients [4]. One of the most significant of these is
the depression that can follow the use of toxic chemotherapy or radiation treatments. These diverse effects make it necessary to evaluate outcome systematically in psychosocial and physical terms [1, 5-11]. Clinical studies of cancer patients, however, rarely include quality of life as an aspect of the outcome. The Karnofsky performance scale (KPS) (Appendix A) emphasizes physical changes in quality of life over psychosocial changes [12, 13]. This pioneering quality-of-life study considers self-care, daily activity, evidence of disease, and ability to work, but it does not deal directly with the emo-
244 tional or cognitive state of the patient and therefore cannot detect depression. Moreover, the concentration on physical aspects of outcome has resulted in a tendency for older patients to receive lower KPS scores than younger patients. This makes it difficult to use the KPS effectively in comparing patients of different age groups. The KPS has nonetheless been used widely in cancer studies, especially in research on the effectiveness of treatment. For instance, it has been used to randomize and select chemotherapy patients [14] and to evaluate the effects of their treatment [15]. The KPS has also been used in treatment planning and prognosis [16] and in the evaluation of treatment outcome [17]. The reliability and validity of the KPS may be increased if the physician asks the patient standardized questions about changes in weight and energy and about difficulties in performing daily activities [18]. Nonetheless, the scores vary with personal and social motivational factors [19, 20]. Even in an improved KPS study [18], only one of seven supplementary topics (reduced energy) can be regarded as a psychosocial dimension. The assessment of KPS needs to be further supplemented with a separate evaluation consisting of standard questions that explore the psychosocial aspects of recovery or decline. Because the patient is in the best position to describe his or her feelings, these questions should be answered directly by the patient rather than indirectly by the physician. Although quality-of-life studies have been undertaken on children with brain tumors [21, 22], none have been performed on adults with brain tumors [4]. Studies assessing quality of life are rare: only one in ten current randomized, prospective clinical trials involving cancer patients considered psychosocial outcomes. Even in trials that do consider quality of life, the assessment of psychosocial recovery is often inadequate. A recent review described 24 different multidimensional scales designed for cancer patients or commonly used with them [23]. Although no brain tumor-specific studies are described, the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ) by Aaronson et al. [24] allowed for site-specific
questions to be added to the general questionnaire. Site-specific studies for patients with lung cancer [24], breast cancer [25], and Hodgkin's disease [26], have proven reliable in assessing quality of life in these patients. Because no such method has been tested for patients with primary brain tumors, we adapted a questionnaire based on the EORTCQLQ method, which dealt with the concerns these patients frequently bring to our attention. We also added questions concerning memory loss, an important deficit that often affects quality of life in these patients [27]. The questionnaire used in this pilot study was multidimensional and included questions that reflected ten aspects of satisfactory quality of life: a high level of energy, enjoyment of leisure time, normal cognitive ability, good social life, congenial work, satisfactory sex, freedom from depression and symptoms, a sense of well-being, and strong memory skills. We hoped to define the relative importance of the various dimensions and to determine which were most strongly influenced by age. By comparing these findings with the KPS scores obtained the same day, we also hoped to determine how each of these scales could be improved by the addition of information provided by the other.
Methods
Between March and July 1988, 208 patients who were 12 years of age or over and had primary tumors of the central nervous system were seen in the outpatient clinic of the Neuro-Oncology service at the University of California, San Francisco. These patients were asked, before their visit with a physician, to fill out the quality of life questionnaire. No attempt was made to select patients on the basis of histological subtype, extent or status of disease, or type of therapy. The patients who agreed to participate signed an informed consent form and filled out the questionnaire. The questionnaire took approximately 15 to 20 minutes to complete and was selfadministered, although some patients answered orally to ensure that the questions were understood and being given careful consideration. Patients filled out the questionnaire only once, even if they
245 returned to the clinic while the study was still in progress. After responding to the questionnaire, patients met with their physicians, who assigned KPS scores to them. When the patients answered the questionnaire, they did not know the results of laboratory or radiographic tests to be discussed during the visit, and members of the medical staff did not know the results of the questionnaire when determining the KPS. Similarly, the staff administering the questionnaire were unaware of individual patients' medical background until later. The questionnaire (Appendix B) asked patients to respond to questions on a scale from 1 ('not at all') to 4 ('very much'). To encourage patients to read questions carefully, questions were written with either a positive or a negative orientation. Some questions asked about abilities or positive feelings; the most satisfactory response to these was 4 ('very much'). Other questions, which asked whether patients were limited or dissatisfied in certain areas, would be answered by the healthiest patients with 1 ('not at all'). Before scoring, answers to questions in the latter group were reversescored so that higher scores indicated more satisfactory quality of life on all questions. Questions 1-46, which asked about energy, leisure, cognition, socializing, work, symptoms, sex life, depression, and well-being, were slightly modified from the EORTC-QLQ questionnaire [24], and the questions on memory, 47-54, were from Wilson et al. [27].
Data analysis For each patient, a mean score was calculated in each dimension. The strength of association between the KPS scores and the quality of life variables was determined by multiple regression analysis, in which KPS was the dependent variable and the quality of life dimensions were independent variables. Age was included as a separate independent variable. The fundamental quality of life dimension of well-being was then used as a dependent variable and was correlated by a stepwise multiple regression analysis with each of the other dimensions, age, and KPS. Relatively few patients answered questions on sex life and work, and data on these dimensions will be reported separately (Fobair, Mackworth, Varghese, and Prados, unpublished data). Pearson correlation was used to compare the overall quality of life scores with the KPS scores. The scores in the quality of life dimensions that were significantly associated with well-being were added together to form a composite quality of life score. Dimensions that did not have a significant relationship with well-being were not included in these calculations. Finally, the quality of life scores and the KPS scores of the patients who survived 1 year or more after these evaluations were compared to those of the patients who died within the first year.
Results Table 1. Characteristics of the group studied (1) (2)
(3) (4) (5)
195 cases" Type of treatment-surgery + radiation therapy/ chemotherapy Sex male - 110 female - 90 Age mean 41 years median 39 years range 12-75 years KPS b mean 88% median 90% range 50-100%
a Five of the 200 patients did not submit complete data; b KPS: Karnofsky performance scale.
Of the 208 patients asked to complete the questionnaire, 200 patients (110 male, 90 female) agreed to participate (Table 1). For various reasons, the other eight patients did not choose to participate. Because this was a pilot study, no attempt was made to determine whether these patients were different from the others in terms of disease status or other factors. Among the 200 patients who completed the questionnaire, the mean age was 41 years (range 12-75 years). The average time between diagnosis by biopsy and response to the questionnaire was 27 months. All patients had undergone surgery and radiation therapy, and most
246 B
A Number of Patients
Number of Patients 25
60 50
2
40 t
30 t 2O 10 0 50
60
70
80
90
100
7
8
9
10
Karnofeky Performance Statue
11
12
13
14
15
16
17
18
19
20
Questionnaire Score
Fig. 1. Distribution of patients' Karnofsky performance scae scores (A) and composite quality of life scores (B) determined from the questionnaire.
had received or were receiving chemotherapy. The tumors were supratentorial in 75% of cases and infratentorial in the remaining cases. Tumors included glioblastoma multiforme, highly anaplastic astrocytomas, and moderately differentiated (lowgrade) astrocytomas. Of the 200 questionnaires, five were incompletely answered and were not used in the analysis, leaving 195 questionnaires. The KPS score was not available for 40 patients, and these patients were excluded from the analysis in calculations involving the KPS; therefore, data for 160 patients were used for these calculations.
K P S scores
The 160 available KPS scores ranged from 50-100 (mean 88, median 90, standard deviation 11.1) (Fig. 1). The KPS scores were predominantly clustered at the high end of the scale: two-thirds of the patients had KPS scores of 90 or 100. The fact that the KPS scores were unevenly distributed made interpretation of the scores difficult. Age had a considerable effect on the KPS scores. Older patients tended to have lower KPS scores; in fact, age had a greater effect on KPS than did any of the other variables we considered (Table 2).
