Smartphone-Based Intervention for University Students with Unhealthy Alcohol Use

Service de mdecine des addictions Smartphone-based secondary prevention intervention for university students with unhealthy alcohol use: a randomized controlled trial Nicolas Bertholet Elodie Schmutz Joseph Studer Ang line Adam Gerhard Gmel John A Cunningham Jennifer McNeely Jean-Bernard Daeppen

Funding: Swiss National Science Foundation, grant 176295 Trial registration: ISRCTN 10007691 No conflict of interest to report

Background Electronic screening and brief intervention is considered an effective method for reducing unhealthy alcohol use Smartphone applications: May be readily adopted, especially among young individuals Allow for multiple-contact and proactive or just in time interventions Linkage possible to additional services Current evidence for smartphone-based interventions efficacy is inconclusive In addition, many applications that are currently available may have harmful effects as they promote drinking

Methods The present study assessed the impact of providing access to a smartphone app for unhealthy alcohol use. The app was adapted from a previous version and co-developped with students. The study was conducted among university students with unhealthy alcohol use, identified by screening, in four higher education institutions in Switzerland.

Methods The study primary hypothesis was that participants randomized to receive access to the smartphone app for unhealthy alcohol use would report less drinking 6 months later, compared to participants who did not receive access to the app.

Methods Participants were recruited at 4 higher education institutions in Switzerland The study was presented on the website/social media of student associations + email Eligibility criteria were: Being a student at the time of recruitment in the study Screening positive for unhealthy alcohol use (defined as AUDIT-C score 4 for men and 3 for women) Reporting smartphone ownership Being willing to complete the follow-up assessments.

Methods The entire study was conducted electronically Follow-up at 3, 6 and 12 months

Smartphone app 6 modules: Personalized feedback Normative feedback Feedback on calorific content of the reported consumption Feedback on health risks Blood alcohol content computation module Self-monitoring tool Goal-setting tool Designated driver tool Fact sheets, contacts

Methods Primary outcome: weekly volume of drinking, in standard drinks, measured at 6 months Secondary outcome: number of heavy drinking days (HDD; i.e. days with 5 drinks or more for men / 4 drinks or more for women) over the past 30 days, at 6 months Additional outcomes: maximum number of drinks on any day over the past 30 days, alcohol related consequences (measured with the Short Inventory of Problems (SIP-2R), and academic performance.

Methods The intervention effect on the outcomes was tested using generalized linear mixed models (GLMM). Participants and recruitment sites were entered as random effects and intervention and time as fixed effects, with an intervention by time interaction term. Models were adjusted for age and sex. Missing data were handled using multiple imputation

Results: flow Screening completed N=3714 Ineligible: No smartphone: N=42 Not willing to complete assessments: N=288 Unhealthy alcohol use N=2694 Eligible participants N=2364 Accepted participation, randomized N=1770 Intervention group N=884 Control group N=886 3 months assessment completed N=846, 96% 3 months assessment completed N=860, 97% 6 months assessment completed N=846, 96% 6 months assessment completed N=851, 96% 12 months assessment completed N=827, 94% 12 months assessment completed N=833, 94%

Participants baseline characteristics Full Sample (N = 1770) Mean / N 22.35 SD / % 3.07 Age Sex Female Male Education program Bachelor Master Doctorate Other AUDIT score Number of standard drinks per week Number of heavy drinking days past 30 days Maximum number of drinks on one occasion, past 30 days Alcohol-related consequences** Academic performance*** 958 812 54.1% 45.9% 1169 533 43 25 9.60 8.59 3.53 7.40 3.83 3.18 66.0% 30.1% 2.4% 1.4% 4.70 8.18 4.02 4.13 3.96 0.79 **: alcohol related consequences were measured with the Short Inventory of Problems (SIP-2R). The minimum possible score is 0, the maximum is 45. The timeframe is the past 3 months. ***: Academic performance was measured on a 1 to 5 scale with the following question: How do you rate your performance in comparison with your fellow students? (with response options : Much worse, Worse, Similar, Better, Much better).

Outcome measures at baseline, 3, 6 and 12 months Outcome measures at baseline, 3, 6 and 12 months Weekly volume of drinking over time Heavy drinking days per month, over time 12 5 11 4.5 10 9 4 8 3.5 7 3 6 5 2.5 baseline 3 months 6 months 12 months baseline 3 months 6 months 12 months control (V) intervention (V) control (HDD) intervention (HDD) Baseline (N = 1770; control: n = 886; intervention: n = 884) Mean (SD) 3 months (N = 1706; control: n = 860; intervention: n = 846) Mean (SD) 6 months (N = 1697; control: n = 851; intervention: n = 846) Mean (SD) 12 months (N = 1660; control: n = 833; intervention: n = 827) Mean (SD) Number of standard drinks per week Total Control Intervention Number of heavy drinking days past 30 days Total Control Intervention 8.59 (8.18) 8.25 (7.65) 8.93 (8.66) 10.44 (10.85) 10.80 (11.33) 10.07 (10.33) 7.39 (7.26) 7.68 (8.11) 7.11 (6.29) 7.32 (7.12) 7.59 (7.88) 7.04 (6.25) 3.53 (4.02) 3.48 (3.83) 3.58 (4.19) 4.35 (4.93) 4.54 (5.13) 4.15 (4.71) 3.21 (3.43) 3.39 (3.65) 3.02 (3.18) 3.17 (3.46) 3.26 (3.47) 3.07 (3.45)

