|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Health Status and Quality of Life in Patients With Early-Stage Hodgkin’s Disease Treated on Southwest Oncology Group Study 9133
From the University of California, Los Angeles, Los Angeles, CA; Dana-Farber Cancer Institute, Boston, MA; Loyola University Stritch School of Medicine, Maywood, IL; Ohio State University Health Center, and Columbus Community Clinical Oncology Program, Columbus, OH; University of Michigan Medical Center, Ann Arbor, MI; Southwest Oncology Group Statistical Center, and Puget Sound Oncology Consortium, Seattle, WA; and University of Rochester School of Medicine, Rochester, NY. Address reprint requests to Southwest Oncology Group (SWOG-9208), Operations Office, 14980 Omicron Dr, San Antonio, TX 78245-3217.
Purpose: We describe the short and intermediate-term quality-of-life (QOL) outcomes in patients treated on a randomized clinical trial in early-stage Hodgkin’s disease (Southwest Oncology Group [SWOG] 9133) comparing subtotal lymphoid irradiation (STLI) with combined-modality treatment (CMT). Patients and Methods: Two hundred forty-seven patients participated in the QOL study (SWOG 9208), completing several standardized instruments (Symptom Distress Scale; Cancer Rehabilitation Evaluation System – Short Form; Medical Outcomes Study 36-Item Short-Form Health Survey Vitality Scale; and a health perception item), as well as questions about work, marital status, and concerns about having children. This article reports on results from baseline before random assignment, at 6 months, and at 1 and 2 years after random assignment. Results: Patients receiving CMT experienced significantly greater symptom distress (P < .0001), fatigue (P = .001), and poorer QOL (P = .015) at 6 months than the STLI patients, reflecting a shorter time since completion of therapy in the CMT arm. Importantly, patients in the two groups did not differ on any outcomes at the 1-and 2-year assessments. Both patient groups reported significantly more fatigue before treatment than healthy reference populations, and fatigue did not improve in either group after treatment. Conclusion: This study demonstrated that patients with early-stage Hodgkin’s disease experience a short-term decrease in QOL and an increase in symptoms and fatigue with treatment, which is more severe with CMT; by 1 year, however, CMT and STLI patients report similar outcomes. Fatigue scores for both arms were lower at baseline than scores for the general population and did not return to normal levels 2 years after random assignment. The mechanisms responsible for this lingering problem warrant further investigation.
DRAMATIC CHANGES in the treatment of Hodgkin’s disease (HD) during the past three decades have transformed a uniformly fatal disease into one that is highly curable.1–4 Patients with HD can expect long-term survival and cure. As a result, there has been substantial interest in refining treatment strategies to offer the best chance of cure while minimizing toxicity. In the early 1990s, there was considerable debate about the necessity for staging laparotomy in early-stage (clinical stage IA and IIA) and favorable-histology HD. This debate was driven by the morbidity of the procedure and the high likelihood of cure with primary radiotherapy, using salvage chemotherapy only at relapse. In addition, there was increasing interest in using short courses of chemotherapy with more limited radiotherapy to maximize cure and minimize toxicity.5 The Southwest Oncology Group (SWOG) designed a treatment protocol to investigate alternative strategies for the management of early-stage HD (SWOG 9133), the early results of which have been published elsewhere.3 SWOG 9133 was a phase III randomized intergroup trial of subtotal lymphoid irradiation (STLI) versus three cycles of doxorubicin and vinblastine followed by STLI (combined-modality therapy [CMT]) in patients with stage IA and II A HD. We describe here the first results from a companion quality-of-life (QOL) study, SWOG 9208, conducted in conjunction with the treatment trial. QOL after treatment for HD has been described in several cross-sectional studies during the past two decades.6–10 Fobair et al9 reported that ongoing fatigue was a major concern for 37% of 403 survivors, which was influenced by age, time since diagnosis, stage of disease, and type of treatment (younger age, longer time since diagnosis, earlier stage, and radiation therapy without chemotherapy were all significantly better). Fatigued survivors also reported higher rates of depression. Other concerns identified were marital disruption, problems with infertility, and less interest in sexual activity. In addition, 29% of this sample was unemployed, with 18% currently looking for work. Other cross-sectional studies examined HD survivors either treated on clinical trials or from large treatment centers,11–14 largely confirming the findings of Fobair et al,9 but also noting that HD survivors performed more poorly on measures of physical and psychosocial function in comparison with either patients with acute leukemia, those with testicular cancer, or healthy population samples. These studies suggested a relationship between outcomes and the intensity of treatments; however, their retrospective and uncontrolled design limits the ability to determine causality. These pioneering studies provided important background for design of the SWOG 9133 QOL study, delineating a number of specific hypotheses and research questions related to acute treatment and long-term survivorship. The objectives of SWOG 9208 were (1) to evaluate prospectively the health status and QOL of early-stage HD patients receiving either STLI or CMT; (2) to describe the short-term effects of the treatments on symptoms and QOL; and (3) to evaluate the intermediate and long-term effects of the two treatments on QOL. This report examines the short-term and intermediate outcomes during the first 2 years after random assignment.
