Effect of computer-assisted cognitive behavior therapy vs usual care on depression among adults in primary care: A randomized clinical trial

Wright, J. H., Owen, J., Eells, T. D., Antle, B., Bishop, L. B., Girdler, R., Harris, L. M., Wright, R. B., Wells, M. J., Gopalraj, R., Pendleton, M. E., & Ali, S. (2022). Effect of computer-assisted cognitive behavior therapy vs usual care on depression among adults in primary care: A randomized clinical trial. JAMA Network Open, 5(2), Article e2146716. https://doi.org/10.1001/jamanetworkopen.2021.46716 

Summary by: Ethan Leon Rosenthal

Introduction

The study by Wright et al. (2022), published in JAMA Network Open, investigates the effectiveness of computer-assisted cognitive behavioral therapy (CCBT) combined with treatment as usual (TAU) compared to TAU alone in treating depression among primary care patients. The rationale behind this research stems from the growing demand for scalable, evidence-based psychological treatments for depression. Traditional CBT is well established as an effective intervention, yet its accessibility remains limited due to shortages of trained therapists and high treatment costs. The authors hypothesized that computer-assisted CBT with limited clinician support could maintain effectiveness while reducing the burden on healthcare systems.

The objectives of the study were:

1. To evaluate the effectiveness of CCBT compared with treatment as usual (TAU) in primary care patients with depression.

2. To evaluate the feasibility of CCBT in a primary care population with substantial numbers of patients with low levels of education, reading skills, and/or internet access.

3. To perform an exploratory analysis of factors associated with differences in treatment outcomes.

4. To evaluate the medical care utilization costs of CCBT compared with TAU.

The authors noted that the economic analysis would be reported separately.

Study Design

This randomized clinical trial (RCT) was conducted in primary care practices in Kentucky, United States. A total of 175 adult participants who scored at least 10 on the Patient Health Questionnaire-9 (PHQ-9) were enrolled and randomly assigned to two groups: CCBT + TAU (n = 95) and TAU alone (n = 80). Most participants had a primary diagnosis of major depressive disorder (MDD), although not all met the diagnostic criteria. The mean age was approximately 47 years.

The study also included substantial numbers of patients with lower incomes, limited educational attainment, and lower reading proficiency, addressing populations often underrepresented in earlier research on digital psychological interventions.

Procedure

Participants in both groups received routine care from their primary care physicians, which could include medication, referrals, or other supportive measures.

The intervention group also received access to a computer-based CBT program called Good Days Ahead (GDA), which included nine interactive lessons focused on CBT principles and techniques. Participants were encouraged to complete the program over 12 weeks and could receive up to 12 weekly telephone support sessions from a master's-level mental health clinician. These sessions were designed to provide guidance, monitor progress, and encourage engagement, with approximately 20 minutes of support planned per week.

To improve accessibility, the researchers also provided laptops with internet access to participants who needed them.

The primary outcome was depression severity, measured by the Patient Health Questionnaire-9 (PHQ-9) at baseline, after 12 weeks of treatment, and at 3- and 6-month follow-ups following treatment completion. Secondary outcomes were measured using the Automatic Thoughts Questionnaire (ATQ) for negative cognitions, the Generalized Anxiety Disorder-7 (GAD-7) for anxiety symptoms, and the Satisfaction with Life Scale (SWLS) for life satisfaction. Treatment satisfaction was also assessed using the Client Satisfaction Questionnaire-8 (CSQ-8).

Results

The findings demonstrated statistically significant improvements in the CCBT + TAU group compared with TAU alone. PHQ-9 scores decreased over time in both treatment groups. However, greater improvements were observed in the CCBT group after treatment (mean difference = −2.5; 95% CI, −4.5 to −0.9; p = .005; Cohen's d = −0.42). These differences were maintained at the 3-month (mean difference = −2.3) and 6-month (mean difference = −3.2) follow-ups.

There were also statistically significant differences favoring CCBT over TAU on the GAD-7, ATQ, and SWLS measures at all assessment points, except for the GAD-7 at the 6-month follow-up.

Importantly, the effect size of CCBT over TAU was moderate (Cohen's d = −0.42) at posttreatment. Response and remission rates were also significantly higher in the CCBT group. At the end of treatment, 58.4% of the CCBT group showed a treatment response compared with 33.1% of the TAU group. Remission rates were 27.3% and 12.0%, respectively.

The results also indicated that completing more GDA modules was associated with greater improvement in depressive symptoms. Furthermore, patients receiving CCBT reported higher treatment satisfaction than those receiving TAU.

Discussion

These results provide evidence that computer-assisted CBT with clinician support can be an effective and potentially scalable option for treating depression in primary care. The findings support the hypothesis that guided digital interventions can improve treatment outcomes while reducing the amount of direct clinician time required.

The researchers reported that participants received an average of less than three hours of clinician support over the entire treatment period. This suggests that CCBT may offer an efficient method of delivering evidence-based psychological care. However, the study did not directly establish its cost-effectiveness.

Wright et al. highlight that one of the major barriers to effective mental health care is limited access to evidence-based psychotherapy. By implementing CCBT within primary care, healthcare systems may be able to provide more accessible treatment to patients who would otherwise face barriers to psychological care.

An important strength of the study was its inclusion of participants with lower incomes, limited education, and restricted internet access. This broadens the relevance of the findings to populations that have often been excluded from previous studies of digital mental health interventions.

However, the study also has several limitations. First, the sample was predominantly female, and participants were recruited from a limited number of primary care practices in Kentucky, which may restrict generalizability to other populations and healthcare settings. Moreover, the follow-up period of six months after treatment completion may not fully capture longer-term outcomes.

Another important limitation is the lack of an active comparison group receiving traditional face-to-face CBT or an equivalent amount of clinician support without the computer program. Therefore, the study cannot establish whether CCBT is equally effective as conventional CBT or determine how much of the observed improvement was attributable specifically to the digital component.

Additionally, treatment as usual was not standardized, and missing follow-up data may have influenced the results. Although CCBT significantly reduced depressive symptoms, a substantial proportion of participants did not achieve remission, indicating that further treatment or additional support may still be necessary.

Conclusion

Overall, the findings suggest that clinician-supported CCBT can be a valuable addition to depression treatment in primary care. The intervention produced greater reductions in depressive symptoms and higher remission rates than treatment as usual alone, while requiring relatively limited clinician contact.

Although further research is needed to clarify its cost-effectiveness and long-term outcomes, the study highlights the potential of digital psychological interventions to improve access to evidence-based care, particularly for patients who face financial, educational, or technological barriers to treatment.