Table 2. Ranking of quality of life variables by their links to KPS a scores (stepwise multiple regression analysis, n = 160b)
Variables
P value
Interpretation
Ranked in order 1) Age (inversely) 2) Well-being
0.0053 ~ 0.0072 J
< 0.01 Very significant
3) Cognition 4) Energy 5) Leisure 6) Socializing
0.0205 "] 0.0249 0.0317 0.0431
< 0.05 Borderline significance
Depression Memory Symptoms
Not statistically significant in regard to their association with the KPS scores
aKPS: Karnofsky performance scale; bForty of the 200 patients did not have KPS scores.
247
Quality of life results Quality of life scores were available for most patients in all of the dimensions, with two exceptions. First, because most patients were not working fulltime, work could not be used in the comparisons. Second, too few patients completed the section on sex life to allow statistically significant comparisons between sex life and other variables. Multiple regression analysis with well-being as the dependent variable and the other dimensions as the independent variables showed that the scores in depression and socializing had the closest associations with well-being (Table 3). Energy was slightly less important, and reduced symptoms had an even weaker association with well-being. No significant association was seen between the scores in well-being and any of the other dimensions (age, leisure, cognition, or memory). Therefore, overall quality of life scores were determined by adding the well-being score to the scores in depression, socializing, energy, and symptoms.
Relationship between the KPS and quality of life The quality of life results were more evenly distributed than the KPS scores (Fig. 1). Among the 160 patients for whom KPS scores were also available, the overall quality of life scores ranged from 7-20 (mean 15.7, standard deviation 2.37). Patients with lower KPS scores also tended to
have lower quality of life scores (Table 4). But among relatively healthy patients, those with KPS scores of 90 had quality of life scores an average of only one point below those of the patients with KPS scores of 100. This small loss was not statistically significant. The lack of a relationship between KPS and quality of life in these patients (constituting two-thirds of this group) showed that the KPS lacks sensitivity at this level, when age is considered separately. Age, however, increased as KPS decreased (Table 2). Thus, the patients with low KPS scores were more likely to be older, making interpretation of KPS scores alone difficult. The KPS scores were also strongly related to well-being. In most respects, the order of association of the quality of life dimensions with the KPS scores was similar to the order of their associations with well-being. However, two differences were seen. First, age was associated with the KPS score even more strongly than was well-being, although age and well-being were not associated with each other (Table 3). Second, the scores in depression were closely associated with the scores in well-being when age was used as a separate and independent variable. Of the 160 patients for whom both quality of life and KPS scores were available, 30 died within 1 year of their evaluation. Among patients who lived longer than 1 year, both the quality of life and KPS scores were significantly higher than among the patients who died. The KPS appeared to be a more reliable predictor of survival than the results of the
Table 3. The other quality of life variables associated with well-being (stepwise multiple regression analysis, n = 195a)
Variables
P value
Interpretation
Ranked in order 1) Depression 2) Socializing
0.0001) 0.0001
< 0.001 Highly significant
3) Energy
0.0026
< 0.01 Very significant
4) Symptoms
0.0284
< 0.05 Borderline significance
Leisure Age Cognition Memory
Not statistically significant in regard to their association with well-being
Five of the 200 patients did not submit complete data.
248 quality of life questionnaire (Table 5). This again suggests that the KPS is more informative when used to evaluate older and more impaired patients.
Discussion
Evaluation of outcome with the KPS can be augmented by the addition of psychosocial evidence. For patients in this study, the KPS was closely related to age and was related to well-being when age was considered separately in the regression analysis (Table 2). The KPS was also related, although less closely, to changes in cognition, energy, leisure, and socializing. However, the relationship between the KPS and quality of life is insignificant among patients who are relatively healthy (KPS 90-100). Because two-thirds of our patients had either minor complaints or none at all, the KPS appears to be of limited value in assessing the quality of life of most outpatients. The difficulty in interpreting these KPS scores may result from the rather vague distinction between KPS 90 and KPS 100, the presence or absence of 'some signs or symptoms of disease' (Appendix A). Although the KPS appears to differentiate between patients with poorer quality of life, its high sensitivity to age (Table 2) makes it more difficult to interpret than quality of life data. A low KPS reading may represent either a real loss of quality of life in a young patient or merely the fact that the patient is older than 40 years of age. The superiority of the KPS in predicting survival may also have been a result of the fact that patients with higher KPS scores were generally younger. Further studies may test whether questions in some categories
of quality of life predict survival but are not influenced by age. The items most closely associated with well-being were depression, socializing, energy, and to a lesser extent, symptoms. Depression, socializing, and symptoms share another characteristic with well-being: they are the only dimensions that are unaffected by the age of the patients (Fobair, Mackworth, Varghese, and Prados, unpublished data). The evidence from these variables can therefore be taken at face value regardless of the age of the patient. With energy and the other six quality of life dimensions, as well as the KPS score, greater levels of impairment are more common in older patients. The greatest clinical advantage of the quality of life questions was that they were much more sensitive to depression than was the KPS. Although high KPS scores have been associated with stronger positive feelings such as well-being [28], they do not correlate at all with the absence of negative feelings Table 5. Retrospective comparison between quality of life and KPS at predicting death within one year of evaluation"
Quality of life scores
Mean Standard deviation Number of patients
Survivors
Deceased
Difference between means
15.7 2.62 130
14.5 3.14 30
1.2
aStandard error of difference: 0.6007, t-value: 1.99976. P: < 0.05 borderline significance. KPS: Karnofsky performance scale. Karnofsky performance scale scores (divided by 10)
Table 4. Mean quality of life scores achieved at given KPS a levels
KPS score
50-70 80 90 100
Survivors
Deceased
Difference between means
89.9 10.3 130
83.4 11.5 30
6.5
Quality of life score (mean + SD) (n = (n = (n = (n =
20) 34) 53) 53)
a KPS: Karnofsky performance scale.
13.9 14.3 16.1 16.8
+ + + +
2.14 1.86 2.09 2.14
Mean Standard deviation Number of patients
Standard error of difference: 0.2224. t-value: 2.92. P: < 0.01 very significant.
249 such as depression. Because depression affects well-being and is often treatable, it must be measured as accurately as possible. Of the many aspects of the patient's supporting network during his or her fight for life, psychosocial factors such as freedom from depression are increasingly more important than the physical factors as recovery proceeds. Depression resulting from illness can be aggravated by treatment plans using chemotherapy [29], radiation therapy [30], or both. Because patients are often reluctant to volunteer details on depression during a short medical interview, psychosocial losses are often less clinically obvious than physical losses such as lack of energy. In the detection of depression, then, specific structured quality of life questions before the medical interview could be helpful. Wells et al. [31] have recommended that questionnaires precede medical interviews; otherwise, depression will be missed in about half of the depressed patients. If this preliminary questioning includes direct and indirect queries about depression-related behaviors, the medical staff will need less time to reach a full psychosocial diagnosis. Despite the limitations of the KPS, especially in patients who are doing well, a case can be made for combining it with assessment of quality of life. One approach would be to use the KPS as a rough screening test and then to administer the more precise quality of life questionnaire. In addition to the inadequate coverage of psychosocial dimensions by the KPS, two factors support a series of psychosocial questions before medical interviews. One is that the participation of patients in making treatment choices is increasing rapidly. This closer doctor-patient collaboration requires that the physician fully understand the feelings of the patient. Good patient care during chemotherapy or steroid therapy, for example, involves close monitoring of emotional states: in some patients, depression may become so severe that the protocol must be changed. The second reason for adding psychosocial evaluation is the recent growth of psychoneuroimmunology. The links between the mind and body are so close that some authors regard the mind and body as a unified system [32]. Because future biogenetic protocols in
cancer research will call for specifications of the psychosocial backgrounds of the patients being treated, improved ways of gathering psychosocial data will be valuable. Because psychosocial treatment can increase the number of immune cells, assessment of pyschosocial factors before and during treatment will be necessary in trials using interleukin-2 or interferon to stimulate the proliferation of such cells [33]. A depressed cancer patient may benefit from a support group consisting of other patients with lifethreatening diseases. Although some authors question the ability of support groups to affect the course of recovery, particularly with advanced forms of cancer [34, 35], the value of attitudinal healing during early stages cannot be discounted. Statistically based scientific data are confirming clinical evidence that emotional factors can affect the function of the immune system and course of a developing cancer [32, 36-39]. Patients whose regular medical treatment for malignant melanomas is supplemented by participation in a support group attain significantly better scores on the Profile of Mood States (POMS) and have more cancer-fighting killer cells than those without such a support group [40]. Moreover, higher mental and physical activity is associated with better immune function and slower growth of malignant melanomas [41]. A standardized means of assessing the quality of life would allow further investigation of such phenomena. It has been suggested that the questionnaire might be given serially at 3-month intervals to detect quality of life trends. This would allow a more detailed item-analysis. Further work is in progress on the use of the quality of life questionnaire in evaluating the results of different forms of treatment and the effects of tumor histology and location.
