|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 4 (February 15), 2004: pp. 714-724 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.06.078 Measuring Quality of Life in Routine Oncology Practice Improves Communication and Patient Well-Being: A Randomized Controlled TrialFrom the Cancer Research UK Clinical Centre-Leeds, Cancer Medicine Research Unit, St James's University Hospital; and Northern and Yorkshire Clinical Trials and Research Unit, Leeds, United Kingdom Address reprint requests to Galina Velikova, MD, Cancer Research UK Clinical Centre-Leeds, Cancer Medicine Research Unit, St James's University Hospital, Beckett St, Leeds LS9 7TF, UK; e-mail: g.velikova{at}cancermed.leeds.ac.uk
PURPOSE: To examine the effects on process of care and patient well-being, of the regular collection and use of health-related quality-of-life (HRQL) data in oncology practice. PATIENTS AND METHODS: In a prospective study with repeated measures involving 28 oncologists, 286 cancer patients were randomly assigned to either the intervention group (regular completion of European Organization for Research and Treatment of Cancer-Core Quality of Life Questionnaire version 3.0, and Hospital Anxiety and Depression Scale on touch-screen computers in clinic and feedback of results to physicians); attention-control group (completion of questionnaires, but no feedback); or control group (no HRQL measurement in clinic before encounters). Primary outcomes were patient HRQL over time, measured by the Functional Assessment of Cancer Therapy-General questionnaire, physician-patient communication, and clinical management, measured by content analysis of tape-recorded encounters. Analysis employed mixed-effects modeling and multiple regression. RESULTS: Patients in the intervention and attention-control groups had better HRQL than the control group (P = .006 and P = .01, respectively), but the intervention and attention-control groups were not significantly different (P = .80). A positive effect on emotional well-being was associated with feedback of data (P = .008), but not with instrument completion (P = .12). A larger proportion of intervention patients showed clinically meaningful improvement in HRQL. More frequent discussion of chronic nonspecific symptoms (P = .03) was found in the intervention group, without prolonging encounters. There was no detectable effect on patient management (P = .60). In the intervention patients, HRQL improvement was associated with explicit use of HRQL data (P = .016), discussion of pain, and role function (P = .046). CONCLUSION: Routine assessment of cancer patients' HRQL had an impact on physician-patient communication and resulted in benefits for some patients, who had better HRQL and emotional functioning.
The emergence of biopsychosocial models of health and the prevalence of chronic diseases in developed countries, led to an interest in measurements of functioning and quality of life, and their applications in research and practice [1,2]. The classical medical system for history taking and recording of findings, which emerged in the 19th century, is focused on diagnosing acute medical problems [3]. It does not easily support monitoring of physical or functional problems over time. Increasingly, oncologists are expected to monitor effects of cancer on patients' physical and psychosocial well-being, and to include these issues in decision making [4-6]. Good communication between health professionals and patients, and the need for comprehensive supportive care, are recognized to be central to the management of cancer patients [7]. Physicians vary in their ability to elicit psychosocial information [8,9]. The measurement of individual patients’ health-related quality of life (HRQL) can be used in clinical practice to facilitate detection of physical or psychological problems, to monitor disease and treatment over time, and thus improve the delivery of medical care [10-12]. Previous research, including studies in oncology, suggested that individual HRQL reports provide useful information to physicians as well as facilitate communication, but have little impact on patients' well-being [13-18]. Two systematic reviews recommended further research, emphasizing the need for evaluation of repeated measurements [19,20]. Recently, electronic methods of collecting data from patients have allowed real-time HRQL measurements and presentation of results to clinicians, making this approach feasible in busy clinical practices [21,22]. We conducted a randomized study to examine the effects of regular repeated collection and feedback of HRQL data to oncologists. We hypothesized that the intervention would have a positive effect on two primary outcomes: patient well-being and process of care (content of physician-patient communication and management decisions). Secondary outcomes were other process measures (tests, drugs, medical records), continuity of care, and patient satisfaction. This article describes the primary outcomes analysis.