SWOG 9133 Treatment Summary SWOG 9133 was a randomized trial for clinically staged patients with stage IA, IEA, IIA and IIEA HD. Patients with unfavorable presentations (eg, infradiaphragmatic disease, large mediastinal masses, or "B" symptoms) were excluded.3 The standard therapy in this trial was STLI, and consisted of sequential mantle and periaortic/spleen fields, to a dose of 36 to 40 Gy for 4 weeks each (1.8 or 2 Gy administered in 20 fractions), using megavoltage radiotherapy in the 4- to 10-MeV range. Patients randomly assigned to the CMT arm were initially treated with doxorubicin (25 mg/m2 intravenously) and vinblastine (6 mg/m2 intravenously) on days 1 and 15 of each 28-day course, for three cycles. At the completion of the third cycle of chemotherapy, staging studies were repeated, and a period of 6 weeks after the last doses of doxorubicin and vinblastine was allowed to elapse before the initiation of STLI (36 to 40 Gy). The primary treatment report contains the full details.3
Eligibility for the QOL Study
Instruments and Assessment Strategy The primary measure of intermediate- and long-term QOL was The Cancer Rehabilitation Evaluation System – Short Form (CARES-SF). CARES-SF23,24 and its parent longer instrument, the CARES,25,26 are validated, multidimensional, cancer-specific measures of QOL, with excellent psychometric properties.26,27 The long form of the CARES has been used in cross-sectional and longitudinal studies of breast cancer patients and is sensitive to change over time.28–30 The CARES-SF provides a global QOL score and five summary scales (physical, psychosocial, marital, medical interaction, sexual).23 Three additional items from the sexual interest and sexual functioning subscales of the CARES were included for more detailed assessment. Demographic information was collected with a brief questionnaire that asked about education, ethnicity, marital status, current employment status, household income, and type of health care insurance. Reproductive history, fertility and plans to have children in the future, date of last menstrual period (women only), and vasectomy and sperm banking (men only) were also queried. The assessment strategy for this trial focused on comparison of major time points when HD survivors would like information about their anticipated function, after receiving one treatment or the other. It was assumed that treatment toxicity could be captured through standard clinical trial reporting; therefore, we did not collect patient reported data while on treatment. Rather, we assessed symptoms and QOL at times in the course of disease and recovery that made clinical sense for patients and health care providers. Thus, we scheduled one assessment after the completion of treatment at a regularly scheduled visit (6 months), and then annually thereafter. We did not specifically plan to assess QOL at the time of relapse, but assumed that if patients required 6 to 9 months of salvage chemotherapy at the time of relapse, it would likely be captured as an effect on symptoms and QOL (including ability to work) at the annual follow-up assessment. For this report, we present data from the first 2 years of the study, and analyses related to relapse will be addressed in subsequent articles when there has been sufficient long-term follow-up of the study sample.