Conclusions
The overlap between the KPS and quality of life scores was moderate (r = 0.48). Each of these two methods was useful in particular groups of patients. Quality of life questions were especially useful for
250 the two-thirds of the patients whose quality of life readings were better than average. In these patients, the KPS readings were often insensitive to the psychological dimensions of quality of life, such as depression. The KPS, however, was useful in evaluating the patients whose quality of life scores were below average. With these patients, the KPS readings were definitely related to impairments in well-being, energy, sex life, and leisure. However, even in these patients, the KPS was not sensitive to depression; in fact, changes in depression had the smallest effect of any of the variables on the KPS readings. The other weakness of the KPS is that it is influenced by the age of the patients. In contrast, some of the quality of life dimensions, such as well-being, socializing, depression, and symptoms, were free from this complication. Because both of these methods have implications for diagnosis, treatment, and research, they should be used together in the evaluation of all brain tumor patients.
Acknowledgements This work was supported by a grant from the National Brain Tumor Foundation, and supported in part by NCI Grant 13525. The authors thank Leslie Dana Pech for interviewing assistance and thank Kathleen Lamborn and Anna Varghese for statistical advice.
Appendix A: The Karnofsky Performance Scale Criteria Scale 100 Normal; no complaints; no signs or symptoms of disease. 90 Able to carry on normal activity; minor signs or symptoms of disease. 80 Normal activity with effort; some signs or symptoms of disease. 70 Cares for self; unable to carry on normal activity or to do active work. 60 Requires occasional assistance, but is able to care for most of his or her needs. 50 Requires considerable assistance and frequent medical care.
40 Disabled; requires special care and assistance. 30 Severely disabled; hospitalization is indicated although death not imminent. 20 Very sick; hospitalization necessary; active supportive treatment necessary. 10 Moribund; fatal processes progressing rapidly. 0 Dead.
Appendix B: The Quality of Life Questionnaire We are interested in learning more about you and your health. There are no 'right' or 'wrong' answers. Please answer all questions yourself by circling the number that best applies to you. The information that you provide will remain strictly confidential.
Energy 1. Can you do hard activities at home like scrubbing floors or lifting or moving heavy furniture? 2. If you wanted to, could you participate in active sports such as swimming, tennis or rowing a boat? 3. If you wanted to, could you run a short distance? 4. Can you walk uphill or upstairs? 5. Can you do moderate work at home like moving a chair or pushing a vacuum cleaner? 6. Can you do light work around the house like dusting or washing dishes? 7. Can you walk around inside the house? 8. Do you need help with eating, dressing or using the toilet?
Not at all
A little
Quite a bit
Very much
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Note: The answers corresponding to the most satisfactory state were either 4 ('very much') or 1 ('not at all') initially on these questionnaire sheets. This was to make sure the patients thought out their answers rather than answering all the ques-
251 tions with one particular number. The computer analysis then converted the raw data so that '4' denoted a very satisfactory state, '3' was satisfactory, '2' was unsatisfactory, and '1' was very unsatisfactory. Not at all Leisure 9. Are you less interested than usual in your hobbies and leisure time activities? 1 10. Are you limited in any way in following your regular hobbies and interests? 1 11. Does your condition keep you from following your regular hobbies and interests? 1 Cognition 12. Do you finish things that you start? l 13. Do you have difficulty reasoning and solving problems; for example, making plans, making decisions, learning new things? 1 14. Do you forget a lot; for example, things that happened recently, where you put things, appointments? 1 15. Do you have difficulty doing activities involving concentration and thinking? l 16. Do you make more mistakes than usual? l 17. Do you have trouble keeping your attention on any activity for long? 1 Socializing 18. Are you going out less to visit people? 1 19. Do you often act irritable toward those around you? 1 20. Are you doing fewer social activities with
A little
2
Quite Very a bit much
3
3
4
2
3
4
3
1
2
3
4
1
2
3
4
1
2
3
4
Answer questions 23 to 26 if you are currently working at a job. If you are not currently working, please go to question 27. Not at all
A little
Quite Very a bit much
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
4
2
2
groups of people? 21. Do you talk less with those around you? 22. Do you stay alone much of the time?
4
Work 23. Are you accomplishing as much as usual at work? 24. Do you often act irritable toward your work associates? 25. Are you working shorter hours? 26. Do you work only for short periods of time or take frequent rests?
Circle the number for each statement that best describes how you felt or behaved this way during the past week.
2
2
3
3
4
4
2
3
4
2
3
4
2
2
2
3
3
3
4
4
4
Symptoms 27. Did you have difficulty with feeling or movement on one side of your hobby? 28. Did you have any epileptic seizures? 29. Did you have headaches? 30. Did you feel energetic? 31. Did you need to rest? 32. Were you tired? 33. All in all, did you feel ill? 34. Did you have trouble sleeping?
1
2
3
4
1
2
3
4
1 1 1 1
2 2 2 2
3 3 3 3
4 4 4 4
1
2
3
4
1
2
3
4
2
3
4
2
3
4
Sex life 35. Were you limited in your level of sexual activity? 1 36. How satisfied were you with your level of sexual activity? l
252 Not at all Depression 37. Did you feel sad? 38. Did you feel that you could not shake off the blues even with help from family and friends? 39. Did you feel lonely? 40. Did you feel depressed? 41. Did you feel tearful? 42. Were you bothered by things that usually don't bother you Well-being 43. Did you feel that things were going your way? 44. How satisfied were you with your level of daily activity 45. How satisfied were you with your physical condition? 46. Overall, how satisfied did you feel with your life? Memory 47. Can you recall people's names? 48. Do you remember appointments without writing them down? 49. Can you recall messages without writing them down? 50. Can you always find the words when you are talking? 51. Can you find your way in a well-known place? 52. Are you able to remember faces? 53. Can you recall where you put things? 54. Are you able to avoid saying the same thing twice?
A little
Quite Very a bit much
1
2
3
4
1 1 1 1
2 2 2 2
3 3 3 3
4 4 4 4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Thank you for participating!