Subjects Patients attending the Leeds Cancer Centre Medical Oncology Clinic at St James's Hospital (Leeds, UK) were eligible if they were commencing treatment, expected to attend the clinic at least three times, fluent in English, not taking part in HRQL studies, and not exhibiting overt psychopathology. Patients were invited to participate in the study after a medical decision had been made during their initial consultation to start cytotoxic or biologic treatment. At their next clinic visit, they were approached with both oral and written information about the study. All oncology consultants and physicians in training (specialist registrars) were invited to participate. The physicians worked in two teams (four to seven physicians each), and over time, patients saw different physicians. The project was approved by the institutional ethical committee. Written informed consent was obtained from patients and clinicians.
Design An optimal experimental design for the study was difficult to achieve for several reasons. The study was conducted in a natural environment (oncology clinics), with two groups of subjects (physicians and patients) who were in continuous complex interactions. The experimental intervention was both at patient level (completion of intervention questionnaires) and physician level (feeding back of HRQL information). Random assignment of physicians was considered, but rejected due to practical limitations. In the Cancer Centre-Leeds (similar to many large oncology practices in the United Kingdom), patient care is provided by teams consisting of four to seven physicians, and over time, patients usually see several different physicians who, if physicians were randomly assigned, might happen to be either in the experimental or the control group. If different clinics were randomly assigned instead of individual physicians, definite differences between patients would result, as the clinics were specialized by cancer site. Therefore, patients were chosen as the units of random assignment, with an analysis of possible physician-sensitizing effect planned at the design stage. Patients providing written informed consent were randomly assigned, and their clinic encounter was audio-recorded and considered as baseline data for the study. Starting with the next clinic visit, patients randomly assigned to groups 1 and 2 regularly completed the touch-screen questionnaires in the waiting room before every encounter, for approximately 6 months. The medical encounters of all patients were tape-recorded. Patient outcomes were measured with a different HRQL instrument (Functional Assessment of Cancer-General [FACT-G] [23]), completed by all patients. In order to separate as much as possible the effects of the intervention questionnaires from the outcome measures, patients were given the outcome questionnaires on paper to complete at home and return by postal service at four time points: after the study baseline encounter, after three on-study encounters (approximately 2 to 3 months), after 4 months, and at study end (approximately 6 months). Process-of-care outcomes were evaluated from audio-taped encounters. Physicians' perceptions of clinical usefulness of the data for individual encounters were assessed.
Sample Size and Random Assignment
Experimental Intervention Touch-screen computers were used, with graphic printouts of results [11,28]. An example of the individual HRQL graphs is presented in Figure 1.
Physicians were trained in interpretation of the questionnaires. A manual was prepared, with description of scales, interpretation of scores, and explanation of the graphs [28,29]. Structured meetings were conducted individually, with each physician to discuss the study and review examples of HRQL and clinical details of real patients. Posters with interpretative information were displayed in clinics. The physicians were asked to review and use the HRQL results during all intervention encounters, unless totally inappropriate. No recommendations for specific responses were made. After seeing each patient, the physicians completed a checklist assessing the clinical usefulness of the HRQL results for the individual encounter. Physicians indicated whether they found the data clinically useful and in what way.
Patient Outcome Measure
Process-of-Care Measures Medical and nonmedical actions were recorded. Medical decisions were defined as decisions on cancer treatment, symptomatic/supportive treatment, investigations, and referrals. Nonmedical actions included advice on lifestyle, coping, and reassurance. The length of the encounters was recorded from the audio tapes.
Coding was performed directly from the audio tapes by three raters (G.V., P.L., L.B.), blinded to patient identity. The interrater reliability was good, with exact agreement of 76% to 100% (median, 95%) and
Analysis
Patient HRQL A random coefficients model was used, since the timing of assessments varied across patients. The model included FACT-G scores over time as the outcome variable; baseline FACT-G score as a covariate; performance status, time, study arm, and study arm x time as fixed effects; and patient and patient x time as random effects. Time was fitted in the model as a continuous covariate, since timing of assessments varied across patients. The intervention group was compared with the control and the attention-control groups according to the study hypothesis and a priori analysis plan. Comparison between the control and attention-control groups was performed in a secondary analysis, which was not part of the original hypothesis testing strategy. Similar models were fitted with subscales scores as outcome variables. The intervention was expected to have a predominant effect on emotional well-being [18].
