Early- and late-term dislocation risk after primary hip arthroplasty in the Medicare population

From 1998 to 2007 Medicare 5% national sample dataset, 39 271 primary total hip arthroplasty (THA) patients were identified. Dislocations within 2 years (early) and after 2 years (late) of primary THA were identified. Cox regression was used to evaluate patient, hospital, and procedure characteristics for risk of dislocation, and 1540 (3.92%) and 451 (1.15%) patients were diagnosed with early and late dislocations, respectively. Dislocation rate at 6 months' follow-up decreased steadily between 1998 and 2007 from 4.21% to 2.14%. Early and late dislocation risks were lower by 35% (P < .001) and 43% (P = .01), respectively, for patients operated during 2004 and 2007 compared with 1998 and 2003. Higher Charlson index scores (i.e., more comorbid conditions) and surgeon volume were significant risk factors (P < or = .04). Decrease in dislocation risks after primary THA seems to coincide with increasing use of larger diameter femoral heads. An awareness of risk factors for dislocation can help surgeons identify high-risk patients so as to prescribe appropriate intervention strategies.