A Brief Intervention to Increase Uptake and Adherence of an Internet-Based Program for Depression and Anxiety (Enhancing Engagement With Psychosocial Interventions): Randomized Controlled Trial

Abstract

Background Psychosocial, self-guided, internet-based programs are effective in treating depression and anxiety. However, the community uptake of these programs is poor. Recent approaches to increasing engagement (defined as both uptake and adherence) in internet-based programs include brief engagement facilitation interventions (EFIs). However, these programs require evaluation to assess their efficacy. Objective The aims of this hybrid implementation effectiveness trial are to examine the effects of a brief internet-based EFI presented before an internet-based cognitive behavioral therapy self-help program (myCompass 2) in improving engagement (uptake and adherence) with that program (primary aim), assess the relative efficacy of the myCompass 2 program, and determine whether greater engagement was associated with improved efficacy (greater reduction in depression or anxiety symptoms) relative to the control (secondary aim). Methods A 3-arm randomized controlled trial (N=849; recruited via social media) assessed the independent efficacy of the EFI and myCompass 2. The myCompass 2 program was delivered with or without the EFI; both conditions were compared with an attention control condition. The EFI comprised brief (5 minutes), tailored audio-visual content on a series of click-through linear webpages. Results Uptake was high in all groups; 82.8% (703/849) of participants clicked through the intervention following the pretest survey. However, the difference in uptake between the EFI + myCompass 2 condition (234/280, 83.6%) and the myCompass 2 alone condition (222/285, 77.9%) was not significant (n=565; χ21=29.2; P=.09). In addition, there was no significant difference in the proportion of participants who started any number of modules (1-14 modules) versus those who started none between the EFI + myCompass 2 (214/565, 37.9%) and the myCompass 2 alone (210/565, 37.2%) conditions (n=565; χ21<0.1; P=.87). Finally, there was no significant difference between the EFI + myCompass 2 and the myCompass 2 alone conditions in the number of modules started (U=39366.50; z=−0.32; P=.75) or completed (U=39494.0; z=−0.29; P=.77). The myCompass 2 program was not found to be efficacious over time for symptoms of depression (F4,349.97=1.16; P=.33) or anxiety (F4,445.99=0.12; P=.98). However, planned contrasts suggested that myCompass 2 may have been effective for participants with elevated generalized anxiety disorder symptoms (F4,332.80=3.50; P=.01). Conclusions This brief internet-based EFI did not increase the uptake of or adherence to an existing internet-based program for depression and anxiety. Individuals’ motivation to initiate and complete internet-based self-guided interventions is complex and remains a significant challenge for self-guided interventions. Trial Registration Australian New Zealand Clinical Trials Registry ACTRN12618001565235; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=375839

Technology-supported strategies for promoting user engagement with digital mental health interventions: A systematic review

“… 32 , 33 , 36 , 43 , 44 , 47 , 49 The remaining seven studies reported no differences between the strategy and no-strategy groups across all measures of engagement. 26 , 28 , 29 , 35 , 39 , 41 , 46 …”

“…All studies offered feedback or suggestions based on users’ past programme use or individual context. These were delivered via in-app push notifications, 30 , 31 , 33 , 43 email, 28 , 29 internet-based platform 26 and text message. 45 Four studies evaluated the effect of tailored feedback on engagement compared to no-strategy.…”

“… 45 One study found that a brief engagement intervention, used in addition to reminders, had no incremental effect on the number of modules started or completed. 26 Finally, two studies investigated whether the time of dissemination and type of content (suggestions vs. insights) of tailored health messages influenced their effectiveness in nudging users to engage with a workplace mental health app. In general, users were more likely to engage with the app when they received a health message than if they had not received it, regardless of the time of day; notably, messages sent in the early afternoon on weekends were most likely to elicit app engagement.…”

“… 39 In particular, reminder-based strategies may have been frowned upon by some users who perceived them as controlling 35 or lacking emotional validation. 46 Apart from one study, 26 there was no information on whether prospective users of the digital interventions were involved in developing these strategies. This absence of end-user perspectives in the design process may have had a negative impact on the potential efficacy of the strategies examined.…”

The Effect of Dissemination Pathways on Uptake and Relative Costs for a Transdiagnostic, Self-guided Internet Intervention for Reducing Depression, Anxiety, and Suicidal Ideation: Comparative Implementation Study

“…It was anticipated that the relative reach (ratio of uptake to exposure) would be highest in the social media pathway, whereas the cost per enrolled participant would consequently be lower in this pathway. In addition, we hypothesized that engagement with the program (based on adherence; module completion) would be lowest among people recruited through social media, as this group had the least possibility of human contact in the recruitment process [ 46 ]; that is, the higher uptake on the web may come at the expense of engagement with the intervention. We did not have any specific hypotheses regarding outcomes for the pharmacy pathway, as this pathway has not been investigated previously in e-mental health implementation research.…”