Health-related quality of life in the finnish trial of screening for prostate cancer

Background: Evidence of the potential impact of systematic screening for prostate cancer (PCa) on health-related quality of life (HRQoL) at a population-based level is currently scarce. Objective: This study aims to quantify the long-term HRQoL impact associated with screening for PCa. Design, setting, and participants: Postal questionnaire surveys were conducted in 1998, 2000, 2004, and 2011 among men in the Finnish PCa screening trial diagnosed with PCa (total n = 7011) and among a random subsample of the trial population (n = 2200). In 2011, for example, 1587 responses were received from men with PCa in the screening arm and 1706 from men in the control arm. In addition, from the trial subsample, 549 men in the screening arm and 539 in the control arm provided responses. Outcome measurements and statistical analysis: Health-state-value scores were compared between the intervention and control arms using three distinct HRQoL measures (15D, EQ-5D, and SF-6D), and statistical significance was assessed using t tests. In addition, differences over repeated assessments of HRQoL between groups were evaluated using generalised estimating equations. Results and limitations: In the 2011 survey, a small but statistically significant difference emerged between the trial arms among men diagnosed with PCa (mean scores, screening vs control arm: 15D: 0.872 vs 0.866, p = 0.14: EQ-5D: 0.852 vs 0.831, p = 0.03: and SF-6D: 0.763 vs 0.756, p = 0.06). Such differences in favour of the screening arm were not found among the sample of men from the trial (15D: 0.889 vs 0.892, p = 0.62: EQ-5D: 0.831 vs 0.852, p = 0.08: and SF-6D: 0.775 vs 0.777, p = 0.88). The slight advantage with screening among men with PCa was reasonably consistent across time in the longitudinal analysis and was strongest among men with early-stage disease. Conclusions: These results show some long-term HRQoL benefit from screening for men with PCa but suggest little impact overall in the trial population.