Administration Procedures
Statistical Considerations and Analytic Plan There is no formal statistical test for whether or not the MAR missing data mechanism holds. Therefore, we assessed sensitivity of results to this assumption by performing the same analyses assuming an alternative informative or nonignorable missing data mechanism, which assumes that the probability of missing data depends on factors other than the observed QOL measures, such as unobserved deteriorating health. We fit pattern-mixture models34,35 with four strata corresponding to patients who dropped out at the baseline, 6-month, 1-year, and 2-year time points, respectively, and separate Normal linear mixed models within each strata. To identify parameters in strata where patients dropped out the earliest, we set the slope equal to zero and carried the last mean value forward. The conclusions were the same as for the MAR analyses and results are reported only for the SDS outcome measure. For tests of proportions, such as the percent of patients with SDS scores greater than 25, the Fisher’s exact test was used. For tests of treatment differences in all subscales, such as the CARES-SF physical subscale, the Wilcoxon rank sum test was used. For tests of binary outcomes, such as compliance, logistic regression analysis was used.
The treatment study was expected to randomly assign approximately 420 patients, with accrual estimated at 60 patients per year. The QOL study started 1 year later than the therapeutic trial. With the later start of the QOL study, we expected to have 288 patients available for intermediate end point comparisons at 1 year (420 - 60 = 360 total patients, minus 72 to account for a 20% rate of incomplete questionnaires). In addition, there were expected to be approximately 200 patients available for medical end point comparisons at 5 years. Assuming 20% attrition related to questionnaire completion over the study period, we expected to have 160 patients available for the long-term QOL analysis at 5 years. Power was estimated to be greater than 0.80 for detecting differences more than 0.29 in the proportion of psychosocial morbidity between the treatment arms at 5 years as measured by the CARES-SF psychosocial scale cutoff score of End points for the QOL study focused on short-term toxicities as well as intermediate and long-term effects. For this protocol, short-term was the period from baseline (before randomization) through the 12-month assessment. The SDS was the primary short-term outcome, with the MOS SF-36 Vitality Scale as a secondary end point. We hypothesized that patients receiving the CMT would report more symptoms and greater fatigue than patients receiving STLI alone, with treatment arm differences most prominent at the 6-month assessment, and possibly continuing up to 1 year after random assignment. Further, we hypothesized that patients receiving CMT would have poorer health perceptions than patients receiving STLI alone. Intermediate effects of treatment are measured at the 2-year assessment, with CARES-SF providing the primary outcome measure.
Patient Characteristics SWOG 9133 was closed early at the recommendation of the Data Safety Monitoring Committee on April 15, 2000, after 348 patients (80% of planned accrual) had been entered and randomly assigned, due to a markedly superior 3-year estimate of failure-free survival for patients with CMT (94%) compared with those treated with STLI (81%).3 For this reason, the sample for the QOL study was somewhat reduced from the planned target of 288 for intermediate end point assignment. Patient characteristics by treatment are shown in Table 1  . Comparison of the QOL study sample with the final SWOG 9133 sample demonstrated slightly more white patients on the STLI arm in the QOL study (94% v 86%, respectively; P = .10).
Compliance With Patient-Reported Data Table 2 presents acceptable submission rates for CARES-SF and SDS baseline and follow-up assessments, with only one submission rate less than 70%. Logistic regression was used to see if the following variables were related to submission of CARES-SF and SDS forms at 2 years: age, sex, relapse, stage of disease, health perception at baseline, marital status, number of children, and employment status. None of these variables was related to drop-out, with P values ranging from .08 to .87 for CARES-SF, and .14 to .88 for the SDS. Submission rates were slightly higher for the SDS. The computer monitoring system for the trial initially gave institutions credit when only the SDS form was submitted; this was later changed to require both forms.
QOL forms were not submitted for a variety of reasons. The following reasons were given, summarized across the 6-, 12-, and 24-month assessment points: patient refused but not for health reasons (25.2%); patient refused a telephone interview after missing a clinic appointment (1.5%); institution staff forgot to administer the form (20%); the patient could not be contacted (2.9%); the patient went off treatment early and could not be contacted for QOL follow-up (2.2%); other reasons (13.3%); death (1.5%); reason not provided (33.3%). Patient refusals (15% to 32%) and no reason provided (24% to 38%) increased considerably over the three follow-up assessments.