To illustrate the clinical significance of between-group differences in FACT-G, the individual changes in FACT-G scores were categorized as improvement (follow-up score minus baseline
Process of Care Multiple regression analysis was used for between-group comparison of number of symptoms/functions discussed. Exploratory analysis identified potential covariates for inclusion in the model. Univariate regression models were fitted with number of symptoms as the outcome and the potential covariate as single explanatory variable. Patient and physician sex (P < .01 and P = .02), number of physicians seen (P < .01), and time on study (P = .09) were identified as possible predictors to be retained in the model. Age, extent of disease, performance status, and diagnosis did not satisfy the prespecified criterion (P < .1).
Subgroup Analysis of Intervention Group
Physicians' Use of HRQL Information All analyses were on an intention-to-treat basis. Significance level was set at 5% for primary hypotheses (FACT-G scores and combined communication scores on EORTC symptoms and functions) and at 1% for other comparisons. The analyses were performed with SPSS Version 11.0 for Windows (SPSS Inc, Chicago, IL) and SAS (SAS/STAT User's Guide, Version 6; SAS Institute, Cary, NC) version 8.02 (SAS [SAS/STAT User's Guide, Version 6, SAS Institute] Institute Inc, Cary, NC).
The study was carried out between January 2000 and July 2001. Patients' progress through the study is presented in Figure 2. Nonparticipants (32% of eligible patients) were older than participants (mean age, 61.7 v 54.9 years; standard deviation [SD], 12.40 v 12.52 years; respectively; t417 = -5.1, P < .001). No statistically significant sex difference was found (  21 = 3.63; P = .06). The attrition rate by 6 months was 35% in the intervention arm, 46% in the attention-control, and 35% in the control arm, due predominantly to disease progression. More than 90% of the patients who remained on study completed the outcome questionnaires at each time point.
Table 1 presents the baseline patients and encounters characteristics, demonstrating a good balance of baseline variables between the study arms. In addition to the baseline clinical variables, data on response to treatment after 3 and 6 months was extracted from the medical records and recorded as complete remission (CR), partial remission (PR), stable, or progressive disease. Response was assessed according to the standard clinical practice, using radiological investigations and tumor markers. The response rates were well balanced between the three study arms for remission (CR + PR: 40%, 41%, and 42% for arms 1, 2, and 3, respectively), stable disease (26%, 26%, and 24% for arms 1, 2, and 3, respectively), and progression (31%, 31%, and 31% for arms 1, 2, and 3, respectively).
All 28 physicians working in the unit participated. There were 17 male and 11 female physicians (median age, 33.5 years; range, 26 to 51 years), 22 specialist registrars, and six consultants. The physicians had differing lengths of medical practice (range, 3 to 27 years) and oncology experience (range, 0 to 24 years). Numbers of encounters per physician were approximately evenly distributed across study arms.
Patient HRQL
Overall, FACT-G scores increased over time in all patients, which was likely due to attrition of ill patients. Patients who discontinued the study had significantly lower baseline FACT-G scores than patients completing the study (mean, 68.1 v 73.6; SD, 16.37 v 17.17; respectively; t260 = 2.54, P = .01). FACT-G scores in the control group increased at a faster rate, and the effect of intervention diminished over time. The interaction time x study arm was not significant (P = .32). Figure 4 illustrates the clinical significance of the results, showing that a larger proportion of patients in the intervention arm had clinically meaningful improvement in HRQL after three interventions. The number needed to treat for one patient to get benefit was 4.2 [31].
Process of Care The comparison of process measures between the three groups is presented in Table 3. The number of EORTC QLQ-C30 symptoms mentioned during the encounters was higher in the intervention group in comparison with the control group (P = .03). More frequent discussion of chronic nonspecific symptoms (difficulty sleeping, lack of appetite, and fatigue) was observed, without prolonging the encounters. As expected, there was no between-group difference in the number of other symptoms discussed (P = .81), suggesting that it was still possible to cover patient and disease-specific problems. No effect on patient management was detected.
Physicians may be expected to differ in their communication or decision-making styles. The possibility of a "doctor" effect was considered for both analyses of patient and process of care outcomes. As the data had hierarchical structure with patients nested within physicians, nested models were fitted with the outcome variable either FACT-G after 3 visits or symptoms discussed. No significant effect for "doctor" was found and adding "doctor" as a random effect in the models did not affect the intervention estimates (details of analysis not shown).
