JMIR Publications

ABSTRACT

Background:

eHealth systems have been increasingly utilized to manage depressive symptoms in patients with somatic illnesses. However, understanding the factors that drive their usage, particularly among breast and prostate cancer patients, remains a critical area of research.

Objective:

This study aimed to determine the factors influencing usage of the NEVERMIND e-health system among breast and prostate cancer patients over 12 weeks, with a focus on the Technology Acceptance Model (TAM).

Methods:

Data from the NEVERMIND trial, encompassing 129 breast and prostate cancer patients, were retrieved. At baseline, participants completed questionnaires detailing demographic data and measuring depressive and stress symptoms, using Beck Depression Inventory-II and Depression, Anxiety, and Stress Scale-21 respectively. Over a 12-week period, patients engaged with the NEVERMIND system, with follow-up questionnaires administered at 4 weeks and after 12 weeks, assessing the system's perceived ease of use and usefulness. Usage log data were collected at the 2-week and 12-week marks. The relationships between sex, education, baseline depressive and stress symptoms, perceived ease of use, perceived usefulness, and system usage at various stages were examined using Bayesian Structural Equation Modeling path analysis, a technique differing from traditional frequentist methods.

Results:

The path analysis was conducted on 100 breast and prostate cancer patients, with a majority of females (66%) and individuals with a college education (81%). Patients reported good mental health scores, with low levels of depression and stress at baseline. System usage was approximately six days in the initial two weeks and around 45 days over the 12-week study period. Results revealed that perceived usefulness was the strongest predictor of system usage at 12-weeks (βuse12w~pu12w=.384), while early engagement (usage at two weeks) moderately predicted later usage (βuse12w~use2w=.239). Notably, uncertain associations were observed between baseline variables (education, sex, and mental health symptoms) and early system usage. A positive residual covariance was found between sex and system usage at 12 weeks (Buse12w~female=0.240), suggesting a possible disparity between breast and prostate cancer patients' system usage.

Conclusions:

The perceived usefulness of the NEVERMIND system and early engagement were both significant predictors of its overall use at the end of the 12-week intervention period. This suggests that in general eHealth implementations, caregivers should educate patients about the benefits and functionalities of such systems, thus enhancing their understanding of potential health impacts. Concentrating resources on promoting early engagement is also essential, given its influence on sustained use. Further research is necessary to clarify remaining uncertainties, enabling us to refine our strategies and maximize the benefits of eHealth systems in healthcare settings.