UC Davis

Background

This study investigates outcomes from two federal grant programs: the Evidence-Based Tele-Behavioral Health Network Program (EB THNP) funded from September 2018 to August 2021 and the Substance Abuse Treatment Telehealth Network Grant Program (SAT TNGP) funded from September 2017 to August 2020. As part of the health services implementation program, the aims of this study were to evaluate outcomes in patient symptoms of depression and anxiety across the programs' 17 grantees and 95 associated sites, with each grantee having data from telehealth patients and from an in-person comparison group.

Methods

The research design is a prospective multi-site observational study. Each grantee provided data on a nonrandomized convenience sample of telehealth patients and an in-person comparison group from sites with similar rural characteristics and during the same time period. Patient characteristics were collected at treatment initiation, and clinical outcome measures were collected at baseline and monthly. The validated clinical outcome measure instruments included the Patient Health Questionnaire-9 (PHQ-9) for depression symptoms and the Generalized Anxiety Disorder-7 (GAD-7) scale for anxiety-related symptoms. Linear mixed models, with grantee as the random effect, were used to determine the association of behavioral health delivery (telehealth versus in-person) on the one-month change in PHQ-9 and GAD-7 while adjusting for covariates.

Results

Across a total of 1,514 patients, one-month change scores were improved indicating that PHQ-9 and GAD-7 scores decreased from baseline to the one-month follow-up at similar rates in both the in-person and telehealth groups. Reduction in scores averaged 2.8 for the telehealth treatment group and 2.9 for the in-person treatment group in the PHQ-9 subsample and 2.0 for the telehealth treatment group and 2.4 for the in-person treatment group in the GAD-7 subsample. There was no statistically significant association between the modality of care (telehealth treatment group versus in-person comparison group) and the one-month change scores for either PHQ-9 or GAD-7. Individuals with higher baseline scores demonstrated the greatest decrease in scores for both measures. Upon adjusting for baseline scores and grantee program, patient demographics were not found to be significantly associated with change in anxiety or depression symptoms.

Conclusion

In our very large pragmatic study comparing behavioral health treatment delivered to a population of patients in rural, underserved communities, we found no clinical or statistical differences in improvements in depression or anxiety symptoms as measured by the PHQ-9 and GAD-7 between patients treated via telehealth or in-person.