Subgroup Analysis of Intervention Group
Physicians' Use of HRQL Information According to the completed checklists after the individual encounters, physicians found the HRQL information clinically "very useful/quite useful" in 43% of encounters, "somewhat useful" in 28%, "little useful" in 21%, and "not useful" (or missing response) in 9%. The HRQL data provided an overall assessment of patients (69%), additional information (33%), or identified problems for discussion (27%). It contributed to patient management in 11% of encounters. Oncologists did not use the HRQL information if "the data was irrelevant for the purpose of the encounter or irrelevant to patients major problems."
As stated in the main study hypotheses, the intervention resulted in a significant increase in the discussion of chronic symptoms and had a positive impact on patients' well-being. Of particular interest is the finding that routine repeated HRQL measurements with feedback of results may lead to benefits for some patients with improvement in their overall and emotional well-being. A larger proportion of patients in the intervention group showed clinically meaningful improvement in HRQL than patients in attention-control and control groups. In addition, the subgroup analysis of the intervention group suggested that the explicit use of HRQL information during encounter was associated with a clinically significant improvement (> 7 points) in patient well-being. These results differ from previous reports which generally show little or no patient benefit [13-20,32]. The difference could reflect our use of repeated measurements and feedback [17,18] and of a cancer-specific instrument as an outcome measure [16]. In our practice patients may see different physicians sequentially so the HRQL information might be more useful than in a single handed practice [15,16]. We trained the physicians to some extent and asked them to use the information regularly, which may have enhanced the effects of the intervention, as suggested by the subgroup analysis of the intervention group [20]. The other similar studies did not appear to have asked the participating physicians to use regularly the available HRQL data [16-18]. Our study suggested that completion of the questionnaires itself may have effect on patient well-being, particularly on symptom control (FACT [physical well-being]), regardless of whether the results are fed back to physicians. This result was difficult to compare with previous research, as the only other study with attention-placebo group did not find any effect on patient well-being [32]. Our study was not primarily designed to compare attention-control with control group and it will be necessary to repeat this observation in future studies. It could be speculated that HRQL questionnaires may encourage patients to discuss more issues, but no significant difference was found in the encounter analysis. Alternatively, physicians' use of HRQL information in the intervention group may have been neither efficient, nor sufficiently frequent and consistent. Oncologists explicitly used the information in only 64% of the third encounters. Note that the positive effect on patient emotional well-being was associated with feedback of data, not just completion of instruments. If future studies confirm the suggestion that regular measurement of HRQL may influence patient well-being (overall quality of life and physical functioning), this will have implications for the design and interpretation of therapeutic clinical trials with HRQL measurement as an outcome. Most clinical trials do not include HRQL assessments as regularly as this study, so the effects are likely to be smaller, if any. Nevertheless, our findings raise interesting points for thought and future research. Routine use of HRQL information during the encounters had an impact on physician-patient communication without prolonging the interviews. Chronic nonspecific symptoms were discussed more frequently. With the simple coding system we used, we could not detect a significant effect on patient management. These results are consistent with other studies, confirming influence on physician-patient interactions, but not medical decisions [15-17,19,33]. The intervention was intended to be limited to the encounters where an HRQL profile was provided. However, it is possible that the experience with the profiles may influence a physician's practice when seeing patients in the control arms. If such a sensitizing effect had developed, it would be expected that the baseline encounters toward the study's end (once physicians had gained experience with the intervention) would show increased discussion of HRQL issues in comparison with early baseline encounters. Higher frequencies of discussing emotional (57% v 42%), physical (40% v 32%), and