Determinants of the use of complementary therapies by patients with cancer

Purpose: We performed a survey of Israeli oncology patients to examine the extent of their use of complementary therapies (CT) and to compare sociodem ographic, psychologic, and medical characteristics, attitudes, and quality of life of users and nonusers of CT. Patients and Methods: Questionnaires were administered to 1,027 patients at tending ambulatory and inpatient hematology or oncology facilities at three hospitals. Medical information was extracted from charts. Univariate and m ultivariate comparisons of users and nonusers of CT were performed. Results: A total of 526 participants (51.2%) had used CT since their diagno sis, and 357 patients (34.9%) had used CT recently (in the past 3 months). Factors that multivariate analysis found to be significantly associated (P < .05) with recent CT use were as follows: female sex; age 35 to 59 years; more education; coming to the hospital by private car; advanced disease sta tus; having a close friend or a relative with cancer; and attending support groups or individual counseling. After controlling for these factors, indi vidually examined psychosocial variables associated with recent CT use incl uded the following (odds ratios [OR] with 95% confidence intervals [CI]): n eeds unmet by conventional medicine (OR, 2.76; 95% CI, 1.95 to 3.89); helpl essness (OR, 1.39; 95% CI, 1.0 to 1.91); incomplete trust in the doctor (OR , 1.49; 95% CI, 1.08 to 2.06); and changed outlook or beliefs since the dia gnosis of cancer (OR, 1.47; 95% CI, 1.07 to 2.02). Functional quality of li fe (including physical, emotional, social, and role function) and symptom ( fatigue and diarrhea) scores were significantly worse for recent CT users c ompared with nonusers, controlling for age, sex, and current disease status . Conclusion: Characteristics associated with CT use include age, sex, educat ion, and advanced disease. Significant associations between CT use and atte nding supportive psychotherapy, unmet needs, helplessness, and worse emotio nal and social function indicate considerable distress, suggesting that inc reased attention to psychosocial needs within oncologic settings is warrant ed. J Clin Oncol 19:2439-2448. (C) 2001 by American Society of Clinical Onc ology.