References 1. Campbell A: Subjective measures of well being. Am Psychol 31: 117-124, 1976 2. Maslow AH: A theory of human motivation. Psychol Rev 50: 370-396, 1943 3. Maslow AH: The Farther Reaches of Human Nature. Penguin Books, New York 42-43, 1976 4. Donovan K, Sanson-Fisher RW, Redman S: Measuring quality of life in cancer patients. J Clin Oncol 7: 959-968, 1989 5. DeHaes JCJM, Van Kippenberg FCE: The quality of life of cancer patients: A review of the literature. Soc Sci Med 20: 80%817, 1985 6. Duncan W: Caring of curing: Conflicts of choice. J R Soc Med 78: 526-536, 1985 7. Mullah F: Seasons of survival: Reflections of a physician with cancer. N Engl J Med 313: 270-273, 1985 8. Najman JM, Levine S: Evaluating the impact of medical care and technologies on the quality of life: A review and critique. Soc Sci Med 15F: 107-115, 1981 9. Silberfarb PM; Psychiatric problems in breast cancer. Cancer 53: 820-824, 1984 10. Stehlin JS, Bach KN: Psychological aspects of cancer therapy. JAMA 197: 140-144, 1986 11. Weisman AD, Worden JW: The existentialplight in cancer: Significance of the first 100 days. Int J Psychiatry Med 7: 1-15, 1976-1977 12. Karnofsky DA, Abelmann WH, Craver LF, Burchenal JH: The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer 1: 634-656, 1948 13. Karnofsky DA, Burchenal JH: The clinical evaluation of chemotherapeutic agents in cancer. In: Macleod CM (ed) Evaluation of Chemotherapeutic Agents. Columbia University Press, New York 199-205, 1949 14. Aisner J, Hansen HH; Commentary: Current status of chemotherapy for non-small cell lung cancer. Cancer Rep 65: 979-986, 1981 15. Tannock I, Murphy K: Reflections on medical oncology: An appeal for better clinical trials and improved reporting of their results. J Clin Oncol 1: 66-70, 1983 16. Stanley KE: Prognostic factors for survival in patients with inoperable lung cancer. J Natl Cancer Inst 65: 25-32, 1980 17. Berry WR, Laszlo J, Cox E, Walker A, Paulson D: Prognostic factors in metastatic and hormonally unresponsive carcinoma of the prostate. Cancer 44: 763-775, 1979 18. Schag CC, Heinrich RL, Ganz PA: Karnofsky performance scale revisited: Reliability, validity, and guidelines. J Clin Oncol 2: 187-193, 1984 19. Buchanan RB, Blarney RW, Durrant KR, Howell A, Paterson AG, Preece PE, Smith DC, Williams CJ, Wilson RG: A randomized comparison of tamoxifen with surgical oophoreetomy in premenopausal patients with advanced breast cancer. J Clin Oncol 4: 1326-1330, 1986 20. Trojanowski T, Peszynski J, Turowski K, Kaminski S, Goscinski I, Reinfus M, Krzyszkowski T, Pyrich M, Bielawski
253
21.
22.
23. 24.
25.
26.
27.
28.
29.
30. 31.
A, Leszczyk C, Bendarzewska B, Staszczak A, Kozniewska H: Postoperative radiotherapy and radiotherapy combined with CCNU chemotherapy for treatment of brain gliomas. J Neuro-Oncol 6: 285-291, 1988 LeBaron S, Zeltzer PM, Zeltzer LK, Scott SE, Marlin AE: Assessment of quality of survival in children with medulloblastoma and cerebellar astrocytoma. Cancer 62: 1215-1222, 1988 Packer RJ, Sutton LN, Atkins TE, Radcliffe J, Bunin GR, D'Angio G, Siegel KR, Schut L: A prospective study of cognitive function in children receiving whole-brain radiotherapy and chemotherapy: 2-year results. J Neurosurg 70: 707-713, 1989 Celia DF, Tulsky DS: Measuring quality of life today: Methodological aspects. Oncology 4: 29-38, 1990 Aaronson NK, Bakker W, Stewart AL, Van Dam FSAM, van Zandwijk S, Yarnold JL, Kirkpatrick A: Multidimensional approach to the measurement of quality of life in lung cancer clinical trials. In: Aaronson NK, Beckmann JH (eds) The Quality of Life of Cancer Patients. Raven Press, New York 63-82, 1987 Levine MN, Guyatt GH, Gent M: Quality of life in stage I1 breast cancer: An instrument for clinical trials. J Clin Oncol 6: 1798-1810, 1988 Fobair P, Hoppe RT, Bloom J, Cox R, Varghese A, Spiegel D: Psychosocial problems among survivors of Hodgkin's disease. J Clin Oncol 4: 805-814, 1986 Wilson B, Cockburn J, Baddeley A, Hiorns R: The development and validation of a test battery for detecting and monitoring everyday memory problems. J Clin Exp Neuropsychol 11: 855-870, 1989 Yates JW, Chalmer B, McKegney FP: Evaluation of patients with advanced cancer using the Karnofsky Performance Status. Cancer 45: 2220-2224, 1980 Jacobsen PB, Andrykowski MA, Redd WH, Die-Trill M, Hakes TB, Kaufman RJ, Currie VE, Holland JC: Nonpharmacologic factors in the development of post-treatment nausea with adjuvant chemotherapy for breast cancer. Cancer 61: 379-385, 1988 Peck A, Boland J: Emotional reactions to radiation treatment. Cancer 40: 180-184, 1977 Wells KB, Hays RD, Burnam A, Rogers W, Greenfield S,
32. 33.
34. 35.
36.
37.
38.
39.
40.
41.
Ware JE Jr: Detection of depressive disorder for patients receiving prepaid or fee-for-service care. Results from the Medical Outcomes Study. JAMA 262: 3298-3302, 1989 Pelletier KR, Herzing DL: Psychoneuroimmunology: Toward a mindbody model. Advances 5: 27-56, 1989 Fawzy FI, Kemeny ME. Fawzy NW, Elashoff R, Morton D, Cousins N, Fahey JL: A structured psychiatric intervention for cancer patients. II. Changes over time in immunological measures. Arch Gen Psychiatry 47: 729-735, 1990 Angell M: Disease as a reflection of the psyche. N Engl J Med 312: 373-375, 1985 Cassileth BR, Lusk EJ, Miller DS, Brown LL, Miller C: Psychosocial correlates of survival in advanced malignant disease? N Engl J Med 312: 1551-1555, 1985 Calabrese JR, Kling MA, Gold PW: Alterations in immunocompetence during stress, bereavement and depression: Focus on neuroendocrine regulation. Am J Psychiatry 144: 1123-1134, 1987 Levy SM, Herberman RB, Lippman M: Correlation of stress factors with sustained depression of natural killer cell activity and predicted prognosis in patients with breast cancer. J Clin Oncol 5: 348-353, 1987 Levy SM, Herberman RB, Maluish AM, Schlien B, Lippman M: Prognostic risk assessment in primary breast cancer by behavioral and immunological parameters. Health Psychol 4: 99-113, 1985 Solomon GF: Psychoneuroimmunology: Interactions between central nervous system and immune system. J Neurosci Res 18: 1-9, 1987 Fawzy FI, Cousins N, Fawzy NW, Kemeny ME, Elashoff R, Morton D: A structured psychiatric intervention for cancer patients. I. Changes over time in methods of coping and affective disturbance. Arch Gen Psychiatry 47: 720725, 1990 Temoshok L: Biopsychosocial studies on cutaneous malignant melanoma: Psychological factors associated with prognostic indicators, progression, psychophysiology and tumor-host response. Soc Sci Med 20: 833-840, 1985
Address for offprints: M.D. Prados, Department of Neurological Surgery, c/o The Editorial Office, 1360 Ninth Avenue, Suite 210, San Francisco, CA 94122, USA
Clinical Study
Quality of life self-reports from 200 brain tumor patients: comparisons with Karnofsky performance scores Norman Mackworth, 1 Patricia Fobair 2 and Michael D. Prados 1
1Neuro-Oncology Service, Brain Tumor Research Center, Department of Neurological Surgery, School of Medicine, University of California, San Francisco, CA 94143;2Department of Radiation Oncology, Division of Radiation Therapy, Stanford University Medical Center, Palo Alto, CA 94305, USA
Key words: Karnofsky performance scale, quality of life, brain tumors, depression Summary To test a method of assessing quality of life, 200 primary brain tumor patients in an outpatient clinic answered a 20-minute questionnaire covering ten aspects of quality of life. These results were compared with Karnofsky performance scale (KPS) scores, taking age into account. Among patients with KPS 90-100 (two-thirds of the patients), the KPS alone was difficult to interpret. The questionnaire, with its specific questions related to the key dimension of well-being, provided a more definitive assessment of status. The central importance of well-being was supported by its strong statistical relationships with other dimensions. Particularly, well-being was related to freedom from depression (p < 0.0001), active social life (p < 0.0001), energy (p < 0.01), and fewer symptoms (p < 0.05). The KPS was more useful in differentiating the other one-third of the patients (KPS 50--80) but was highly sensitive to age. KPS scores, therefore, may have been unreliable. Depression, reported by half of the patients, was not predicted by the KPS when age was excluded from the regression analysis but was related to the scores in well-being and socializing. Neither depression, well-being, nor socializing was influenced by age. Thus, the questionnaire directly assessed these central, emotionally based variables, particularly among patients with satisfactory KPS. Such an assessment is especially crucial as a supplement to the KPS in evaluating brain tumor survivors, whose emotional well-being is often severely challenged by treatment after surgical excision.