role functioning (28% v 19%) were observed in the late sample. The differences were not statistically significant, but some degree of sensitizing physicians, especially to emotional problems, could not be excluded. This effect would have conservative impact on the results, tending to reduce between-group differences. From a broader perspective, such sensitizing effect is not unwanted, and is an interesting starting point for future research. If HRQL summaries can stimulate changes in physicians' communication, they could potentially be used together with communication training programs to improve and maintain physicians' communication skills. Alternative explanations for the results should be considered. Regular completion of the touch-screen questionnaires could train patients and influence their scores. To reduce this effect, a different outcome questionnaire was used (FACT-G) and was completed at patients' homes. A "Hawthorne effect" may occur with patients in groups 1 and 2, showing improvement due to participation in the research. However, all patient encounters, including control group encounters, were continuously recorded, and research assistants approached all study patients at each visit to remind them of the study. The improvement of HRQL over time in all patients is undoubtedly due to attrition of ill patients. An attrition rate of above 30% was observed, which is not dissimilar to other longitudinal studies [16]. The missing data problem was addressed to some extent in the analysis by the mixed-effects model, which assumes that drop-out is not related to intervention. This assumption was checked using logistic regression with drop-out as outcome and study arm was not found to influence attrition (P = .31). Finally, a nonrespondent rate of 30% suggests that this intervention may be unsuitable for some patients (lung cancer, older patients), though patients seemed to decline participation due to the burden of additional data collection for evaluation of the intervention (n = 39). The study was conducted in a specialized cancer center with experience in HRQL research. Further studies are necessary to evaluate the intervention in other disease and treatment settings with a more diverse physician and patient population. In conclusion, our results indicated that routine repeated HRQL assessment in individual patients is a feasible and effective approach for improving medical practice. This simple intervention had a positive impact on physician-patient communication and improved some patients' HRQL and emotional functioning. This approach has a potential for improving clinical practice and deserves further evaluation in multicenter trials and in different cancer patient populations.
The authors indicated no potential conflicts of interest.
We thank all patients and oncologists participating in this study. We are grateful to the nursing and administrative staff members in the clinics who helped run the study. We are grateful to Dr Neil Aaronson for helpful comments on an earlier version of the manuscript.
Supported by grants from Cancer Research UK (formerly Imperial Cancer Research Fund; G.V., A.B.S., L.B., P.L., and P.J.S.), the National Lotteries Charities Board (G.V.), and National Health Service Research and Development (J.M.B., P.M.B.). Presented in part as oral presentations at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31–June 3, 2003, and at the 9th Annual Conference of the International Society for Quality of Life Research, Orlando, FL, October 30–November 2, 2002. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Engel GL: The need for a new medical model: A challenge for biomedicine. Science 196:129–136, 1977 2. Greenfield S, Nelson EC: Recent developments and future issues in the use of health status assessment measures in clinical settings. Med Care 30:23–41, 1992 (suppl 5) 3. McWhinney I: The need for a transformed clinical method, in Stewart M, Roter D (eds): Communicating with Medical Patients. Newbury Park, UK, SAGE Publications, 1989, pp 25–40 4. Holland JC: Cancer's psychological challenges. Sci Am 275:158–161, 1996[Medline] 5. NCCN practice guidelines for the management of psychosocial distress: National Comprehensive Cancer Network. Oncology (Huntingt) 13:113–147, 1999 6. Ganz PA: Quality of life and the patient with cancer: Individual and policy implications. Cancer 74:1445–1452, 1994 (suppl 4)[CrossRef][Medline] 7. Department of Health (UK): The NHS Cancer Plan. London, United Kingdom, Department of Health Publication, 2000 8. Fallowfield L, Ratcliffe D, Jenkins V, et al: Psychiatric morbidity and its recognition by doctors in patients with cancer. Br J Cancer 84:1011–1015, 2001[CrossRef][Medline]