Short-Term Effects of Treatment
As confirmed by the model that adjusted for baseline effects, there was a significant increase in SDS scores from baseline to 6 months for both treatment arms (P = .001 for STLI; P < .0001 for CMT). The change was much more dramatic on the CMT arm, with a 28.4% increase on the CMT arm compared with an 8.4% increase on the STLI arm (P < .0001). As a result, at 6 months, SDS scores are significantly higher (ie, worse) on the CMT arm compared with the STLI arm (P < .0001). SDS scores on both arms return to baseline levels at 1 year and remain the same until 2 years (Fig 1  ). Under the alternative pattern-mixture model, the difference between the treatments in SDS scores at the 6-month time point and in the change in SDS scores at 6 months were also significant, with P < .0001.
Patients on the CMT arm also experienced more fatigue at 6 months compared with the STLI arm (Fig 2
Intermediate-Term Effects of Treatment Figure 3 presents global CARES-SF scores for the two treatment arms, showing a trend that is similar to the other instruments. None of the baseline variables (age, sex, or stage) had a significant effect on CARES-SF scores, so they were excluded from further analyses. There was a significant increase (worsening) in CARES-SF scores from baseline to 6 months within the CMT arm (75.5%; P = .004), but not in the STLI arm. At 6 months, CARES-SF scores were significantly higher (worse) in the CMT arm than the STLI arm (P = .015). Scores returned to baseline levels for both arms by year 1. Intermediate-term effects by treatment arm were not observed at year 2 for the total CARES-SF score.
In exploratory evaluations of the CARES-SF Summary Scales for Physical, Psychosocial, Sexual, Marital and Medical Interaction, there were no significant differences for any of the comparisons at the 1- and 2-year assessments. For the 6-month assessment, there was no significant difference between the two treatments for the Marital or Medical Interaction Summary scales. The Psychosocial Summary Scale showed a marginally significant difference favoring the STLI arm; the test was adjusted for multiple comparisons (P = .01). Regarding the CARES-SF psychosocial cutoff score used for power calculations, we found that, as with other QOL outcomes, only at 6 months did the two arms differ in the proportion of patients at or above a score of 0.615; a significantly larger proportion of patients in the combined arm reported scores at or above this cutoff point (62.4%) compared with those in the STLI arm (43.6%). At 6 months, the Physical and Sexual Summary Scales were significantly worse in the CMT arm (Physical, P < .0001; Sexual, P = .006; data not shown). To further understand the differential impact of CMT on sexual health, we examined scores from the Sexual Interest and Sexual Functioning subscales of the Sexual Summary Scale. At 6 months, an increase in problems with sexual interest seemed to be the primary contributor to the poorer Sexual Summary scores in the CMT group compared with STLI group (P = .0009), while patients in both groups reported increased sexual dysfunction, which was slightly worse in the CMT group (P = .044; data not shown).
Figure 4
Impact on Social and Demographic Outcomes We examined changes in the following over time: marital or partner status (single, married, or divorced), plans to have children (no, yes, or uncertain), concern for having/fathering a child in the future (very concerned, concerned, only somewhat concerned, or not at all concerned), and employment status (working, retired, on leave, unemployed, disabled, student, or other). Figure 5 (A-D) shows, by treatment arm, the percentage of patients married, not planning to have children, concerned or very concerned about having children, and working, respectively, over the 2-year assessment period. We found no significant differences between treatment arms except for employment (Fig 5D), for which the percentage of patients working at baseline was marginally lower on the CMT arm (P = .08) and was significantly lower at 6 months (P = .003). By year 1, there were no significant differences in working status between the two arms.
