Depression is the leading cause of disability worldwide [1], and those diagnosed with major depressive disorder (MDD) are up to 7.6 times more likely to die by suicide compared to the general population [2]. Depressive symptoms are associated with severe and chronic impairment in occupational and social domains [3], decreased quality of life [4], and adverse physical health outcomes [5]. Depression is a significant problem for US military veterans, with 13% of post-9/11 war veterans meeting criteria for current MDD [6] and 46.5% of female veterans and 36.3% of male veterans meeting criteria for lifetime MDD [7]. The high prevalence of depressive symptoms in veterans not only has negative ramifications for social and interpersonal functioning, occupational participation, and quality of life [8] but also markedly increases the risk for suicide. On average, 543 veterans die by suicide each month [9], and MDD is the second most important predictor of veteran suicide among all recorded mental health diagnoses [10]. Depressive comorbidity among veterans with posttraumatic stress disorder (PTSD) compared to PTSD alone is also associated with elevated rates of disability in cognitive (Cohen d=1.03) and mental health (Cohen d=1.49) domains. Veterans with comorbid PTSD and MDD also reported decreases in quality of life (Cohen d=0.84) and a 2-fold risk of suicide compared to those with PTSD alone [11]. Given the functional impairments and elevated suicidal risk associated with depressive symptoms, it is essential to provide veterans with timely and appropriate intervention.

Addressing depression before the symptoms become severe and entrenched could substantially improve outcomes. Even mild to moderate and subthreshold depressive symptoms are associated with significant disease burden, including increases in disability and mortality and decreases in quality of life [12]. The rates of transition to MDD for individuals with subthreshold depressive symptoms are exceedingly high both in the short term [13] and long term [14]. When subthreshold symptoms progress to MDD, symptoms become chronic in the vast majority of cases, with 85% of individuals with MDD experiencing recurrent episodes even after receiving specialty mental health treatment [15]. Evidence suggests that interventions targeted at individuals with mild to moderate symptoms, including those with subthreshold symptoms of MDD, are effective in restoring functioning and preventing the progression to severe disease [16].

Although depressive symptoms often go undetected at mild to moderate or subthreshold levels in community settings [17], depression symptom screenings are routinely performed on a population level at US Department of Veterans Affairs (VA) facilities. This provides a key opportunity to identify veterans with depressive symptoms, even before these symptoms reach the diagnostic threshold, allowing for rapid and high-quality intervention. Patient preference studies indicate that individuals with depressive symptoms prefer to receive psychotherapy over medication [18]. Although the VA has made significant strides in improving access to psychotherapy, including through the expansion of primary care–mental health integration [19], many veterans still struggle to access adequate psychotherapy. Research shows that the majority of veterans (61%) with diagnosed mental health disorders only attend 1 session of psychotherapy [20]. Shame and stigma around depressive symptoms and seeking mental health treatment, as well as difficulties attending mental health treatments due to a lack of time or transportation, are known barriers to adequate and timely intervention and care [21]. Although the expansion of virtual mental health care (ie, telehealth) within VA facilities is effective at overcoming certain logistical barriers to treatment, telehealth is as resource intensive as traditional face-to-face psychotherapy, and access is dependent on provider availability. Furthermore, VA telehealth services might be less appropriate for individuals with mild to moderate symptoms, whose symptom severity level might not merit individual psychotherapy, but who would nonetheless benefit from help. Thus, despite increased detection of depressive symptoms and an identified need for intervention, a significant treatment gap exists.

Internet-delivered interventions are well positioned to fill this treatment gap because these interventions are low cost, easily accessible, and scalable. When compared to telehealth or traditional face-to-face mental health care, internet-delivered interventions could alleviate veterans’ stigma concerns and provide increased flexibility of access in the presence of competing demands of work and childcare. Positive experiences with internet-delivered interventions could also facilitate the eventual transition to traditional mental health care if needed [22]. Meta-analyses of results from community samples provide evidence for the effectiveness of internet-delivered interventions for depressive symptoms, including acceptance and commitment therapy [23], cognitive behavioral therapy [24,25], and mindfulness-based approaches [26].