9. Passik SD, Dugan W, McDonald MV, et al: Oncologists' recognition of depression in their patients with cancer. J Clin Oncol 16:1594–1600, 1998 10. Lohr KN. Applications of health status assessment measures in clinical practice: Overview of the third conference on advances in health status assessment. Med Care 30:1–14, 1992 (suppl 5)[CrossRef][Medline] 11. Velikova G, Brown JM, Smith AB, et al: Computer-based quality of life questionnaires may contribute to doctor-patient interactions in oncology. Br J Cancer 86:51–59, 2002[CrossRef][Medline] 12. Wagner AK, Vickrey BG: The routine use of health-related quality of life measures in the care of patients with epilepsy: Rationale and research agenda. Qual Life Res 4:169–177, 1995[CrossRef][Medline] 13. Rubenstein LV, Calkins DR, Young RT, et al: Improving patient function: A randomized trial of functional disability screening. Ann Intern Med 111:836–842, 1989 14. Rubenstein LV, McCoy JM, Cope DW, et al: Improving patient quality of life with feedback to physicians about functional status. J Gen Intern Med 10:607–614, 1995[Medline] 15. Wagner AK, Ehrenberg BL, Tran TA, et al: Patient-based health status measurement in clinical practice: A study of its impact on epilepsy patients' care. Qual Life Res 6:329–341, 1997[CrossRef][Medline]
16. Detmar SB, Muller MJ, Schornagel JH, et al: Health-related quality-of-life assessments and patient-physician communication: A randomized controlled trial. JAMA 288:3027–3034, 2002 17. Taenzer P, Bultz BD, Carlson LE, et al: Impact of computerized quality of life screening on physician behaviour and patient satisfaction in lung cancer outpatients. Psychooncology 9:203–213, 2000[CrossRef][Medline]
18. McLachlan SA, Allenby A, Matthews J, et al: Randomized trial of coordinated psychosocial interventions based on patient self-assessments versus standard care to improve the psychosocial functioning of patients with cancer. J Clin Oncol 19:4117–4125, 2001 19. Espallargues M, Valderas JM, Alonso J: Provision of feedback on perceived health status to health care professionals: A systematic review of its impact. Med Care 38:175–186, 2000[CrossRef][Medline] 20. Greenhalgh J, Meadows K: The effectiveness of the use of patient-based measures of health in routine practice in improving the process and outcomes of patient care: A literature review. J Eval Clin Pract 5:401–416, 1999[CrossRef][Medline] 21. Pouwer F, Snoek FJ, van der Ploeg HM, et al: A comparison of the standard and the computerized versions of the Well-being Questionnaire (WBQ) and the Diabetes Treatment Satisfaction Questionnaire (DTSQ). Qual Life Res 7:33–38, 1998[CrossRef][Medline]
22. Velikova G, Wright EP, Smith AB, et al: Automated collection of quality of life data: A comparison of paper and computer-touchscreen questionnaires. J Clin Oncol 17:998–1007, 1999
23. Cella DF, Tulsky DS, Gray G, et al: The Functional Assessment of Cancer Therapy scale: Development and validation of the general measure. J Clin Oncol 11:570–579, 1993 24. Cella D, Eton DT, Fairclough DL, et al: What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire? Results from Eastern Cooperative Oncology Group (ECOG) Study 5592. J Clin Epidemiol 55:285–295, 2002[CrossRef][Medline] 25. Cohen J: Statistical power analysis for behavioral sciences. Hillsdale, NJ, Laurence Erlbaum Associates, 1988
26. Aaronson NK, Ahmedzai S, Bergman B, et al: The European Organization for Research and Treatment of Cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 85:365–376, 1993 27. Zigmond AS, Snaith RP: The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 67:361–370, 1983[Medline]
28. Velikova G, Wright P, Smith AB, et al: Self-reported quality of life of individual cancer patients: Concordance of results with disease course and medical records. J Clin Oncol 19:2064–2073, 2001 29. Fayers P, Weeden S, Curran D: EORTC QLQ-C30 Reference Values. Brussels, Belgium. EORTC Publication (D/1998/6136/002), 1998 30. Brown H, Prescott R: Applied mixed models in medicine. Chichester, England, John Wiley and Sons Ltd, 1999
31. Guyatt GH, Juniper EF, Walter Sd, et al: Interpreting treatment effects in randomized trials. BMJ 316:690–693, 1998 32. Kazis LE, Callahan LF, Meenan RF, et al: Health status reports in the care of patients with rheumatoid arthritis. J Clin Epidemiol 43:1243–1253, 1990[CrossRef][Medline]
33. Detmar SB, Muller MJ, Schornagel JH, et al: Role of health-related quality of life in palliative chemotherapy treatment decisions. J Clin Oncol 20:1056–1062, 2002 Submitted June 18, 2003; accepted December 5, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
coefficients of 0.48 to 1.00 (median, 0.86). Weekly meetings were conducted to achieve consensus. 
+7), no change (between +6.9 and -6.9), or deterioration (change 
-7.0). Proportions of patients in each category and number of patients needed "to treat" for one patient to get benefit were calculated [