This article describes short- and intermediate-term QOL outcome data for HD patients participating in SWOG 9208, a companion study to SWOG 9133. We believe that this is one of the first prospective QOL studies in patients with HD, and as such, it provides important new outcome data on this patient population. Consistent with our hypotheses, at 6 months, patients receiving CMT experienced greater severity of symptoms, more fatigue, and poorer QOL compared with patients receiving STLI alone. The Physical and Sexual Summary Scales of CARES-SF were the most significantly affected dimensions of QOL in the CMT arm, probably reflecting the significant increase in symptoms and fatigue reported by these patients. However, these results must be interpreted in light of the more protracted treatment course in the CMT group, with completion of therapy very close to the 6-month assessment, providing less time for the possibility of recovery, in comparison with the STLI group. Nevertheless, these results provide information to patients who may want to know "How will I be feeling 6 months from now?" Because the timing of the assessment differed in length of time since the end of treatment, we cannot conclude that the CMT is more toxic, but we can say that patients who receive CMT versus STLI can expect to still be experiencing more symptoms and poorer QOL than those with radiation alone 6 months after the initiation of treatment for HD. Importantly, these differences are short-lived, and by the 1-year assessment, there were no significant differences in symptoms or QOL, which was contrary to our expectations. For the three main outcomes in this study (symptoms, fatigue, QOL), the scores at 1 year were similar to baseline scores before treatment, without further improvement at the 2-year assessment. Patient perception of general health paralleled the findings for the other QOL measures, with the CMT arm doing worse in the short-term. It is noteworthy that patients in neither group returned to their baseline health perception score, suggesting a persistent decrement in perceived health associated with the diagnosis and treatment of HD. Another important finding from this study was the increased fatigue level for both study groups (CMT = 45.9 and STLI = 49.7) at baseline. These scores are lower than scores for general population samples on this measure.22 For example, for healthy men aged 35 to 44 years, the mean Vitality Scale score is 65.5, and for women aged 35 to 44 years, it is 59.4. The standard deviation for this scale is about 18.5. Thus, before any treatment, these early-stage HD patients report scores that are about a half SD below normal and more consistent with scores from older patients with ischemic heart disease.22 While fatigue is a known symptom of HD, we did not expect it to be so prominent in patients with favorable prognosis disease without B symptoms. Furthermore, if this were a manifestation of the disease, then it would be expected to improve subsequent to treatment and induction of remission. For these patients, the Vitality Scale scores at 1 and 2 years are slightly below the baseline score, and are substantially lower than comparison data from a breast cancer survivor sample after adjuvant treatment and radiotherapy.37 Our observations regarding fatigue are consistent with the larger literature on HD from cross-sectional studies.9–12 In those studies, excess fatigue was often attributed to the late effects of treatment. However, our prospective data suggest that increased fatigue is a substantial problem at diagnosis and that it does not remit with effective treatment. Baseline fatigue could be disease-related, while that experienced subsequently could be a result of treatment. However, fatigue in a large sample of breast cancer survivors was not related to treatment intensity,37 nor is this the case for comparison of the treatment arms in this study of HD. These observations suggest a need for further specific research on the etiology of fatigue in HD survivors, since it may be multifactorial, and could relate to ongoing disturbance in proinflammatory cytokines that may be part of the underlying disease process or a result of treatments.38 We plan to explore the relationship between fatigue, patient characteristics, symptoms, and QOL in this patient sample in a future report. In conclusion, we found that patients with early-stage HD in both treatment groups experienced a short-term increase in symptoms, fatigue, and poorer QOL as a result of treatment, which was more severe in the CMT group at 6 months after diagnosis due to more prolonged treatment; however, by 1 year after random assignment, outcomes in the two treatment groups were indistinguishable. In this study, which may be the first prospective clinical trial in HD to collect QOL data with an instrument that has population norms,39 we have identified increased fatigue in favorable HD patients at diagnosis that persists after successful, curative treatment. Fatigue is sustained out to 2 years of follow-up and is a long-term problem for these survivors. Further research is needed to understand the etiology of this important complication of HD.
The authors indicated no potential conflicts of interest.
We thank the patients and research staff at SWOG and CALGB institutions for making this study possible. Special thanks to Enid Zuckerman for coordinating the collection of QOL data in CALGB institutions, and Laura Abraham and Amber Paklit for providing reminder telephone calls for the QOL assessments.
This investigation was supported in part by the following Public Health Service Cooperative Agreement grant numbers awarded by the National Cancer Institute, Department of Health and Human Services, Washington, DC: CA38926, CA32102, CA46282, CA04920, CA65261, CA32291, CA27057, CA46368, CA22433, CA45377, CA20319, CA13612, CA35176, CA58415, CA76447, CA45807, CA58723, CA12644, CA63845, CA46113, CA76132, CA58861, CA35192, CA12213, CA35281, CA52654, CA35128, CA76429, CA04919, CA42777, CA63844, CA58416, CA58658, CA16385, CA67575, CA35996, CA14028, CA37981, CA35090, CA35262, CA58348, CA28862, CA76462, CA35119, CA35431. Dr Ganz is also supported by a Clinical Research Professorship from the American Cancer Society, Atlanta, GA.
