Background and purpose — Mortality and revision risks are important issues during shared decision-making for total hip arthroplasty (THA) especially in elderly patients. We examined mortality and revision rates as well as associated patient and prosthesis factors in primary THA for osteoarthritis (OA) in patients ≥ 80 years in the Netherlands. Patients and methods — We included all primary THAs for OA in patients ≥ 80 years in the period 2007–2019. Patient mortality and prosthesis revision rates were calculated using Kaplan-Meier survival analyses. Risk factors for patient mortality and prosthesis revision were analyzed using multivariable Cox regression analysis adjusted for age, sex, ASA class, fixation method, head size, and approach. Results — Mortality was 0.2% at 7 days, 0.4% at 30 days, 2.7% at 1 year, and 20% at 5 years. Mortality was higher in males and higher ASA class, but did not differ between fixation methods. The 1-year revision rate was 1.6% (95% CI 1.5–1.7) and 2.6% (CI 2.5–2.7) after 5 years. Multivariable Cox regression analysis showed a higher risk of revision for uncemented (hazard ratio [HR] 1.6; CI 1.4–1.8) and reverse hybrid THAs (HR 2.9; CI 2.1–3.8) compared with cemented THAs. Periprosthetic fracture was the most frequently registered reason for revision in uncemented THAs. Interpretation — Mortality is comparable but revision rate is higher after uncemented compared with cemented THA in patients 80 years and older, indicating that cemented THA might be a safer option in this patient group.