Deprexis (GAIA AG) [27,28] is an internet-delivered intervention that integrates acceptance and commitment therapy, cognitive behavioral therapy, and psychodynamic and mindfulness-based approaches. The program consists of 10 core modules targeting depressive symptoms and associated functional impairment. Engagement with content is completely self-guided, and the program is accessed remotely. A randomized controlled trial (RCT) of Deprexis in a general population sample found that it was effective in reducing depressive symptoms, improving well-being, and decreasing disability [29]. Despite these positive results in the general population, it is unknown whether Deprexis is an effective treatment for veterans with depressive symptoms, given the unique composition of the veteran population in terms of gender ratio, barriers to treatment, the severity of comorbid conditions, and sociodemographic characteristics. Furthermore, there is a need to understand whether this low-intensity, internet-delivered treatment for depression also addresses functional impairments in veterans. A critical gap in the field is testing the effects of evidence-based mental health interventions on recovery outcomes. The primary goal in previous trials was to test the effect of Deprexis on depressive symptoms. Although improvements in functional domains after Deprexis treatment have been reported [29], the conducted assessments did not span the full range of functional outcomes. There is a need to centralize the assessment of psychosocial functioning in mental health delivery research generally and the study of Deprexis specifically.

Deprexis has not yet been rigorously evaluated with a veteran population in a VA medical center setting. This study aims to fill this critical gap in the literature by testing whether Deprexis is acceptable and effectively decreases depressive symptoms and improves functional outcomes in veterans (N=152) presenting for VA health care with mild to moderate depressive symptoms. The objective of aim 1 is to assess perceptions, needs, and preferences as they relate to Deprexis through conducting posttreatment assessment interviews with veterans (n=16-20) and identify the components that improve veterans’ experience and uptake of Deprexis. The results obtained in aim 1 will be used to inform the rollout of RCT procedures under aim 2. Aim 2a involves completing an RCT comparing an 8-week course of Deprexis to a treatment-as-usual (TAU) control condition. We hypothesize that veterans engaged in Deprexis will show improvements on measures of functioning and decreases in depressive symptoms compared to the TAU control group. To generate hypotheses for future study, the objective of exploratory aim 2b will be to examine whether demographic variables, baseline psychopathology, credibility, and Deprexis use moderate treatment effects on the primary outcomes.

This study is an individually randomized, parallel, 2-group therapy trial. A total of 132 veterans will be randomized in a 1:1 fashion to either a Deprexis intervention or a TAU control group.

Recruitment will take place within the VA Central Texas Healthcare System. All assessment and intervention components will be completed remotely through Qualtrics (Qualtrics International Inc) [30] and the Deprexis website, respectively.

Potential participants include all veterans irrespective of sex or gender and of all ages and races and ethnicities who (1) are able to comprehend and sign the informed consent form; (2) have reliable access to the internet and a computer, tablet device, or smartphone; (3) exhibit mild to moderate (not very severe) levels of depression (Quick Inventory of Depressive Symptomatology–Self-Report [QIDS-SR] [31] score between 6 and 20 at the time of screening); and (4) are stable on psychotropic medications (defined as no medication change in the 30 days before study entry). The last criterion is included to ensure that the symptoms evaluated during the pretreatment assessment are due to underlying psychiatric conditions and not because of starting or stopping medications. Veterans will be excluded from study participation if they (1) endorse any positive symptoms of a psychotic disorder (assessed via the Psychiatric Diagnostic Screening Questionnaire [PDSQ] [32]), (2) screen positive for bipolar I disorder (assessed via the Mood Disorder Questionnaire [33]), or (3) report current suicidal risk (assessed via Beck Depression Inventory–Second Edition [BDI-II] item 9 [34]).

