Feasibility of a Digital Coaching Program for Improving Mental Well-Being and Emotional Intelligence: Pragmatic Retrospective Cohort Study

Abstract

Background: Within the past decade, digital coaching programs (DCPs) have emerged as an evidence-based modality to improve mental well-being and emotional intelligence (EI), although there is limited evidence in real-world contexts.

Objective: This pragmatic retrospective cohort study aims to determine the preliminary effectiveness of a DCP in improving mental well-being and EI within a real-world context. We hypothesized that there would be a significant increase in mental well-being and EI.

Methods: This study included 588 people who voluntarily enrolled in an 8-week, blended care DCP offered through their employers from October 2021 to August 2024. The DCP included routine check-ins and consultations with certified coaches. Participants completed the World Health Organization-Five Well-Being Index (WHO-5) at baseline and then weekly until the end of the program, as well as the Brief Emotional Intelligence Scale-10 (BEIS-10) at baseline and the end of the program. Multivariable linear mixed models examined changes in WHO-5 (biweekly) and BEIS-10 (pre-post) scores, adjusting for age, gender, program engagement, and program completion. Multivariable logistic regression models evaluated correlates of clinically meaningful improvements on the WHO-5 (ie, at least a 10-point improvement). We calculated a reliable change index (RCI) for the BEIS-10 and the proportion of participants meeting the RCI criterion from baseline to end of treatment.

Results: In multivariate linear mixed models adjusting for demographics and program characteristics, we observed a significant increase in WHO-5 scores (baseline x¯=45.6; week 8 x¯=66.3; Cohen's d=1.98; P<.001). Over half of the sample (55.4%) experienced a clinically meaningful improvement on the WHO-5. Multivariable logistic regression found that higher engagement was associated with an increased odds of a clinically meaningful improvement on the WHO-5 (adjusted odds ratio [aOR] 1.002, 95% CI 1.001-1.003), while program noncompletion (aOR 0.27, 95% CI 0.15-0.50) and higher baseline well-being (aOR 0.91, 95% CI 0.89-0.92) were associated with reduced odds. BEIS-10 scores also significantly increased from baseline to the end of the program after adjusting for relevant correlates (baseline x¯=37.6; week 8 x¯=41.2; Cohen's d=1.32; P<.001). The estimated RCI on the BEIS-10 was approximately 5, with 19.7% experiencing a meaningful improvement.

Conclusions: These results demonstrate that DCPs can be a viable option for individuals looking to improve their mental well-being. Additional efforts should focus on establishing reliable change metrics for EI measures. Studies using hybrid effectiveness-implementation trial designs are now needed to further evaluate the real-world effectiveness of this program.

Keywords: digital health; emotional intelligence; epidemiology; implementation science; mental health services; quality of life.

Figure 1.. Patterns of engagement during an 8-week digital coaching program (N=588).

Figure 2.. Change in WHO-5 scores during an 8-week digital coaching program (N=588). Error bands indicate 95% CIs. WHO-5 scores range from 0 to 100, with higher scores indicating better mental well-being. Scores of 50 or less indicate poor mental health requiring clinical attention. WHO-5: World Health Organization-Five Well-Being Index.

Figure 3.. Multivariate associations with clinically significant improvement in mental well-being (N=588) Program noncompletion was defined as engaging in the program for less than 4 weeks. Reference groups are included parenthetically. Total active minutes, age, baseline WHO-5, and baseline BEIS-10 do not show visible CIs due to their extremely narrow ranges relative to other predictors. * indicates continuous variables. WHO-5: World Health Organization-Five Well-Being Index; BEIS-10: Brief Emotional Intelligence Scale-10.