Introduction Quality of life has long been defined as a sense of well-being derived from a person's current experience of life as a whole [1]. This sense depends on the person's ability to satisfy his or her basic needs, including sense of safety, social belonging, selfesteem, the attainment of potential, and, most fundamentally, physiological well-being [2, 3]. Cancer and its treatments can threaten all of these aspects of quality of life. In fact, as many as 35 distinct psychosocial losses have been described in cancer patients [4]. One of the most significant of these is
the depression that can follow the use of toxic chemotherapy or radiation treatments. These diverse effects make it necessary to evaluate outcome systematically in psychosocial and physical terms [1, 5-11]. Clinical studies of cancer patients, however, rarely include quality of life as an aspect of the outcome. The Karnofsky performance scale (KPS) (Appendix A) emphasizes physical changes in quality of life over psychosocial changes [12, 13]. This pioneering quality-of-life study considers self-care, daily activity, evidence of disease, and ability to work, but it does not deal directly with the emo-
244 tional or cognitive state of the patient and therefore cannot detect depression. Moreover, the concentration on physical aspects of outcome has resulted in a tendency for older patients to receive lower KPS scores than younger patients. This makes it difficult to use the KPS effectively in comparing patients of different age groups. The KPS has nonetheless been used widely in cancer studies, especially in research on the effectiveness of treatment. For instance, it has been used to randomize and select chemotherapy patients [14] and to evaluate the effects of their treatment [15]. The KPS has also been used in treatment planning and prognosis [16] and in the evaluation of treatment outcome [17]. The reliability and validity of the KPS may be increased if the physician asks the patient standardized questions about changes in weight and energy and about difficulties in performing daily activities [18]. Nonetheless, the scores vary with personal and social motivational factors [19, 20]. Even in an improved KPS study [18], only one of seven supplementary topics (reduced energy) can be regarded as a psychosocial dimension. The assessment of KPS needs to be further supplemented with a separate evaluation consisting of standard questions that explore the psychosocial aspects of recovery or decline. Because the patient is in the best position to describe his or her feelings, these questions should be answered directly by the patient rather than indirectly by the physician. Although quality-of-life studies have been undertaken on children with brain tumors [21, 22], none have been performed on adults with brain tumors [4]. Studies assessing quality of life are rare: only one in ten current randomized, prospective clinical trials involving cancer patients considered psychosocial outcomes. Even in trials that do consider quality of life, the assessment of psychosocial recovery is often inadequate. A recent review described 24 different multidimensional scales designed for cancer patients or commonly used with them [23]. Although no brain tumor-specific studies are described, the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ) by Aaronson et al. [24] allowed for site-specific
questions to be added to the general questionnaire. Site-specific studies for patients with lung cancer [24], breast cancer [25], and Hodgkin's disease [26], have proven reliable in assessing quality of life in these patients. Because no such method has been tested for patients with primary brain tumors, we adapted a questionnaire based on the EORTCQLQ method, which dealt with the concerns these patients frequently bring to our attention. We also added questions concerning memory loss, an important deficit that often affects quality of life in these patients [27]. The questionnaire used in this pilot study was multidimensional and included questions that reflected ten aspects of satisfactory quality of life: a high level of energy, enjoyment of leisure time, normal cognitive ability, good social life, congenial work, satisfactory sex, freedom from depression and symptoms, a sense of well-being, and strong memory skills. We hoped to define the relative importance of the various dimensions and to determine which were most strongly influenced by age. By comparing these findings with the KPS scores obtained the same day, we also hoped to determine how each of these scales could be improved by the addition of information provided by the other.
Methods
Between March and July 1988, 208 patients who were 12 years of age or over and had primary tumors of the central nervous system were seen in the outpatient clinic of the Neuro-Oncology service at the University of California, San Francisco. These patients were asked, before their visit with a physician, to fill out the quality of life questionnaire. No attempt was made to select patients on the basis of histological subtype, extent or status of disease, or type of therapy. The patients who agreed to participate signed an informed consent form and filled out the questionnaire. The questionnaire took approximately 15 to 20 minutes to complete and was selfadministered, although some patients answered orally to ensure that the questions were understood and being given careful consideration. Patients filled out the questionnaire only once, even if they
245 returned to the clinic while the study was still in progress. After responding to the questionnaire, patients met with their physicians, who assigned KPS scores to them. When the patients answered the questionnaire, they did not know the results of laboratory or radiographic tests to be discussed during the visit, and members of the medical staff did not know the results of the questionnaire when determining the KPS. Similarly, the staff administering the questionnaire were unaware of individual patients' medical background until later. The questionnaire (Appendix B) asked patients to respond to questions on a scale from 1 ('not at all') to 4 ('very much'). To encourage patients to read questions carefully, questions were written with either a positive or a negative orientation. Some questions asked about abilities or positive feelings; the most satisfactory response to these was 4 ('very much'). Other questions, which asked whether patients were limited or dissatisfied in certain areas, would be answered by the healthiest patients with 1 ('not at all'). Before scoring, answers to questions in the latter group were reversescored so that higher scores indicated more satisfactory quality of life on all questions. Questions 1-46, which asked about energy, leisure, cognition, socializing, work, symptoms, sex life, depression, and well-being, were slightly modified from the EORTC-QLQ questionnaire [24], and the questions on memory, 47-54, were from Wilson et al. [27].
Data analysis For each patient, a mean score was calculated in each dimension. The strength of association between the KPS scores and the quality of life variables was determined by multiple regression analysis, in which KPS was the dependent variable and the quality of life dimensions were independent variables. Age was included as a separate independent variable. The fundamental quality of life dimension of well-being was then used as a dependent variable and was correlated by a stepwise multiple regression analysis with each of the other dimensions, age, and KPS. Relatively few patients answered questions on sex life and work, and data on these dimensions will be reported separately (Fobair, Mackworth, Varghese, and Prados, unpublished data). Pearson correlation was used to compare the overall quality of life scores with the KPS scores. The scores in the quality of life dimensions that were significantly associated with well-being were added together to form a composite quality of life score. Dimensions that did not have a significant relationship with well-being were not included in these calculations. Finally, the quality of life scores and the KPS scores of the patients who survived 1 year or more after these evaluations were compared to those of the patients who died within the first year.
Results Table 1. Characteristics of the group studied (1) (2)
(3) (4) (5)
195 cases" Type of treatment-surgery + radiation therapy/ chemotherapy Sex male - 110 female - 90 Age mean 41 years median 39 years range 12-75 years KPS b mean 88% median 90% range 50-100%
a Five of the 200 patients did not submit complete data; b KPS: Karnofsky performance scale.
Of the 208 patients asked to complete the questionnaire, 200 patients (110 male, 90 female) agreed to participate (Table 1). For various reasons, the other eight patients did not choose to participate. Because this was a pilot study, no attempt was made to determine whether these patients were different from the others in terms of disease status or other factors. Among the 200 patients who completed the questionnaire, the mean age was 41 years (range 12-75 years). The average time between diagnosis by biopsy and response to the questionnaire was 27 months. All patients had undergone surgery and radiation therapy, and most
246 B
A Number of Patients
Number of Patients 25
60 50
2
40 t
30 t 2O 10 0 50
60
70
80
90
100
7
8
9
10
Karnofeky Performance Statue
11
12
13
14
15
16
17
18
19
20
Questionnaire Score
Fig. 1. Distribution of patients' Karnofsky performance scae scores (A) and composite quality of life scores (B) determined from the questionnaire.
had received or were receiving chemotherapy. The tumors were supratentorial in 75% of cases and infratentorial in the remaining cases. Tumors included glioblastoma multiforme, highly anaplastic astrocytomas, and moderately differentiated (lowgrade) astrocytomas. Of the 200 questionnaires, five were incompletely answered and were not used in the analysis, leaving 195 questionnaires. The KPS score was not available for 40 patients, and these patients were excluded from the analysis in calculations involving the KPS; therefore, data for 160 patients were used for these calculations.
