BACKGROUND: The treatment of osteoporosis among postmenopausal women represents a major public health challenge because long-term therapy is needed to prevent fractures and chronic disability. Low patient compliance with prescribed osteoporosis treatments can severely distort the validity of controlled clinical trials. Raloxifene and alendronate have been shown to reduce the incidence of osteoporotic fracture in postmenopausal women in well-conducted randomized trials, but few data are available on the rate of adherence to these treatments in routine clinical practice. OBJECTIVE: The primary aim of this study was to assess the compliance of postmenopausal women at risk for osteoporotic fractures who were treated with raloxifene hydrochloride (RLX) versus alendronate sodium (ALN) during a 12-month observational period in a routine clinical setting. Secondary objectives were the assessment of factors that might contribute to noncompliance and patient satisfaction. METHODS: This open-label, prospective, multicenter, nonrandomized, observational, comparative study was conducted at 154 centers across Spain. Assignment to either RLX or ALN treatment was determined by the physician and was based on each patient's clinical profile. Compliance with RLX (60-mg tablet once daily) versus ALN (10-mg tablet once daily) was assessed using 3 different compliance assessment tools: the Morisky-Green test, the Autocompliance test, and the Compliance Questionnaire. A logistic regression model was used to assess different factors affecting compliance. Patient satisfaction was also assessed using a questionnaire. Adverse events (AEs) were collected as reasons for discontinuation in the Compliance Questionnaire and at the discontinuation visit. RESULTS: A total of 902 women (RLX group, n = 476; ALN group, n = 426) were included in the study (mean age, 64.4 [6.9] years). Overall, patients in the RLX group reported significantly better compliance than patients in the ALN group, as collected either by the Morisky-Green test (68.7% vs 54.0%; P < 0.001) or the Autocompliance test (94.7% vs 90.6%; P = 0.033). More patients discontinued treatment prematurely in the ALN group compared with the RLX group (25.8% vs 16.4%; P < 0.001). The age-adjusted relative risk for discontinuation was 1.4-fold higher for women treated with ALN than for those treated with RLX (95% CI, 1.21-1.61). The main reason for premature discontinuation was due to AEs (RLX 4.8% vs ALN 11.0%; P < 0.001). The proportion of patients with gastrointestinal AEs was 9.9% in the ALN group and 3.4% in the RLX group (P < 0.001). Only treatment and type of physician were independent covariates of treatment compliance. After 12 months of observation, significantly more patient in the RLX group were satisfied or very satisfied with their treatment than patients in the ALN group (P < 0.001). CONCLUSION: In this study of postmenopausal women at risk for osteoporotic fractures, compliance with 12-month treatment with daily RLX was higher than with daily ALN in clinical setting. RLX showed significant benefits compared with ALN in terms of compliance assessed by means of the Morisky-Green and Autocompliance tests and the patients' self-reported satisfaction.