A randomized controlled trial comparing pragmatic interventions to improve mammogram uptake in a non-compliant population.

Abstract

11008 Background: Despite over 20 years of a national breast cancer screening program, mammogram uptake in Singapore, a highly developed Asian country, remains under 40%. Limited large-scale prospective data exist on effective pragmatic interventions to boost screening participation. Methods: 9000 Singaporean women aged 50-69, eligible for free biennial screening, non-compliant (at least 1 prior mammogram but overdue by > 2 years from last screen), with no breast cancer history and registered on hospital’s mobile health app, were randomly selected from a tertiary hospital’s electronic medical records and allocated to one of the 5 arms in a 2:1:1:1:1 ratio. Arm 1 (control; n = 3000) received physical mailer reminders (MR). Four intervention arms (n = 1500 each) received MR plus- Arm 2: US$7.4 voucher on screening completion, Arm 3: US$3700 lottery chance, Arm 4: video health message, and Arm 5: health concierge for appointment scheduling. Arms 1–5 also received the reminders via 3 push notifications (PN), spaced 3 weeks apart, through the app. A non-interventional (NI) cohort (n = 3000) fulfilling similar criteria served as comparison. Primary outcome was mammogram uptake rate within 4 months of reminder in Arm 1 vs Arms 2-5. Secondary outcomes included participation rate in Arm 1 vs NI group. Exploratory analysis included cost-effectiveness of intervention and app utility in shaping health behaviour. Results: 74% of the 12000 participants were Chinese, followed by Malay (13%) and Indian (7%), reflecting Singapore’s demographics. Median age was 62 years (67% aged 60-69). 89% lived in public and 11% in private housing. 99% lived within 5 km of a mammogram facility. Median interval from last mammogram was 5.2 years (range: 2.0–21.5); longer in women aged 60-69 vs 50-59 (6.1 vs 3.9 years, p < 0.0001). Mammogram uptake in the NI group was 3.3% and increased to 11.2% with MR alone (Arm 1; RR 3.28, 95% CI 2.60–3.96, p < 0.01). Uptake in Arm 5 was slightly higher than in Arm 1 (13.8% vs 11.2%, RR 1.23, 95% CI 1.07-1.38, p = 0.01), but not significant after adjusting for multiple comparisons (p-adj = 0.11). Arms 2-4 (12.1%, 11.3%, 10.8%) showed no difference vs Arm 1. Intervention cost per compliant subject was US$7 in Arm 1, increasing to $13, $28, $32 and $40 in Arms 2-5 respectively. Only 4.1% (3.1–4.6%) in the 5 intervention arms accessed the webpage via PN and < 1% clicked the self-help booking link. Conclusions: A physical MR significantly improved mammogram participation in a non-compliant population, but additional monetary incentives or health message offered no further benefit. A concierge service slightly improved uptake but at a 6-fold higher cost. PNs had limited utility in this older population. A low-cost MR program is scalable and piloting concierge services for targeted groups may enhance cost-effectiveness. Further research should explore barriers to digital intervention engagement in this age group. Clinical trial information: NCT06733155 .