1. Rosenberg SA, Kaplan HS: The evolution and summary results of the Stanford randomized clinical trials of the management of Hodgkin’s disease: 1962–1984. Int J Radiat Oncol Biol Phys 11:5–22, 1985[Medline] 2. Mendenhall NP: The role of radiation in the management of Hodgkin’s disease: An update. Cancer Invest 17:47–55, 1999[Medline]
3. Press OW, LeBlanc M, Lichter AS, et al: Phase III randomized intergroup trial of subtotal lymphoid irradiation versus doxorubicin, vinblastine, and subtotal lymphoid irradiation for stage IA to IIA Hodgkin’s disease. J Clin Oncol 19:4238–4244, 2001
4. Prosnitz LR: Reducing treatment-related morbidity and mortality in early-stage Hodgkin’s disease and why the recent Southwest Oncology Group Trial is not the way to go. J Clin Oncol 20:2225–2228, 2002
5. Horning SJ, Hoppe RT, Mason J, et al: Stanford-Kaiser Permanente G1 study for clinical stage I to IIA Hodgkin’s disease: Subtotal lymphoid irradiation versus vinblastine, methotrexate, and bleomycin chemotherapy and regional irradiation. J Clin Oncol 15:1736–1744, 1997 6. Kornblith AB, Anderson J, Cella DF, et al: Hodgkin disease survivors at increased risk for problems in psychosocial adaptation: The Cancer and Leukemia Group B. Cancer 70:2214–2224, 1992[CrossRef][Medline] 7. Kornblith AB, Anderson J, Cella DF, et al: Comparison of psychosocial adaptation and sexual function of survivors of advanced Hodgkin disease treated by MOPP, ABVD, or MOPP alternating with ABVD. Cancer 70:2508–2516, 1992[CrossRef][Medline]
8. Kornblith AB, Herndon JE, Zuckerman E, et al: Comparison of psychosocial adaptation of advanced stage Hodgkin’s disease and acute leukemia survivors: Cancer and Leukemia Group B. Ann Oncol 9:297–306, 1998
9. Fobair P, Hoppe RT, Bloom J, et al: Psychosocial problems among survivors of Hodgkin’s disease. J Clin Oncol 4:805–814, 1986
10. Bloom JR, Fobair P, Gritz E, et al: Psychosocial outcomes of cancer: A comparative analysis of Hodgkin’s disease and testicular cancer. J Clin Oncol 11:979–988, 1993
11. Knobel H, Havard LJ, Brit LM, et al: Late medical complications and fatigue in Hodgkin’s disease survivors. J Clin Oncol 19:3226–3233, 2001
12. Loge JH, Abrahamsen AF, Ekeberg O, et al: Hodgkin’s disease survivors more fatigued than the general population. J Clin Oncol 17:253–261, 1999 13. Loge JH, Abrahamsen AF, Ekeberg, et al: Fatigue and psychiatric morbidity among Hodgkin’s disease survivors. J Pain Symptom Manage 19:91–99, 2000[CrossRef][Medline] 14. Greil R, Holzner B, Kemmler G, et al: Retrospective assessment of quality of life and treatment outcome in patients with Hodgkin’s disease from 1969 to 1994. Eur J Cancer 35:698–706, 1999[CrossRef][Medline] 15. McCorkle R, Quint-Benoliel J: Symptom distress, current concerns and mood disturbance after diagnosis of life-threatening disease. Soc Sci Med 17:431–438, 1983[CrossRef][Medline] 16. McCorkle R, Benoliel JQ, Donaldson G, Georgiadou F, Moinpour C, Goodell B: A randomized clinical trial of home nursing care for lung cancer patients. Cancer 64:1375–1382, 1989[CrossRef][Medline] 17. McCorkle R, Cooley ME, Shea JA: A User’s Manual for the Symptom Distress Scale. Philadelphia, PA, University of Pennsylvania School of Nursing, 1998 18. Ware JE Jr, Sherbourne CD: The MOS 36-item short-form health survey (SF-36), I: Conceptual framework and item selection. Med Care 30:473–483, 1992[Medline] 19. McHorney CA, Ware JE Jr, Raczek AE: The MOS 36-Item Short-Form Health Survey (SF-36), II: Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 