JMIR Publications

ABSTRACT

Background:

Patient safety in obstetrics can be enhanced through theory-based communication interventions, mitigating risks of preventable adverse events (PAE). Digital interventions could be a way of scaling up and expanding interventions.

Objective:

To find out to what extent patient safety and individual determinants of PAE (perceived safety, communication behavior, outcome expectancies, self-efficacy, planning) can be improved through a training theory-based on the health action process approach and delivered via the internet in form of a web-app.

Methods:

A mixed methods design was applied. Qualitative data were collected through semi-structured interviews with N = 20 app users and analyzed thematically with qualitative content analysis (1) using MAXQDA2020. Quantitatively, a nationwide online sample of N = 651 participants was recruited with provided informed consent and were randomized to the waitlist control group (CG; NCG = 324), or intervention group (IG; NIG = 327) with immediate access to the app. Baseline values were drawn from the pre-app survey (T0), and post-intervention values from T2 (after Module 3) for the IG, and at T1 (before Module 1) for the CG. Missing data were imputed using Multiple Imputation by Chained Equations (MICE; m = 122 datasets). All test variables were linearly rescaled to a 0–100 metric to ensure comparability. Bayesian mixed-effects models were fitted for each test variable with group×time as the primary interaction term, including informed priors from previous evidence (age, education, family status).

Results:

Qualitative findings revealed that participants perceived improved confidence, empathy, and preparedness for childbirth, and valued the app’s clarity, usability, and relevance to real clinical encounters. In the qualitative study, 75% of the study participants (n = 15) worked through all modules and those who discontinued (n = 5) reported time constraints or lack of interest. Duration of engagement ranged from a few days to several months. Increased empathy and perspective taking was mentioned most often as learnings. In the quantitative study, the app produced meaningful improvements across most evaluated constructs. Posterior estimates for the group×time interaction indicated significant effects of −0.30 (95% CrI [−0.33, −0.26]) for perceived safety, 0.38 (95% Credible Interval [0.35, 0.41]) for communication behavior, 0.21 (95% CrI [0.17, 0.24]) for self-efficacy, and 0.83 (95% CrI [0.79, 0.86]) for action planning. In contrast, there were no interaction effects of group×time for outcome expectancies (0.03, 95% Crl [-0.07, 0.01]).

Conclusions:

The app effectively enhanced patient safety and volitional indicators of safe communication (behavior, self-efficacy, planning) in line with the theory and hypotheses. Together, the qualitative and quantitative findings show that internet-delivered interventions can produce measurable improvements in communication behavior and perceived patient safety when designed with interactive, user-centered, and theory-based elements. Future implementations should strengthen personalization, feedback, and adaptive engagement strategies to sustain use and extend behavioral transfer beyond obstetric contexts. Clinical Trial: ClinicalTrials.gov Identifier: NCT03855735; https://classic.clinicaltrials.gov/ct2/show/NCT03855735