K P S scores
The 160 available KPS scores ranged from 50-100 (mean 88, median 90, standard deviation 11.1) (Fig. 1). The KPS scores were predominantly clustered at the high end of the scale: two-thirds of the patients had KPS scores of 90 or 100. The fact that the KPS scores were unevenly distributed made interpretation of the scores difficult. Age had a considerable effect on the KPS scores. Older patients tended to have lower KPS scores; in fact, age had a greater effect on KPS than did any of the other variables we considered (Table 2).
Table 2. Ranking of quality of life variables by their links to KPS a scores (stepwise multiple regression analysis, n = 160b)
Variables
P value
Interpretation
Ranked in order 1) Age (inversely) 2) Well-being
0.0053 ~ 0.0072 J
< 0.01 Very significant
3) Cognition 4) Energy 5) Leisure 6) Socializing
0.0205 "] 0.0249 0.0317 0.0431
< 0.05 Borderline significance
Depression Memory Symptoms
Not statistically significant in regard to their association with the KPS scores
aKPS: Karnofsky performance scale; bForty of the 200 patients did not have KPS scores.
247
Quality of life results Quality of life scores were available for most patients in all of the dimensions, with two exceptions. First, because most patients were not working fulltime, work could not be used in the comparisons. Second, too few patients completed the section on sex life to allow statistically significant comparisons between sex life and other variables. Multiple regression analysis with well-being as the dependent variable and the other dimensions as the independent variables showed that the scores in depression and socializing had the closest associations with well-being (Table 3). Energy was slightly less important, and reduced symptoms had an even weaker association with well-being. No significant association was seen between the scores in well-being and any of the other dimensions (age, leisure, cognition, or memory). Therefore, overall quality of life scores were determined by adding the well-being score to the scores in depression, socializing, energy, and symptoms.
Relationship between the KPS and quality of life The quality of life results were more evenly distributed than the KPS scores (Fig. 1). Among the 160 patients for whom KPS scores were also available, the overall quality of life scores ranged from 7-20 (mean 15.7, standard deviation 2.37). Patients with lower KPS scores also tended to
have lower quality of life scores (Table 4). But among relatively healthy patients, those with KPS scores of 90 had quality of life scores an average of only one point below those of the patients with KPS scores of 100. This small loss was not statistically significant. The lack of a relationship between KPS and quality of life in these patients (constituting two-thirds of this group) showed that the KPS lacks sensitivity at this level, when age is considered separately. Age, however, increased as KPS decreased (Table 2). Thus, the patients with low KPS scores were more likely to be older, making interpretation of KPS scores alone difficult. The KPS scores were also strongly related to well-being. In most respects, the order of association of the quality of life dimensions with the KPS scores was similar to the order of their associations with well-being. However, two differences were seen. First, age was associated with the KPS score even more strongly than was well-being, although age and well-being were not associated with each other (Table 3). Second, the scores in depression were closely associated with the scores in well-being when age was used as a separate and independent variable. Of the 160 patients for whom both quality of life and KPS scores were available, 30 died within 1 year of their evaluation. Among patients who lived longer than 1 year, both the quality of life and KPS scores were significantly higher than among the patients who died. The KPS appeared to be a more reliable predictor of survival than the results of the
Table 3. The other quality of life variables associated with well-being (stepwise multiple regression analysis, n = 195a)
Variables
P value
Interpretation
Ranked in order 1) Depression 2) Socializing
0.0001) 0.0001
< 0.001 Highly significant
3) Energy
0.0026
< 0.01 Very significant
4) Symptoms
0.0284
< 0.05 Borderline significance
Leisure Age Cognition Memory
Not statistically significant in regard to their association with well-being
Five of the 200 patients did not submit complete data.
248 quality of life questionnaire (Table 5). This again suggests that the KPS is more informative when used to evaluate older and more impaired patients.
Discussion
Evaluation of outcome with the KPS can be augmented by the addition of psychosocial evidence. For patients in this study, the KPS was closely related to age and was related to well-being when age was considered separately in the regression analysis (Table 2). The KPS was also related, although less closely, to changes in cognition, energy, leisure, and socializing. However, the relationship between the KPS and quality of life is insignificant among patients who are relatively healthy (KPS 90-100). Because two-thirds of our patients had either minor complaints or none at all, the KPS appears to be of limited value in assessing the quality of life of most outpatients. The difficulty in interpreting these KPS scores may result from the rather vague distinction between KPS 90 and KPS 100, the presence or absence of 'some signs or symptoms of disease' (Appendix A). Although the KPS appears to differentiate between patients with poorer quality of life, its high sensitivity to age (Table 2) makes it more difficult to interpret than quality of life data. A low KPS reading may represent either a real loss of quality of life in a young patient or merely the fact that the patient is older than 40 years of age. The superiority of the KPS in predicting survival may also have been a result of the fact that patients with higher KPS scores were generally younger. Further studies may test whether questions in some categories
of quality of life predict survival but are not influenced by age. The items most closely associated with well-being were depression, socializing, energy, and to a lesser extent, symptoms. Depression, socializing, and symptoms share another characteristic with well-being: they are the only dimensions that are unaffected by the age of the patients (Fobair, Mackworth, Varghese, and Prados, unpublished data). The evidence from these variables can therefore be taken at face value regardless of the age of the patient. With energy and the other six quality of life dimensions, as well as the KPS score, greater levels of impairment are more common in older patients. The greatest clinical advantage of the quality of life questions was that they were much more sensitive to depression than was the KPS. Although high KPS scores have been associated with stronger positive feelings such as well-being [28], they do not correlate at all with the absence of negative feelings Table 5. Retrospective comparison between quality of life and KPS at predicting death within one year of evaluation"
Quality of life scores
Mean Standard deviation Number of patients
Survivors
Deceased
Difference between means
15.7 2.62 130
14.5 3.14 30
1.2
aStandard error of difference: 0.6007, t-value: 1.99976. P: < 0.05 borderline significance. KPS: Karnofsky performance scale. Karnofsky performance scale scores (divided by 10)
Table 4. Mean quality of life scores achieved at given KPS a levels
KPS score
50-70 80 90 100
Survivors
Deceased
Difference between means
89.9 10.3 130
83.4 11.5 30
6.5
Quality of life score (mean + SD) (n = (n = (n = (n =
20) 34) 53) 53)
a KPS: Karnofsky performance scale.
13.9 14.3 16.1 16.8
+ + + +
2.14 1.86 2.09 2.14
Mean Standard deviation Number of patients
Standard error of difference: 0.2224. t-value: 2.92. P: < 0.01 very significant.
