skip to page content
Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Systems Research

Go to the VA ORD website
Go to the QUERI website

C19 20-396 – HSR Study

C19 20-396
Changes in the Delivery of Evidenced Based Psychotherapies for Depression and PTSD as the Result of COVID-19 Pandemic
Diana M Mendez, PhD
Orlando VA Medical Center, Orlando, FL
Orlando, FL
Funding Period: July 2020 - March 2021
The VA/DOD Clinical Practice Guidelines for Major Depressive Disorder (MDD) and Posttraumatic Stress Disorder (PTSD) recommend evidence-based psychotherapies (EBPs) among the first-line treatments for both conditions. The onset of the COVID-19 pandemic led to abrupt social isolation and an unexpected change in the delivery format of mental health interventions. This change was characterized by the immediate cancellation of face-to-face encounters and the increased use of tele-mental health (TMH) and telephone encounters. With an unprecedented surge in demand to conduct psychotherapy remotely, VA mental health providers have had to utilize the telephone modality in treatment delivery, although its efficacy compared to TMH has not been well established.

The purpose was to determine the overall impact of the COVID-19 onset on the EBP treatment adherence for veterans with depression and PTSD across VA systems across the US and territories. Two specific aims addressed were: (1) To compare rates of attrition from EBP in veterans with depression and PTSD pre- and post-onset of the COVID-19 pandemic; (2) to compare the outcomes of EBP for depression and PTSD between telephone and video delivery modalities during and following the onset of the COVID-19 pandemic.

Retrospective national cohort using VA CDW data on the VINCI platform of all Veterans enrolled at a VAMC or HCS, who received evidence-based psychotherapy treatment sessions for PTSD and depression between September 1, 2017 through September 1, 2021. Identification of patients engaged in EBP treatment was recognized by utilization of the five (5) treatment templates: Acceptance and Commitment Therapy for Depression (ACT-D), Cognitive-Behavioral Therapy for Depression (CBT-D), Interpersonal Psychotherapy for Depression (IPT-D), Cognitive Processing for PTSD (CPT), and Prolonged Exposure Therapy for PTSD (PE).

Sample inclusion criteria included: (1) having a template recorded for initial session in one (and only one) of the 5 EBP templates, (2) the initial session taking place prior to May 19, 2020, which would give at least 120 days to complete the treatment, (3) an existent demographic profile, (4) being alive within 120 days after the initial session, and (5) not having missing values on sex, marital status, race, or ethnicity indicators. Among the 78,126 Veterans who met the inclusion criteria the demographics were mean age = 47.2 ±13.8 years, 21.8% female, 25.8% Black, 9.1% Hispanic.
Multiple logistic regressions were used to estimate the probability of treatment fidelity -pre and post the COVID-19 pandemic onset.

Prior to the onset of the COVID-19 pandemic (March 1, 2020- i.e., when first documented COVID-19 related death was documented in the US), the overall treatment completion rate for MDD EBPs (i.e. ACT, CBT-D, and IPT) was 24.5% [ACT (n= 5,559) =31.6%; CBT-D (n=13,634) = 21.7%, IPT (n=2,432) = 40.7%]. Pre-pandemic completion rates for PTSD EBPs was 23.1% [CPT (n=42,148) = 18.1%; PE (n=11,014) = 38.2%]. Overall, higher completion rates were noted among Veterans who were older, white and among those who never married. No seasonal variation was observed.

Among Veterans who started their EBP after the onset of the pandemic (n=4,439) the odds (OR) of completing the recommended minimum number of sessions were: (1) ACT-D (n=379) = 1.17 (0.88-1.55; p=0.3); (2) CBT-D (n=775) = 1.46 (1.19-1.79; p<.001); (3) IPT-D (n=194) = 1.14 (0.77-1.67; p=0.51); (4)CPT (n=2,525) =1.26 (1.11-1.42; p<.001); and (5) PE (n=566) =0.74 (0.58-0.93; p=0.011).

