Background: Diabetes is a very common chronic disease that exerts massive physiological and psychological burdens on patients. The digitalization of mental health care has generated effective e-mental health approaches, which offer an indubitable practical value for patient treatment. However, before implementing and optimizing e-mental health tools, their acceptance and underlying barriers and resources should be first determined for developing and establishing effective patient-oriented interventions.

Objective: This study aims to assess the acceptance of e-mental health interventions among patients with diabetes and explore its underlying barriers and resources.

Methods: A cross-sectional study was conducted in Germany from April 9, 2020, to June 15, 2020, through a web-based survey for which patients were recruited via web-based diabetes channels. The eligibility requirements were adult age (18 years or older), a good command of the German language, internet access, and a diagnosis of diabetes. Acceptance was measured using a modified questionnaire, which was based on the well-established Unified Theory of Acceptance and Use of Technology (UTAUT) and assessed health-related internet use, acceptance of e-mental health interventions, and its barriers and resources. Mental health was measured using validated and established instruments, namely the Generalized Anxiety Disorder Scale-7, Patient Health Questionnaire-2, and Distress Thermometer. In addition, sociodemographic and medical data regarding diabetes were collected.

Results: Of the 340 participants who started the survey, 261 (76.8%) completed it and the final sample comprised 258 participants with complete data sets. The acceptance of e-mental health interventions in patients with diabetes was overall moderate (mean 3.02, SD 1.14). Gender and having a mental disorder had a significant influence on acceptance (P<.001). In an extended UTAUT regression model (UTAUT predictors plus sociodemographics and mental health variables), distress (β=.11; P=.03) as well as the UTAUT predictors performance expectancy (β=.50; P<.001), effort expectancy (β=.15; P=.001), and social influence (β=.28; P<.001) significantly predicted acceptance. The comparison between an extended UTAUT regression model (13 predictors) and the UTAUT-only regression model (performance expectancy, effort expectancy, social influence) revealed no significant difference in explained variance (F_10,244=1.567; P=.12).

Conclusions: This study supports the viability of the UTAUT model and its predictors in assessing the acceptance of e-mental health interventions among patients with diabetes. Three UTAUT predictors reached a notable amount of explained variance of 75% in the acceptance, indicating that it is a very useful and efficient method for measuring e-mental health intervention acceptance in patients with diabetes. Owing to the close link between acceptance and use, acceptance-facilitating interventions focusing on these three UTAUT predictors should be fostered to bring forward the highly needed establishment of effective e-mental health interventions in psychodiabetology.