249 such as depression. Because depression affects well-being and is often treatable, it must be measured as accurately as possible. Of the many aspects of the patient's supporting network during his or her fight for life, psychosocial factors such as freedom from depression are increasingly more important than the physical factors as recovery proceeds. Depression resulting from illness can be aggravated by treatment plans using chemotherapy [29], radiation therapy [30], or both. Because patients are often reluctant to volunteer details on depression during a short medical interview, psychosocial losses are often less clinically obvious than physical losses such as lack of energy. In the detection of depression, then, specific structured quality of life questions before the medical interview could be helpful. Wells et al. [31] have recommended that questionnaires precede medical interviews; otherwise, depression will be missed in about half of the depressed patients. If this preliminary questioning includes direct and indirect queries about depression-related behaviors, the medical staff will need less time to reach a full psychosocial diagnosis. Despite the limitations of the KPS, especially in patients who are doing well, a case can be made for combining it with assessment of quality of life. One approach would be to use the KPS as a rough screening test and then to administer the more precise quality of life questionnaire. In addition to the inadequate coverage of psychosocial dimensions by the KPS, two factors support a series of psychosocial questions before medical interviews. One is that the participation of patients in making treatment choices is increasing rapidly. This closer doctor-patient collaboration requires that the physician fully understand the feelings of the patient. Good patient care during chemotherapy or steroid therapy, for example, involves close monitoring of emotional states: in some patients, depression may become so severe that the protocol must be changed. The second reason for adding psychosocial evaluation is the recent growth of psychoneuroimmunology. The links between the mind and body are so close that some authors regard the mind and body as a unified system [32]. Because future biogenetic protocols in
cancer research will call for specifications of the psychosocial backgrounds of the patients being treated, improved ways of gathering psychosocial data will be valuable. Because psychosocial treatment can increase the number of immune cells, assessment of pyschosocial factors before and during treatment will be necessary in trials using interleukin-2 or interferon to stimulate the proliferation of such cells [33]. A depressed cancer patient may benefit from a support group consisting of other patients with lifethreatening diseases. Although some authors question the ability of support groups to affect the course of recovery, particularly with advanced forms of cancer [34, 35], the value of attitudinal healing during early stages cannot be discounted. Statistically based scientific data are confirming clinical evidence that emotional factors can affect the function of the immune system and course of a developing cancer [32, 36-39]. Patients whose regular medical treatment for malignant melanomas is supplemented by participation in a support group attain significantly better scores on the Profile of Mood States (POMS) and have more cancer-fighting killer cells than those without such a support group [40]. Moreover, higher mental and physical activity is associated with better immune function and slower growth of malignant melanomas [41]. A standardized means of assessing the quality of life would allow further investigation of such phenomena. It has been suggested that the questionnaire might be given serially at 3-month intervals to detect quality of life trends. This would allow a more detailed item-analysis. Further work is in progress on the use of the quality of life questionnaire in evaluating the results of different forms of treatment and the effects of tumor histology and location.
Conclusions
The overlap between the KPS and quality of life scores was moderate (r = 0.48). Each of these two methods was useful in particular groups of patients. Quality of life questions were especially useful for
250 the two-thirds of the patients whose quality of life readings were better than average. In these patients, the KPS readings were often insensitive to the psychological dimensions of quality of life, such as depression. The KPS, however, was useful in evaluating the patients whose quality of life scores were below average. With these patients, the KPS readings were definitely related to impairments in well-being, energy, sex life, and leisure. However, even in these patients, the KPS was not sensitive to depression; in fact, changes in depression had the smallest effect of any of the variables on the KPS readings. The other weakness of the KPS is that it is influenced by the age of the patients. In contrast, some of the quality of life dimensions, such as well-being, socializing, depression, and symptoms, were free from this complication. Because both of these methods have implications for diagnosis, treatment, and research, they should be used together in the evaluation of all brain tumor patients.
Acknowledgements This work was supported by a grant from the National Brain Tumor Foundation, and supported in part by NCI Grant 13525. The authors thank Leslie Dana Pech for interviewing assistance and thank Kathleen Lamborn and Anna Varghese for statistical advice.
Appendix A: The Karnofsky Performance Scale Criteria Scale 100 Normal; no complaints; no signs or symptoms of disease. 90 Able to carry on normal activity; minor signs or symptoms of disease. 80 Normal activity with effort; some signs or symptoms of disease. 70 Cares for self; unable to carry on normal activity or to do active work. 60 Requires occasional assistance, but is able to care for most of his or her needs. 50 Requires considerable assistance and frequent medical care.
40 Disabled; requires special care and assistance. 30 Severely disabled; hospitalization is indicated although death not imminent. 20 Very sick; hospitalization necessary; active supportive treatment necessary. 10 Moribund; fatal processes progressing rapidly. 0 Dead.
Appendix B: The Quality of Life Questionnaire We are interested in learning more about you and your health. There are no 'right' or 'wrong' answers. Please answer all questions yourself by circling the number that best applies to you. The information that you provide will remain strictly confidential.
Energy 1. Can you do hard activities at home like scrubbing floors or lifting or moving heavy furniture? 2. If you wanted to, could you participate in active sports such as swimming, tennis or rowing a boat? 3. If you wanted to, could you run a short distance? 4. Can you walk uphill or upstairs? 5. Can you do moderate work at home like moving a chair or pushing a vacuum cleaner? 6. Can you do light work around the house like dusting or washing dishes? 7. Can you walk around inside the house? 8. Do you need help with eating, dressing or using the toilet?
Not at all
A little
Quite a bit
Very much
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Note: The answers corresponding to the most satisfactory state were either 4 ('very much') or 1 ('not at all') initially on these questionnaire sheets. This was to make sure the patients thought out their answers rather than answering all the ques-
251 tions with one particular number. The computer analysis then converted the raw data so that '4' denoted a very satisfactory state, '3' was satisfactory, '2' was unsatisfactory, and '1' was very unsatisfactory. Not at all Leisure 9. Are you less interested than usual in your hobbies and leisure time activities? 1 10. Are you limited in any way in following your regular hobbies and interests? 1 11. Does your condition keep you from following your regular hobbies and interests? 1 Cognition 12. Do you finish things that you start? l 13. Do you have difficulty reasoning and solving problems; for example, making plans, making decisions, learning new things? 1 14. Do you forget a lot; for example, things that happened recently, where you put things, appointments? 1 15. Do you have difficulty doing activities involving concentration and thinking? l 16. Do you make more mistakes than usual? l 17. Do you have trouble keeping your attention on any activity for long? 1 Socializing 18. Are you going out less to visit people? 1 19. Do you often act irritable toward those around you? 1 20. Are you doing fewer social activities with
A little
2
Quite Very a bit much
3
3
4
2
3
4
3
1
2
3
4
1
2
3
4
1
2
3
4
Answer questions 23 to 26 if you are currently working at a job. If you are not currently working, please go to question 27. Not at all
A little
Quite Very a bit much
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
4
2
2
groups of people? 21. Do you talk less with those around you? 22. Do you stay alone much of the time?
4
Work 23. Are you accomplishing as much as usual at work? 24. Do you often act irritable toward your work associates? 25. Are you working shorter hours? 26. Do you work only for short periods of time or take frequent rests?
Circle the number for each statement that best describes how you felt or behaved this way during the past week.
2
2
3
3
4
4
2
3
4
2
3
4
2
2
2
3
3
3
4
4
4
Symptoms 27. Did you have difficulty with feeling or movement on one side of your hobby? 28. Did you have any epileptic seizures? 29. Did you have headaches? 30. Did you feel energetic? 31. Did you need to rest? 32. Were you tired? 33. All in all, did you feel ill? 34. Did you have trouble sleeping?
1
2
3
4
1
2
3
4
1 1 1 1
2 2 2 2
3 3 3 3
4 4 4 4
1
2
3
4
1
2
3
4
2
3
4
2
3
4
Sex life 35. Were you limited in your level of sexual activity? 1 36. How satisfied were you with your level of sexual activity? l
252 Not at all Depression 37. Did you feel sad? 38. Did you feel that you could not shake off the blues even with help from family and friends? 39. Did you feel lonely? 40. Did you feel depressed? 41. Did you feel tearful? 42. Were you bothered by things that usually don't bother you Well-being 43. Did you feel that things were going your way? 44. How satisfied were you with your level of daily activity 45. How satisfied were you with your physical condition? 46. Overall, how satisfied did you feel with your life? Memory 47. Can you recall people's names? 48. Do you remember appointments without writing them down? 49. Can you recall messages without writing them down? 50. Can you always find the words when you are talking? 51. Can you find your way in a well-known place? 52. Are you able to remember faces? 53. Can you recall where you put things? 54. Are you able to avoid saying the same thing twice?