31:247–263, 1993[Medline] 20. McHorney CA, Ware JE Jr, Lu JF, et al: The MOS 36-item Short-Form Health Survey (SF-36), III: Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 32:40–66, 1994[Medline] 21. Ware JE Jr: Monitoring and evaluating health services. Med Care 23: 705–709, 1985[CrossRef][Medline] 22. Ware JE Jr: SF-36 Health Survey: Manual and Interpretation Guide. Boston, MA, The Health Institute, New England Medical Center, 1993 23. Schag CA, Ganz PA, Heinrich RL: CAncer Rehabilitation Evaluation System-short form (CARES-SF): A cancer specific rehabilitation and quality of life instrument. Cancer 68:1406–1413, 1991[CrossRef][Medline] 24. Ganz PA, Schag CA, Lee JJ, et al: The CARES: A generic measure of health-related quality of life for patients with cancer. Qual Life Res 1:19–29, 1992[CrossRef][Medline] 25. Schag CC, Heinrich RL, Ganz PA: Cancer Inventory of Problem Situations: An Instrument for Assessing Cancer Patients’ Rehabilitation Needs. J Psychosoc Oncol 1:11–24, 1983[Medline] 26. Schag CA, Heinrich RL: Development of a comprehensive quality of life measurement tool: CARES. Oncology (Huntingt) 4:135–138, 1990 27. Schag CA, Heinrich RL, Aadland RL, et al: Assessing problems of cancer patients: Psychometric properties of the cancer inventory of problem situations. Health Psychol 9:83–102, 1990[CrossRef][Medline] 28. Ganz PA, Schag AC, Lee JJ, et al: Breast conservation versus mastectomy: Is there a difference in psychological adjustment or quality of life in the year after surgery? Cancer 69:1729–1738, 1992[CrossRef][Medline] 29. Schag CA, Ganz PA, Polinsky ML, et al: Characteristics of women at risk for psychosocial distress in the year after breast cancer. J Clin Oncol 11:783–793, 1993[Abstract] 30. Ganz PA, Rowland JH, Desmond K, et al: Life after breast cancer: Understanding women’s health-related quality of life and sexual functioning. J Clin Oncol 16:501–514, 1998[Abstract] 31. Rubin DB: Formalizing subjective notions about the effect of nonrespondents in sample surveys. J Am Stat Assoc 72:538–543, 1977[CrossRef] 32. Laird NM, Ware JH: Random-effects models for longitudinal data. Biometrics 38:963–974, 1982[CrossRef][Medline] 33. SAS Institute I: SAS/STAT User’s Guide, Version 8.0. Cary, NC, SAS Institute Inc, 1999
34. Little RJA: A class of pattern-mixture models for normal and incomplete data. Biometrika 81:471–483, 1994 35. Fairclough DL: Design and analysis of quality of life studies in clinical trials: Interdisciplinary. Boca Raton, FL, Chapman & Hall/CRC Press, 2002 36. Ganz PA, Hirji K, Sim MS, et al: Predicting psychosocial risk in patients with breast cancer. Med Care 31:419–431, 1993[Medline]
37. Bower JE, Ganz PA, Desmond KA, et al: Fatigue in breast cancer survivors: Occurrence, correlates, and impact on quality of life. J Clin Oncol 18:743–753, 2000
38. Bower JE, Ganz PA, Aziz N, et al: Fatigue and proinflammatory cytokine activity in breast cancer survivors. Psychosom Med 64:604–611, 2002 39. Kaasa S, Knobel H, Loge JH, et al: Hodgkin’s disease: Quality of life in future trials. Ann Oncol 9:137–145, 1998 (suppl 5) Submitted January 7, 2003; accepted June 17, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
25 reflects moderate to severe symptom problems.
level at .007 to account for multiple comparisons across the follow-up period, accounting for drop-out. Our sample size of 165 patients at 2 years slightly exceeds the n = 160 required for the power calculations presented here. 
