PURPOSE: Recent studies have shown that symptomatic venous thromboembolism (VTE) frequently develops after hospital discharge following total hip arthroplasty (THA). Risk factors associated with the development of symptomatic VTE, and the efficacy of extended medical prophylaxis to prevent late VTE, are not well defined. METHODS: Using California Medicare discharge data from 1995 through 1997, we identified 298 individuals ≥65 years who were re-hospitalized for symptomatic VTE <91 days after elective unilateral THA. We compared these cases with 599 unmatched controls. Hospital records of the surgical admission were carefully abstracted to obtain information about demographic, surgical and medical variables potentially associated with the development of VTE. Outcomes were confirmed by review of the hospitalization for VTE. RESULTS: Alone or in combination, 70% of control patients were treated with pneumatic compression, 43% with warfarin, 28% with enoxaparin and 6% with unfractionated heparin. Warfarin was given to 29% after discharge. In bivariate analysis, a body mass index (BMI) of ≥25 was strongly associated with rehospitalization for VTE [OR = 2.5, CI 1.8–3.5]. In a risk-adjusted model, independent predictors of VTE included: age over 85 years (vs 65–74) [OR = 2.3, CI 1.2–4.5], female sex [OR = 1.4, CI 1.0–1.9], ambulating before the second post-operative day [OR = 0.7, CI 0.5–0.9], and prior VTE [OR = 3.7, CI 1.8–8.0]. Using patients with a BMI <25 who received no form of prophylaxis (except TED stockings) as the referent group, the only patients who had significantly lower odds of VTE were patients with a BMI <25 who received the following forms of prophylaxis: pneumatic compression alone [OR = 0.4, CI 0.1–1.0], pneumatic compression with enoxaparin [OR = 0.35, CI 0.1–0.9] and pneumatic compression with warfarin [OR = 0.19, CI 0.1–0.5]. Similarly, warfarin use after hospital discharge was associated with absence of VTE, but this was significant only among patients with a BMI <25 [OR = 0.5, CI 0.2–0.8]; for BMI> 25 [OR = 0.7, CI 0.5–1.1]. CONCLUSION: Higher BMI was strongly associated with rehospitalization for symptomatic VTE after total hip arthroplasty. Pneumatic compression and oral anticoagulation after hospital discharge were both associated with significantly lower odds of developing VTE, but only among patients with a BMI <25. Further studies are clearly needed to confirm our findings. In order to further reduce the incidence of symptomatic VTE after THA, efforts should focus on improving the effectiveness of in-hospital as well as extended medical prophylaxis among overweight and obese patients.