Background and purpose — Femoral neck fractures are commonly treated with cemented or uncemented hemiarthroplasties (HA). We evaluated differences in mortality and revision rates in this fragile patient group. Patients and methods — From January 1, 2007 until December 31, 2016, 22,356 HA procedures from the Dutch Arthroplasty Register (LROI) were included. For each HA, follow-up until death, revision, or end of follow-up (December 31, 2016) was determined. The crude revision rate was determined by competing risk analysis. Multivariable Cox regression analyses were performed to evaluate the effect of fixation method (cemented vs. uncemented) on death or revision. Age, sex, BMI, Orthopaedic Data Evaluation Panel (ODEP) rating, ASA grade, surgical approach, and previous surgery were included as potential confounders. Results — 1-year mortality rates did not differ between cemented and uncemented HA. 9-year mortality rates were 53% (95% CI 52–54) in cemented HA compared to 56% (CI 54–58) in uncemented HA. Multivariable Cox regression analysis showed similar mortality between cemented and uncemented HA (HR 1.0, CI 0.96–1.1). A statistically significantly lower 9-year revision rate of 3.1% (CI 2.7–3.6) in cemented HA compared with 5.1% (CI 4.2–6.2) in the uncemented HA was found with a lower hazard ratio for revision in cemented compared with uncemented HA (HR 0.56, CI 0.47–0.67). Interpretation — Long-term mortality rates did not differ between patients with a cemented or uncemented HA after an acute femoral neck fracture. Revision rates were lower in cemented compared with uncemented HA.