A logistic regression with a binary outcome was used to evaluate the rate of early termination of therapy. The outcome was operationalized as reaching the minimum recommended number therapy session for that template completion during the first 120 days after the initial therapy session (12 for CPT, 8 for all others). Predictors included EPB template, age at initial session, sex, marital status, race, ethnicity, and timing of session relative to the onset of COVID-19 pandemic ( pre vs post). Fitting the model to a stratified subsample of data, logistic regression revealed that after adjusting for other predictors, some templates had significantly higher odds of completing the therapy than others: using ACT as a reference point, patients in IPT and PEI templates were 50% and 44% percent more likely to complete therapy, respectively, and patients in CBT-D and CPT were 36% and 50% percent less likely to complete therapy, respectively. Females were 10% more likely to complete therapy than males (OR = 1.10, p = 0.042). Any marital status other than "Never Married" reduced the probability of carrying the treatment to fidelity: being Divorced reduced patients' chanced by 14% (OR = .86, p = .009) and being Married by 10% (OR = .90, p = 0.06).

Examination of the role of race and ethnicity confirmed the expectation of health disparities: Non-Whites (OR = 0.90, p = .022) and Latinx (OR = .87, p = .059) exhibited decreased likelihood of finishing the treatment. Against our initial hypothesis, the onset of social distancing measures associated with COVID-19 resulted in 17% increase (OR = 1.17, p = .056) in likelihood of carrying the treatment to fidelity.
When evaluating the same statistical model against the data for individual templates, we observed similar patterns, however some templates exhibiting more pronounced effects than others. For example, the effect of being married or divorced almost always gaged around 15-20% reduction in the likelihood of completing the treatment across all templates, with some minor variation in uncertainty. Being Non-White or Latinx almost always had mild (10-15%) adverse effect on the probability of carrying the treatment to fidelity, but not in CBT-D and IPT templates. The effect of whether the EBP session occurred before or after the onset of COVID-19 varied by template: for CBT-D (OR: 1.46, p < .0001) and CPT (OR = 1.26, p < .0001) it increased the odds treatment completion, for PE (OR = .74, p = .011) it decreased it, and for ACT (OR = 1.17, p = .3) and IPT (OR = 1.14, p = .5) this effect was not significant.

While developing the proposal for this project, data was expected to be obtained via clinical notes in which EBP providers record the modality of the session (face-to-face, telehealth, or telephone) by checking of the three respective checkboxes. However, only 64% of patient notes associated with EBP visits of our cohort included unambiguous information about the modality of the session. Of these 64%(606,023 ), 69.8% (480,306) identified the session as face-to-face, 16.3% ( 111,784) as Telehealth, and 2.2 % (14,933) as Telephone modality. After the onset of COVID-19, it was noted that providers were more likely to document the delivery modality in the notes; thereby, indicating that clinical documentation underwent improvement during the post COVID-19 onset period. It was also noted that during this period, there were still some face to face visits for each EBP although the majority of visits were either through telehealth or phone. The number of telehealth sessions were between 2.5 to 5 times the number of phone sessions depending on the template. Trends also indicate an overall decrease in the average scores on the mental health assessments post COVID-19 onset although we were unable to distinguish between EBPs and session modalities.

Information regarding veteran engagement and adherence to evidence-based psychotherapies for depression and PTSD is imperative, as these are treatments that have been proven to significantly improve the aforementioned conditions and to reduce suicidal ideation. Our findings demonstrate that EBP completion rates, as evidence by clinical documentation, in some cases increased with the shift to telehealth almost entirely to non in-person formats after the COVID-19 pandemic onset. This highlights that VA efforts to ensure continuity of care from a psychotherapeutic standpoint were optimal and yielded positive results. However, our findings also indicate possible need for increased treatment engagement interventions efforts due to high drop-out rates, both, pre and post COVID-19 onset. They also confirm potential need to increase culturally sensitive engagement efforts when providing evidence-based psychotherapy treatment to veterans who self-identify as non-white, who were shown to be even more likely to prematurely disengage from all treatments. Findings from this study can impact mental health clinical practice, and guide future trainings, guidelines for clinical documentation, and mental health treatment interventions.


None at this time.

DRA: Mental, Cognitive and Behavioral Disorders
DRE: Treatment - Observational, TRL - Applied/Translational
Keywords: Depression, PTSD
MeSH Terms: None at this time.

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.