Healthcare trajectories after an initial diagnosis for mood/anxiety disorders via telehealth versus in-person visit among pregnant women without pre-existing mood/anxiety disorders: Latent growth analyses of electronic medical records
Mahip Acharya, Cari A. Bogulski, Mir M. AliBackground
With telehealth commonly used for mental health diagnosis and management during pregnancy, it is unknown how it affects utilization trajectories.
Methods
We conducted a retrospective cohort study using a national database of electronic medical records (Epic Cosmos), January 2018–November 2025, among pregnant women without pre-existing anxiety, depression, or bipolar disorders. The initial in-person outpatient or telehealth visit/date during pregnancy with a diagnosis of anxiety/depression/bipolar was considered index visit/date. The sample was followed from the index date for 360 days. Four outcomes with anxiety/depression/bipolar diagnoses were studied for each 30-day follow-up interval: 1) emergency department (ED) visits, 2) hospitalizations, 3) in-person outpatient/office visits, and 4) telehealth visits. Multigroup latent growth models were estimated. Average marginal effects (AME) were reported.
Results
The final sample had 121,818 in-person outpatient and 7,763 telehealth initiators. The average ages were 28.9 years (in-person) and 30.2 years (telehealth); 84% (in-person) and 87% (telehealth) were urban residents. The AMEs of telehealth use for ED visits and hospitalizations were (in percentage points), respectively, -0.128 (-0.216; -0.040) and -0.222 (-0.432; -0.012). Telehealth initiators were less likely to have in-person outpatient visits (AME: -3.607 (-4.028; -3.185)) and more likely to have telehealth visits (AME: 2.741 (2.380; 3.102)).
Conclusion
The initial diagnosis of anxiety, depression, or bipolar conditions via telehealth was associated with similar rates of ED visits and hospitalizations, lower rates of outpatient visits, and higher rates of subsequent telehealth visits.