For this study, veterans presenting with mild to moderate symptoms of depression will be recruited. Recruitment letters, including a link to the Qualtrics survey, will be sent out by US mail to veterans identified through the medical record who screen positive on the Patient Health Questionnaire (PHQ)-2 clinical reminder screen (a score of ≥3). The initial eligibility criterion is purposefully broad (ie, a positive PHQ-2 screen vs a positive PHQ-9 screen) because we do not want to exclude treatment-receptive veterans with mild depressive symptoms. Veterans’ telephone numbers will be obtained from the medical record, and a research technician will contact veterans by telephone to ensure that they have received their letters, answer any questions, and provide them with a link to the Qualtrics survey if needed. Flyers at VA primary care and medical specialty clinics will also be posted and distributed, and information about the study and participant inclusion and exclusion criteria will be provided to VA primary care providers.

Deprexis is an internet-delivered treatment for depressive symptoms and related functional impairment. The intervention draws from various theoretical frameworks and consists of 12 modules: 10 core content modules, an introductory module, and a summary module. Deprexis, which is designed to be interactive involves answering questions and learning techniques and concepts through instruction and examples. Deprexis uses a software technology (Broca) that tailors content dynamically to the users’ responses, resulting in a simulated conversational flow. To increase motivation, Deprexis features symptom tracking, with accompanying graphical and text feedback, worksheets and summaries in printable format, audio recordings, and illustrations. The content focuses on improving depressive symptoms (eg, cognitive restructuring of depressogenic thoughts) and decreasing functional deficits associated with depressive symptoms (eg, behavioral activation, interpersonal skills, and healthy lifestyle choices). The program is self-paced, and every module takes between 10 and 60 minutes to complete. Participants are provided a free Deprexis voucher and will access the intervention through the Deprexis website. Textbox 1 [27] presents an overview of session content, while Figure 1 depicts a Deprexis sample presentation.

Veterans randomized to the control condition will have access to standard VA treatment resources, including mental health treatment (they will be provided with a list of available resources).

Veterans can choose to discontinue study involvement at any time without facing any consequences. We will actively monitor for any serious adverse events occurring during their study involvement. Modifications to intervention components or procedures may be made at the discretion of the principal investigator (PI; RP), pending approval from the institutional review board (IRB) and other relevant stakeholders.

Adherence will be optimized using strategies from previous Deprexis trials. All veterans will be sent brief SMS text messages to increase engagement. These SMS text messages will target behavioral and cognitive processes that are known to maintain depressive symptoms and will include reminders to use the Deprexis program. Research staff will also have access to veterans’ real-time Deprexis use data, and based on this use data, short encouraging emails will be sent to them by research staff. These emails will range from acknowledging the veterans’ use of the Deprexis program to encouraging further engagement and addressing feelings of discouragement. Research staff will also reach out to veterans who have not activated their free Deprexis vouchers to provide technical support if needed. Study staff will initiate contact with veterans if there is no engagement with Deprexis over a 2-week period.

To participate, veterans need to be stable (no changes in the last 30 days) on medications prescribed for a mental health disorder. Veterans randomized to either treatment group are allowed to start or continue any nonstudy treatment resources, including mental health treatment, during their participation in the trial.

Figure 2 shows the participant timeline in detail.

In aim 1, qualitative veteran feedback on the Deprexis intervention and study procedures will be obtained; no quantitative analyses will be completed. Broadly, the posttreatment interview questions will cover the following content areas: (1) motivation toward, and barriers to, treatment and participation in the study; (2) Deprexis content, presentation, adherence, and usability; (3) perceptions of study procedures; and (4) changes experienced in mood, behavioral engagement, or interpersonal interactions after Deprexis treatment.

In aim 2a, we will test whether Deprexis is effective for decreasing depressive symptoms and improving functional outcomes in veterans presenting for VA health care with mild to moderate depressive symptoms. The primary outcomes for this aim are depressive symptoms (assessed with the QIDS-SR-16) and functioning (assessed with the World Health Organization Disability Assessment Schedule II [WHODAS-II] and the Sheehan Disability Scale [SDS]). The secondary outcomes are other psychopathology (assessed with the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition [PCL-5] and the PDSQ), social functioning (assessed with the Interpersonal Support Evaluation List–Short Form [ISEL-SF] and the Inventory of Interpersonal Problems–Short Circumplex Form [IIP-SC]), and quality of life (assessed with the Quality of Life Scale [QLS]).