assessment of intervention efficacy, intention assessment of intervention efficacy, intention- - to to- -treat analysis treat analysis Standard drinks per week heavy drinking days, past 30 days Maximum number of drinks, past 30 days IRR 95% CI 0.99 0.94, 1.04 Alcohol-related consequences* Academic performance** IRR 1.05 95% CI 0.96, 1.14 IRR 1.00 95% CI 0.90, 1.11 IRR 0.96 95% CI 0.86, 1.07 b -0.01 95% CI -0.08, 0.06 Intervention (ref. control) Time (ref. baseline) 3 months 6 months 12 months Time by Intervention interaction Intervention by 3 months Intervention by 6 months Intervention by 12 months 1.26 0.91 0.89 1.19, 1.33 0.86, 0.96 0.84, 0.95 1.26 0.97 0.94 1.18, 1.35 0.90, 1.04 0.87, 1.01 1.02 0.98 0.95 0.98, 1.05 0.94, 1.01 0.92, 0.99 0.87 0.94 0.91 0.82, 0.93 0.89, 1.00 0.86, 0.97 0.07 0.00 0.02 0.02, 0.11 -0.04, 0.04 -0.02, 0.06 0.89 0.82, 0.96 0.90 0.82, 0.99 0.94 0.89, 0.99 0.96 0.88, 1.04 0.01 -0.05, 0.07 0.88 0.81, 0.96 0.88 0.80, 0.97 0.96 0.91, 1.01 1.01 0.93, 1.10 0.04 -0.02, 0.09 0.88 0.81, 0.96 0.91 0.82, 1.01 0.98 0.93, 1.03 0.92 0.84, 1.01 0.00 -0.06, 0.06

Conclusions In this study, giving access to a smartphone application was associated with less drinking The impact on drinking was found at the primary time point (6 months follow-up) on the primary outcome measure (number of standard drinks per week) and on the secondary outcome measure (heavy drinking days). The impact was sustained at 12 months for the primary outcome (number of standard drinks per week) but not for heavy drinking days. There was a significant impact on the maximum number of drinks per occasion at 3 months. No impact was observed on alcohol related consequences or on self- reported academic performance.

Conclusions The studied intervention requires fewer resources than face-to-face interventions (limited costs, no need to specifically hire and train health care providers to perform screening and brief interventions), can be offered to large groups using existing communication channels (emails, social media) and has good adoption. Potential for widespread implementation

Caveats Assessments based on self report only The study was conducted during the COVID-19 pandemic (although not at its height) Recruitment started when students came back on campuses after a lockdown period and at a time when restrictions related to COVID-19 were being partially lifted. This may explain the observed general increase in drinking observed at 3 months in the control and intervention groups, and matched the lifting of restrictions on alcohol availability.

Internet based intervention Prototype 0 Pilot study Literature review, expert meetings 10 individual semi-structured interviews (after >= 2weeks of use) Analysis + research team group discussion on suggested modifications Validation with the application developers team (practicability) Development of prototype 1 TEST 1 11* individual semi-structured interviews (after >= 2weeks of use) Analysis + research team group discussion on suggested modifications Validation with the application developers team (practicability) Development of prototype 2 TEST 2 PHASE II (randomized trial) Development of final application

Participants baseline characteristics Full Sample (N = 1770) Intervention group (N = 884) Control group (N = 886) Comparison Intervention vs. Control P-value Mean / N SD / % Mean / N 22.24 SD / % Mean / N 22.45 SD / % Age 22.35 3.07 2.85 3.27 .157 Gender .199 Female 958 54.1% 465 52.6% 493 55.6% Male 812 45.9% 419 47.4% 393 44.4% Education program .385 Bachelor 1169 66.0% 598 67.6% 571 64.4% Master 533 30.1% 253 28.6% 280 31.6% Doctorate 43 2.4% 20 2.3% 23 2.6% Other 25 1.4% 13 1.5% 12 1.4% AUDIT score 9.60 4.70 9.67 4.84 9.53 4.55 .863 Number of standard drinks per week 8.59 8.18 8.93 8.66 8.25 7.65 .470 Number of heavy drinking days past 30 days Maximum number of drinks on one occasion, past 30 days Alcohol-related consequences** 3.53 4.02 3.58 4.19 3.48 3.83 .953 7.40 4.13 7.41 4.16 7.39 4.10 .942 3.83 3.96 3.82 4.05 3.84 3.88 .993 Academic performance*** 3.18 0.79 3.18 0.80 3.19 0.78 .913