A little
Quite Very a bit much
1
2
3
4
1 1 1 1
2 2 2 2
3 3 3 3
4 4 4 4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Thank you for participating!
References 1. Campbell A: Subjective measures of well being. Am Psychol 31: 117-124, 1976 2. Maslow AH: A theory of human motivation. Psychol Rev 50: 370-396, 1943 3. Maslow AH: The Farther Reaches of Human Nature. Penguin Books, New York 42-43, 1976 4. Donovan K, Sanson-Fisher RW, Redman S: Measuring quality of life in cancer patients. J Clin Oncol 7: 959-968, 1989 5. DeHaes JCJM, Van Kippenberg FCE: The quality of life of cancer patients: A review of the literature. Soc Sci Med 20: 80%817, 1985 6. Duncan W: Caring of curing: Conflicts of choice. J R Soc Med 78: 526-536, 1985 7. Mullah F: Seasons of survival: Reflections of a physician with cancer. N Engl J Med 313: 270-273, 1985 8. Najman JM, Levine S: Evaluating the impact of medical care and technologies on the quality of life: A review and critique. Soc Sci Med 15F: 107-115, 1981 9. Silberfarb PM; Psychiatric problems in breast cancer. Cancer 53: 820-824, 1984 10. Stehlin JS, Bach KN: Psychological aspects of cancer therapy. JAMA 197: 140-144, 1986 11. Weisman AD, Worden JW: The existentialplight in cancer: Significance of the first 100 days. Int J Psychiatry Med 7: 1-15, 1976-1977 12. Karnofsky DA, Abelmann WH, Craver LF, Burchenal JH: The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer 1: 634-656, 1948 13. Karnofsky DA, Burchenal JH: The clinical evaluation of chemotherapeutic agents in cancer. In: Macleod CM (ed) Evaluation of Chemotherapeutic Agents. Columbia University Press, New York 199-205, 1949 14. Aisner J, Hansen HH; Commentary: Current status of chemotherapy for non-small cell lung cancer. Cancer Rep 65: 979-986, 1981 15. Tannock I, Murphy K: Reflections on medical oncology: An appeal for better clinical trials and improved reporting of their results. J Clin Oncol 1: 66-70, 1983 16. Stanley KE: Prognostic factors for survival in patients with inoperable lung cancer. J Natl Cancer Inst 65: 25-32, 1980 17. Berry WR, Laszlo J, Cox E, Walker A, Paulson D: Prognostic factors in metastatic and hormonally unresponsive carcinoma of the prostate. Cancer 44: 763-775, 1979 18. Schag CC, Heinrich RL, Ganz PA: Karnofsky performance scale revisited: Reliability, validity, and guidelines. J Clin Oncol 2: 187-193, 1984 19. Buchanan RB, Blarney RW, Durrant KR, Howell A, Paterson AG, Preece PE, Smith DC, Williams CJ, Wilson RG: A randomized comparison of tamoxifen with surgical oophoreetomy in premenopausal patients with advanced breast cancer. J Clin Oncol 4: 1326-1330, 1986 20. Trojanowski T, Peszynski J, Turowski K, Kaminski S, Goscinski I, Reinfus M, Krzyszkowski T, Pyrich M, Bielawski
253
21.
22.
23. 24.
25.
26.
27.
28.
29.
30. 31.
A, Leszczyk C, Bendarzewska B, Staszczak A, Kozniewska H: Postoperative radiotherapy and radiotherapy combined with CCNU chemotherapy for treatment of brain gliomas. J Neuro-Oncol 6: 285-291, 1988 LeBaron S, Zeltzer PM, Zeltzer LK, Scott SE, Marlin AE: Assessment of quality of survival in children with medulloblastoma and cerebellar astrocytoma. Cancer 62: 1215-1222, 1988 Packer RJ, Sutton LN, Atkins TE, Radcliffe J, Bunin GR, D'Angio G, Siegel KR, Schut L: A prospective study of cognitive function in children receiving whole-brain radiotherapy and chemotherapy: 2-year results. J Neurosurg 70: 707-713, 1989 Celia DF, Tulsky DS: Measuring quality of life today: Methodological aspects. Oncology 4: 29-38, 1990 Aaronson NK, Bakker W, Stewart AL, Van Dam FSAM, van Zandwijk S, Yarnold JL, Kirkpatrick A: Multidimensional approach to the measurement of quality of life in lung cancer clinical trials. In: Aaronson NK, Beckmann JH (eds) The Quality of Life of Cancer Patients. Raven Press, New York 63-82, 1987 Levine MN, Guyatt GH, Gent M: Quality of life in stage I1 breast cancer: An instrument for clinical trials. J Clin Oncol 6: 1798-1810, 1988 Fobair P, Hoppe RT, Bloom J, Cox R, Varghese A, Spiegel D: Psychosocial problems among survivors of Hodgkin's disease. J Clin Oncol 4: 805-814, 1986 Wilson B, Cockburn J, Baddeley A, Hiorns R: The development and validation of a test battery for detecting and monitoring everyday memory problems. J Clin Exp Neuropsychol 11: 855-870, 1989 Yates JW, Chalmer B, McKegney FP: Evaluation of patients with advanced cancer using the Karnofsky Performance Status. Cancer 45: 2220-2224, 1980 Jacobsen PB, Andrykowski MA, Redd WH, Die-Trill M, Hakes TB, Kaufman RJ, Currie VE, Holland JC: Nonpharmacologic factors in the development of post-treatment nausea with adjuvant chemotherapy for breast cancer. Cancer 61: 379-385, 1988 Peck A, Boland J: Emotional reactions to radiation treatment. Cancer 40: 180-184, 1977 Wells KB, Hays RD, Burnam A, Rogers W, Greenfield S,
32. 33.
34. 35.
36.
37.
38.
39.
40.
41.
Ware JE Jr: Detection of depressive disorder for patients receiving prepaid or fee-for-service care. Results from the Medical Outcomes Study. JAMA 262: 3298-3302, 1989 Pelletier KR, Herzing DL: Psychoneuroimmunology: Toward a mindbody model. Advances 5: 27-56, 1989 Fawzy FI, Kemeny ME. Fawzy NW, Elashoff R, Morton D, Cousins N, Fahey JL: A structured psychiatric intervention for cancer patients. II. Changes over time in immunological measures. Arch Gen Psychiatry 47: 729-735, 1990 Angell M: Disease as a reflection of the psyche. N Engl J Med 312: 373-375, 1985 Cassileth BR, Lusk EJ, Miller DS, Brown LL, Miller C: Psychosocial correlates of survival in advanced malignant disease? N Engl J Med 312: 1551-1555, 1985 Calabrese JR, Kling MA, Gold PW: Alterations in immunocompetence during stress, bereavement and depression: Focus on neuroendocrine regulation. Am J Psychiatry 144: 1123-1134, 1987 Levy SM, Herberman RB, Lippman M: Correlation of stress factors with sustained depression of natural killer cell activity and predicted prognosis in patients with breast cancer. J Clin Oncol 5: 348-353, 1987 Levy SM, Herberman RB, Maluish AM, Schlien B, Lippman M: Prognostic risk assessment in primary breast cancer by behavioral and immunological parameters. Health Psychol 4: 99-113, 1985 Solomon GF: Psychoneuroimmunology: Interactions between central nervous system and immune system. J Neurosci Res 18: 1-9, 1987 Fawzy FI, Cousins N, Fawzy NW, Kemeny ME, Elashoff R, Morton D: A structured psychiatric intervention for cancer patients. I. Changes over time in methods of coping and affective disturbance. Arch Gen Psychiatry 47: 720725, 1990 Temoshok L: Biopsychosocial studies on cutaneous malignant melanoma: Psychological factors associated with prognostic indicators, progression, psychophysiology and tumor-host response. Soc Sci Med 20: 833-840, 1985
Address for offprints: M.D. Prados, Department of Neurological Surgery, c/o The Editorial Office, 1360 Ninth Avenue, Suite 210, San Francisco, CA 94122, USA