Exploratory analyses in aim 2b will examine whether demographic variables (sex, gender, age, race, ethnicity, and education), PTSD symptom severity (assessed with the PCL-5), comorbidity (assessed with the PDSQ), treatment credibility (assessed with the Credibility and Expectancy Questionnaire), frequency of study support, and Deprexis use moderate treatment effects.

The following primary outcomes (assessed at all time points) are administered through Qualtrics:

The QIDS-SR-16 [31] is a 16-item self-report measure of depressive symptom severity. This measure takes 5 to 7 minutes to complete and assesses the 9 Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-V), symptom criterion domains for depression (eg, “There is no change in my usual appetite”) and has been shown to be highly internally consistent (Cronbach α=0.86). The QIDS-SR-16 demonstrates concurrent validity with measures assessing similar outcomes (r=0.86-0.65) and is sensitive to symptom change [31]. Scores of >6 are indicative of meaningful depressive symptoms. The QIDS-SR-16 will serve as a primary eligibility criterion, specifically a QIDS-SR-16 score between 6 and 20. To verify the duration of depressive symptoms, we will include an item adapted from the Clinical Interview Schedule–Revised, which will assess the duration of depressive symptoms on a 6-point scale.

The WHODAS-II [35] is a 36-item self-report questionnaire, which takes 5 to 20 minutes to complete. The measure assesses functional disability across 7 domains: understanding and communicating, getting around, getting along with people, life activities, work, participation in society, and self-care (eg, “In the last 30 days, how much difficulty did you have in concentrating and doing something for 10 minutes?”). The WHODAS-II total score is scaled from 0 to 100, with higher scores indicating greater disability. The WHODAS-II has high test-retest reliability (r=0.98) and moderate to good concurrent validity with measures assessing similar outcomes (r=0.45-0.65), and it is sensitive to change in functional disability after treatment [35].

The SDS [36] is a 3-item self-report measure of symptom-related disability (eg, “The symptoms have disrupted your work/school life”) that has been used in previous Deprexis trials. The SDS takes approximately 5 minutes to complete, and items are summed (range 0-30), with higher scores indicating greater disability. The SDS was developed as a global measure of the impact of mental illness on work and school activities, family relationships, and social functioning. The SDS is highly internally consistent (Cronbach α=0.89) and has good reliability (r=0.73and high construct validity [37]; in addition, it is sensitive to treatment effects [38].

The secondary outcomes listed in the following sections are assessed at all time points.

The PDSQ [32] is a 125-item questionnaire containing 13 subscales (eg, MDD, anxiety disorders, and alcohol and drug abuse and dependence) that is used to screen for symptoms of mental health disorders. The PDSQ takes 15 to 20 minutes to complete, with scoring and cutoffs varying by individual scale [34]. This scale shows high internal consistency (Cronbach α=0.82); high test-retest reliability (r=0.84); and good discriminant, convergent (mean r=0.17), and concurrent (mean r=0.72) validity [32].

The BDI-II [34] is a 21-item self-report measure assessing depression symptom severity, completed in approximately 10 minutes. Items are scored on a 4-point scale (range 0-63), with higher scores reflecting increased endorsement of depressive symptoms. A score of ≥15 indicates the presence of clinically significant depressive symptoms [39]. The BDI-II has high internal consistency (Cronbach α=0.90), good to high test-retest reliability (r=0.73-0.96), and it correlates highly with interview-based measures of depression (r=0.66-0.75) [40]. The BDI-II will be used as a secondary measure of depression and to assess the suicidal ideation exclusion criteria.

The IIP-SC [41] is a 32-item self-report measure assessing the interference of depressive thoughts, behaviors, and symptoms across various psychosocial domains (eg, “It is hard for me to join in on groups”). The measure takes 10 to 15 minutes to complete. It consists of 8 scales consisting of 4 items each. Scores are averaged (range 0-4), with higher scores indicating greater interpersonal problems. The IIP-SC is sensitive to treatment effects and has demonstrated good to high internal consistency (Cronbach α=0.64-0.87) and good to high test-retest reliability (r=0.56-0.81) [42].