outcome measures at baseline, 3, 6 and 12 outcome measures at baseline, 3, 6 and 12 months months Baseline (N = 1770; control: n = 886; intervention: n = 884) 3 months (N = 1706; control: n = 860; intervention: n = 846) 6 months (N = 1697; control: n = 851; intervention: n = 846) 12 months (N = 1660; control: n = 833; intervention: n = 827) Maximum number of drinks on one occasion, past 30 days Total Control Intervention Alcohol-related consequences* Total Control Intervention Academic performance** Total Control Intervention 7.40 (4.13) 7.39 (4.10) 7.41 (4.16) 7.34 (4.46) 7.59 (4.83) 7.09 (4.04) 7.09 (4.28) 7.25 (4.41) 6.93 (4.14) 6.97 (4.26) 7.05 (4.49) 6.90 (4.02) 3.83 (3.96) 3.84 (3.88) 3.82 (4.05) 3.32 (3.78) 3.37 (3.73) 3.26 (3.84) 3.65 (4.13) 3.62 (3.92) 3.67 (4.33) 3.33 (3.76) 3.46 (3.85) 3.20 (3.67) 3.18 (0.79) 3.19 (0.78) 3.18 (0.80) 3.25 (0.79) 3.25 (0.79) 3.25 (0.78) 3.20 (0.73) 3.19 (0.74) 3.21 (0.73) 3.21 (0.73)a 3.21 (0.72)b 3.21 (0.73)c **: alcohol related consequences were measured with the Short Inventory of Problems (SIP-2R). The minimum possible score is 0, the maximum is 45. The timeframe is the past 3 months. ***: Academic performance was measured on a 1 to 5 scale with the following question: How do you rate your performance in comparison with your fellow students? (with response options : Much worse, Worse, Similar, Better, Much better).

Results: flow Screening completed N=3714 Ineligible: No smartphone: N=42 Not willing to complete assessments: N=288 Unhealthy alcohol use N=2694 Eligible participants N=2364 Accepted participation, randomized N=1770 Intervention group N=884 (app downloaded n=738, 83%) Control group N=886 (control procedure completed n=846, 95%) 3 months assessment completed N=846, 96% Included in analyses N=884 3 months assessment completed N=860, 97% Included in analyses N=886 6 months assessment completed N=846, 96% Included in analyses N=884 6 months assessment completed N=851, 96% Included in analyses N=886 12 months assessment completed N=827, 94% Included in analyses N=884 12 months assessment completed N=833, 94% Included in analyses N=886

assessment of intervention efficacy, per assessment of intervention efficacy, per- - protocol analysis protocol analysis Standard drinks per week heavy drinking days, past 30 days Maximum number of drinks, past 30 days Alcohol-related consequences* Academic performance** IRR 95% CI IRR 95% CI IRR 95% CI IRR 95% CI b 95% CI Intervention (ref. control) 1.04 0.95, 1.13 1.02 0.91, 1.13 1.00 0.94, 1.05 0.96 0.86, 1.08 -0.01 -0.08, 0.07 Time (ref. baseline) 3 months 1.26 1.19, 1.33 1.30 1.21, 1.39 1.03 0.99, 1.07 0.87 0.82, 0.93 0.07 0.02, 0.11 6 months 0.92 0.86, 0.97 1.00 0.93, 1.07 0.99 0.95, 1.03 0.94 0.88, 1.00 0.01 -0.05, 0.03 12 months 0.90 0.85, 0.96 0.97 0.90, 1.05 0.97 0.93, 1.00 0.92 0.86, 0.99 0.01 -0.03, 0.06 Time by Intervention interaction Intervention by 3 months 0.89 0.82, 0.97 0.86 0.77, 0.95 0.92 0.87, 0.97 0.95 0.86, 1.04 0.00 -0.06, 0.07 Intervention by 6 months 0.87 0.80, 0.94 0.84 0.75, 0.93 0.94 0.89, 0.99 1.00 0.91, 1.10 0.04 -0.03, 0.10 Intervention by 12 months 0.86 0.79, 0.94 0.87 0.78, 0.97 0.95 0.90, 1.00 0.89 0.81, 0.98 0.00 -0.06, 0.06

Study announcement: email/social media/posters Study website Randomization Anonymous screening and eligibility assessment Baseline assessment FU assessemnts: 3, 6, 12 months email email email Access to the app (link) No access to the app