AAT-App+: A double-blind, pilot randomised controlled trial of a novel personalised smartphone intervention to reduce hazardous and harmful alcohol consumption among middle–older adults

BACKGROUND: Given increased rates of problematic alcohol use among middle-older age adults, novel, accessible alcohol interventions are needed. This pilot RCT explored the feasibility, acceptability, and preliminary effectiveness of personalised smartphone-delivered alcohol approach bias modification (ApBM) in Australians (≥55 years) drinking at hazardous/harmful levels (AUDIT ≥ 8). METHOD: Double-blinded parallel two-arm pilot RCT in Victoria, Australia (July 2022-January 2023). Community-based ≥ 55-year-olds received 'AAT-App+' (n = 96) or sham training (n = 92) over four-weeks. Feasibility was defined as recruitment of ≥ 100 participants within the recruitment window and 50 % adherence to four sessions. Acceptability was defined as User Version Mobile Application Rating Scale (uMARS) mean scores ≥ 3. The primary preliminary effectiveness outcome was past-week standard drinks at week four relative to baseline. RESULTS: Feasibility and acceptability were good, with n = 188 recruited, 69 % training adherence, and 'good' uMARS functionality/aesthetics and 'acceptable' quality ratings among responders (n = 94). Retention rates were 56 % for primary effectiveness outcome. Linear mixed modelling using intention-to-treat analysis showed a significant mean reduction of 6.9 past-week standard drinks relative to baseline overall (p<.001), but no significant group-by-time interaction at the primary endpoint (β = -1.90, p=.43, 95 %CI [-6.62, 2.82]). The groups differed significantly favouring ApBM after two-weeks of training (β = -5.28, p=.03, 95 %CI [-10.01;-0.56]). There were no adverse events throughout the training window. CONCLUSION: Personalised smartphone ApBM is feasible and acceptable to middle-to-older adults, but we found no evidence to suggest it reduced alcohol consumption any more than sham-training beyond the acute period. Further work is needed prior to large-scale RCT progression to mitigate attrition and optimise study design.