Mobile Health Adoption in High-Risk Pregnancies Using Cluster Analysis of Biopsychosocial Outcomes: Observational Longitudinal Cohort Study.

BACKGROUND

The use of mobile technologies during high-risk pregnancy, placing patients at the center of care, affords them self-management and easier access to health information.

OBJECTIVE

This study aims to understand the health perception of pregnant women at the beginning of high-risk antenatal care, the usability of a mobile health app-the Health Assistant-and to compare maternal-fetal outcomes between users and nonusers of the app.

METHODS

This is an observational longitudinal cohort study that looked into clusters of high-risk pregnant women admitted to antenatal care at the maternity unit of a public university hospital in southern Brazil between April 2022 and November 2023. Pregnant women who did not have a compatible smartphone to download the app or who did not have internet access were excluded from the study. According to systematic randomization, one patient was allocated to the app group and the other to the control group. They all answered an inclusion questionnaire (Q1), and those in the app group were instructed to use the Health Assistant app to prepare for their first antenatal appointment, which would take place in a few weeks' time, when they would answer the Brazilian version of the Mobile App Usability Questionnaire. After childbirth, maternal-fetal outcomes were assessed. Student 2-tailed t test, Mann-Whitney test, Fisher exact test, and the chi-square test were used for statistical analysis. A hierarchical cluster analysis was performed using the Ward method and the Euclidean squared distance measure.

RESULTS

The sample contained 111 pregnant women, of whom 55 (49.5%) were allocated to the app group and 56 (50.5%) to the control group. Of the 55 pregnant women who used the app, 21 (38.2%) demonstrated adherence, with an average Mobile App Usability Questionnaire score of 6.2 (SD 1.0). Clustering included 110 pregnant women, and the dendrogram resulted in three clusters, which show several significant differences in terms of family income, medical history, medication adherence, and lifestyle habits. Cluster 2 had the lowest adherence to the app (P=.08) and attended significantly fewer antenatal appointments (6.9 appointments) as compared with Clusters 1 (10.3) and 3 (9.1; P=.006). Cesarean section was more frequent in Cluster 3 (n=41, 95.3%) as compared with Clusters 1 (n=12, 27.9%) and 2 (n=5, 20.8%), P<.001.

CONCLUSIONS

Cluster analysis, revealing different profiles of pregnant women, allowed us to identify groups that would benefit from personalized approaches and digital interventions to improve self-awareness and gestational outcomes. The Health Assistant app showed good usability in this context.