The ISEL-SF [43] is a 12-item self-report measure assessing perceived social support across four subscales: (1) appraisal support (the perceived availability of someone to discuss issues of personal importance), (2) tangible assets support (the perceived availability of material aid), (3) belonging support (the perceived availability of others to interact with socially), and (4) self-esteem support (the perceived availability of others with whom one compares favorably). The measure takes approximately 5 minutes to complete, with higher scores indicating the presence of increased social support. There is evidence that perceived social support improves after internet-delivered interventions for depression [44]. The ISEL-SF has been found to be highly internally consistent (Cronbach α=0.83) and has demonstrated moderate convergent construct validity with other measures of social support (r=0.45) [45].

The QLS [46] is a 16-item self-report measure of quality of life across various domains, including material and physical well-being; relationships; social, community, recreational, and civic activities; personal development; and fulfillment and independence. Ratings for domains (eg, “Health, being physically fit, and vigorous”) are assigned on a 7-point scale from “delighted” to “terrible” (range 12-116), with higher scores indicating a better quality of life. The measure takes approximately 10 minutes to complete. The QLS has high to excellent internal consistency (Cronbach α=0.82-0.92) and high test-retest reliability (r=0.78-0.84), and it has demonstrated good convergent validity with similar measures (r=0.67-0.75) [47]. Furthermore, the QLS is able to discriminate between populations with expected differences in quality of life (ie, those with chronic illness and those without) [48].

The PCL-5 [49] will be used to identify whether PTSD is a potential moderator of symptom change and to examine whether Deprexis produces meaningful changes in PTSD symptoms. The version of the PCL-5 used (PCL-5 with Criterion A) will include an assessment of the presence of a traumatic event meeting criterion A (the first criterion for diagnosing PTSD according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition). If no traumatic event is endorsed, veterans will not complete the remaining PCL-5 questions. Veterans who endorse exposure to a criterion A traumatic event will complete the 20-item PCL-5, which assesses PTSD symptom severity on a 5-point scale (eg, “In the past month, how much were you bothered by ‘Repeated, disturbing, and unwanted memories of the stressful experience?’”). Scores range from 0 to 80, with scores of >31 being indicative of probable PTSD (ref). The measure takes approximately 10 minutes to complete. The measure has excellent internal consistency (Cronbach α=0.94), high test-retest reliability (r=0.82) [49], and moderate convergent and discriminant validity (r=0.31-0.60) [49]. The PCL-5 is as sensitive to clinical change that occurs between pre- and posttreatment assessments as gold standard interview-based measures of PTSD symptoms (alerting convergent validity: r=0.94 and contrast converting validity: r=0.92) [50].

The following moderators and covariates are measured at baseline:

Several manuscripts are planned to align with the various study aims. In addition, results will be presented at research conferences, and summaries will be drafted for media releases. Author eligibility and order will be determined by the study PI.

The protocol has been approved by the Central Texas Veterans Affairs Healthcare System IRB (protocol 2024-001). Any changes to the study protocol will be promptly communicated to relevant stakeholders.

A waiver of informed consent will be obtained for the completion of the eligibility screen. Informed consent will be obtained for the remainder of the study procedures. A watermarked paper copy of the informed consent will be included with the study recruitment letter. An electronic version of the consent form will be included in the Qualtrics eligibility survey. Veterans will be asked to review the consent form, and they will be asked whether they have any questions or concerns. Veterans who indicate that they have questions or concerns will be contacted by telephone by the research technician. Once all questions and concerns are responded to, the research technician will verbally ask whether the veteran is still interested in the study and if the veteran answers in the affirmative will send a link to DocuSign consent via encrypted email. Veterans who indicate that they do not have questions or concerns will be asked in the Qualtrics eligibility screen whether they are interested in participating in the study. If they answer in the affirmative, they will be sent a link to DocuSignconsent via encrypted email.

No participant data will be released in any form unless the investigators or study personnel are mandated to do so by law (eg, because of risk to self or others). All study data will be kept in accordance with VA guidelines, which require that electronic and paper data be kept for the duration of the study and for 6 fiscal years thereafter (for details on data storage and disposition, refer to VA Records Control Schedule 10-1 [69]). Participants will be assigned a study ID, which is linked to identifying information in a password-protected spreadsheet, which is stored behind the VA firewall. Study personnel who have access to study data will receive training in confidentiality policies and procedures.

