To describe variability in the prescription of biologics (B-DMARDs) for patients with rheumatoid arthritis (RA) in hospitals in Spain, and to explore which characteristics of the patient, the doctor and the hospital are associated with this variability. Cross-sectional multicentric study in 46 rheumatology services of the National Health System. Medical records of 1188 randomly selected patients were reviewed. The association of each variable with B-DMARD prescription was analyzed using simple logistic regressions. Multilevel logistic regression models were created to analyze variability among centers. 36.8% of patients had received B-DMARD. The proportion of patients being treated with B-DMARDs varied between 3.6 and 71.4% depending on the center. Association of prescription of B-DMARD with patient age (OR = 0.958, 95% CI = 0.947-0.968, p < 0.001), longer disease duration (OR = 1.05, 95% CI = 1.032-1.069, p < 0.001), higher CRP levels (OR = 1.022, 95% CI = 1.003-1.042, p = 0.023), and higher number of hospitalizations (OR = 1.286, 95% CI = 1.145-1.446, p < 0.001) was observed. With regard to the center characteristics, the existence of telephone consultations (OR = 1.438, 95% CI = 1.037-1.994, p = 0.03) and the number of beds (OR = 1.045, 95% CI = 1.001-1.091, p = 0.044) were positively associated with prescription of B-DMARDs. Patient variables explained 34.04% of the variability among centers. By adjusting for patient and hospital characteristics, it went up to 83.71%. There is variability in the prescription of B-DMARDs for patients with RA among hospitals which is associated, to a greater extent, with the center characteristics. B-DMARDs prescription could be partly explained by other factors not covered by the current study including the provider's attitudes towards biologics and other hospital characteristics.