All IRB-approved study team members may have access to participant-level data pending approval from the study PI. The full protocol will be available at the discretion of the study PI.

Research examining computer-delivered interventions in veterans lags behind that conducted in general population samples. This study will gather the qualitative experiences of veterans with a computerized intervention and will be the first RCT of a self-guided, computerized intervention for depressive symptoms and functional impairment in veterans. The results from the described study will demonstrate whether the positive results in general population trials of computerized interventions generalize to the veteran population. Furthermore, we broaden the target population of these types of interventions by including veterans with subthreshold symptoms of depression. The described trial has an explicit focus on functional outcomes because there is evidence that symptomatic recovery and functional improvements do not consistently improve in tandem after mental health interventions. Deprexis has previously been associated with significant improvement in disability and well-being; however, these were secondary outcomes. This study centralizes and significantly expands the assessment of psychosocial functioning across domains.

Depressive symptoms, independent of comorbid trauma symptoms, are an important driver of functional impairment and suicidality in veterans. When depressive symptoms co-occur with PTSD, these symptoms often persist after trauma-focused treatments. Thus, depressive symptoms should be an urgent target of intervention in veterans, even in the context of complex comorbid conditions. There is evidence that the depression treatment gap commonly described in community samples also exists for veterans who receive their care at VA facilities [20]. Various reasons for the existence of this treatment gap have been suggested, and it seems likely that multiple individual, logistical, and institutional barriers may prevent veterans with depressive symptoms from receiving appropriate care. Cost-effective, accessible computerized interventions would broaden the reach of depressive treatments available within the VA system.

Importantly, the VA system is an appropriate site for testing an intervention such as Deprexis. The broad screening for depressive symptoms that occurs at all VA clinics (ie, outside of specialty mental health clinics) and the integrated mental health record allow for the timely identification of veterans presenting with relatively mild depressive symptoms. There is an opportunity to prevent chronicity and disease progression by focusing treatment on mild to moderate depressive symptoms. There is evidence that Deprexis may be especially effective for individuals with lower depressive symptom severity [53]. Our long-term vision is that brief, computerized interventions such as Deprexis can be integrated in a stepped-care approach, providing rapid intervention for veterans with mild to moderate depressive symptoms, while reserving resource-intensive treatments such as face-to-face psychotherapy for individuals with severe symptomatology.

Alongside the aforementioned strengths, this study has several limitations. The choice of a TAU control group is a relatively weak comparator. Follow-up studies comparing Deprexis to active treatment conditions would provide further evidence for the relative impact of the intervention. All assessments administered in this study are self-report questionnaires, which are prone to certain types of bias (eg, desirability bias). If the efficacy of Deprexis is supported in this trial, examining whether change also generalizes to other assessments, such as ecological momentary assessment or tracking, may be merited. Relatedly, because no clinical interviews are conducted, we are not able to assess the presence or absence of clinical diagnoses. Finally, use will be closely monitored in this trial, which may not be feasible in clinical settings. Automating use reminders may be an important step to promote successful uptake in the VA system.

Despite the aforementioned limitations, we believe that the described study will provide much needed initial evidence for the effectiveness of computerized interventions broadly and Deprexis specifically for veterans presenting with depressive symptoms and functional impairment. The results obtained in this study could inform the rollout of targeted computer-delivered interventions that address common barriers to mental health care in veterans. This is especially pertinent given that demand for VA behavioral health care outpaces supply, resulting in increased outsourcing of behavioral health care in the form of community care referrals [70]. Building a diverse arsenal of psychosocial interventions, including computerized interventions, potentially allows more veterans to receive behavioral health care within the VA system. Furthermore, giving veterans additional mental health care options could improve overall engagement with treatment [71]. We believe that the results of this study could potentially extend and advance routine care for veterans with depressive symptoms, which is an essential step toward closing the treatment gap and providing the highest quality care for veterans with complex mental health needs.