This remotely delivered randomized clinical trial, conducted in a community health care system in the Metro-North Boston region during the COVID-19 pandemic, compared the effectiveness of MBCT-R supported by CHA-MW (MBCT-R+CHA-MW) with CHA-MW alone on change in depression symptoms over 24 weeks. As a secondary analysis, the study investigated the effectiveness of iCBT+CHA-MW versus CHA-MW alone on change in depression symptoms. Within- and between-group effects for depression were assessed and compared at 24 weeks. As a secondary outcome, participants’ behavioral health visit use during the 24-week study period was analyzed to investigate between-group differences as well as sensitivity analyses to measure the potential effects of visit use on the measured mental health outcomes.

This study was conducted in accordance with the ethical principles of the World Medical Association’s Declaration of Helsinki, as revised in 2024. All participants provided written informed consent before participating in the study. The CHA Institutional Review Board (reference number: Legacy-CHA-IRB-1141-05-20) approved the study, and the rights, privacy, and confidentiality of all participants were protected throughout the study. This study was preregistered on ClinicalTrials.gov (NCT04595084) and monitored by the National Center for Complementary and Integrative Health (NCCIH) and a NCCIH-approved Independent Monitoring Committee [48]. Participants received financial compensation, up to US $160, for their participation. Payment was given at 5 timepoints, including US $20 after the screening or consent visit, US $20 after completion of baseline surveys, US $60 after the week 12 assessment, US $40 after the week 24 assessment, and US $20 for a completion bonus. Upon completion, participants could also become eligible for an optional substudy on stress reactivity (not reported here), which included an additional 3 payment timepoints: US $20 for a second screening or consent following main study completion, US $20 to US $40 for saliva collection, and US $20 to US $40 for pre– and/or post–daily diaries.

Participants were recruited through 2 main methods. First, emails were sent to all adult primary care patients in the health care system inviting them to participate in CHA-MW [21]. Second, primary care providers referred patients with anxiety, depression, or COVID-19–related distress to CHA-MW. Once individuals registered for CHA-MW, the CHA-MW team automatically invited them to complete the first CAT-MH assessment. Those who met preliminary study inclusion criteria via CAT-MH symptom severity scores were eligible and invited by the CHA-MW team to learn more about joining this study. Study research coordinators screened all patients who expressed interest in joining the study for eligibility through a standardized process that included telephone interviews, self-report surveys, and electronic health record review. Inclusion criteria required participants to be current CHA patients aged between 18 and 70 years with English proficiency, able to participate in teleconference meetings, and having a depression score of 50‐75 measured via CAT-MH Depression module (CAT-DEP) [49], representing mild to moderately severe depression symptoms. Exclusion criteria included psychosis (CAT-Psychosis>60) [50], history of bipolar I disorder or current mania (CAT-Mania>70) [51], acute suicidality (CAT-Suicide ≥71) [52], severe depression (CAT-DEP>75) [49,51], severe PTSD (CAT-PTSD>75) [51], severe substance use disorder or positive toxicology for cocaine, unprescribed opioids, stimulants, or benzodiazepines in the 3 months before screening, current treatment with antipsychotic medication, mood stabilizer or benzodiazepine equivalent of 3 mg per day of lorazepam, current participation in another experimental research study, mindfulness group experience within the past year, or anticipated hospitalization or incarceration in the next 6 months. Eligible participants reviewed informed consent via REDCap (Research Electronic Data Capture) [53] and completed a consent assessment to ensure that they understood key aspects of study procedures. Those who scored greater than 90% completed the consent process and were able to complete baseline surveys. Study reporting followed the CONSORT (Consolidated Standards of Reporting Trials). The trial protocol is provided in Multimedia Appendix 2.

Upon completion of the baseline survey battery, participants were randomized in randomly ordered blocks of 5 or 10 with a 2:2:1 ratio to MBCT-R+CHA-MW, iCBT+CHA-MW, or CHA-MW according to a computer-generated sequence (STATA random generator). The study methodologist (GP) provided randomization lists to the study senior research coordinator (AC) and/or data analyst (JB), all of whom were blinded to participant identity prior to and during randomization. When a participant was eligible, research coordinators informed the senior research coordinator, data analyst, or the methodologist who, in turn, provided the randomization status, as research coordinators themselves did not have access to source the blinded randomization sequence. A research coordinator then called the participant by phone to inform them of the process and timing for the MBCT-R or iCBT program, as well as the process for ongoing CHA-MW surveys.

Participants were not blinded to their assigned study arm, nor were the coprimary investigator (CF), research coordinators, data analyst, or methodologists after randomization took place. Finally, the contact primary investigator (ZS-O) was blinded to the assigned study arm when reviewing data quality, adverse events, and protocol violations, and was not fully unblinded until all primary outcome data were collected and the database was locked.

All interventions were delivered between baseline and study week 8. CHA-MW monitoring and support was provided to all participants, while the other 2 interventions were added to CHA-MW.

Staff monitored adverse events at each study visit and through REDCap surveys administered at weeks 8, 16, and 24. Events were assessed for severity, relatedness, and expectedness, and were regularly reviewed by a data safety and monitoring board approved by the NCCIH.

Randomization was assessed by comparing the baseline characteristics of participants assigned to all 3 intervention arms using t tests, chi-square tests, or Fisher exact tests. For the primary outcome regarding depression, we conducted an intention-to-treat, repeated-measures analysis using generalized linear mixed-effects models (mixed) with a week by study group interaction term to estimate differences between MBCT-R+CHA-MW vs CHA-MW in depression severity (CAT-DEP) from baseline to 24-week follow-up. As a secondary outcome, we also examined the change in depression from baseline to week 24 comparing iCBT+CHA-MW with CHA-MW.

For the completer analysis, using the a priori per protocol completer definition (completion of at least 6 sessions of either MBCT-R or iCBT), we estimated the difference between participants who did and did not complete their allocated intervention within the MBCT-R+CHA-MW and iCBT+CHA-MW arms over time by adding a 3-way interaction term representing study week, group, and completer status.

For the secondary outcome of mental health clinician visit utilization during the 24-week study period, we used logistic regression to compare the proportion of participants with at least 1 mental health clinician visit across study groups. We also conducted post hoc moderation analyses to investigate potential impacts of differential visit usage on both primary and exploratory mental health outcomes.

Finally, 7 exploratory mediators of change were analyzed, chosen based on published studies of MBCT, CBT, and other related interventions, including anxiety, perceived stress, interoceptive awareness, experiential avoidance, emotional dysregulation, self-compassion, and decentering. We conducted a difference-in-differences intention-to-treat repeated-measures analysis using linear mixed-effects models with a study week by group interaction term to estimate the relative changes from baseline to week 24.

Assuming a 2-sided significance level α=.05, we calculated using Stata (version 16; StataCorp LLC) [74] that a sample of 200 participants randomized 2:2:1 would have provided 80% power to detect an effect size of d=−0.54 for between-group differences on the primary aim comparison of changes in depression severity (CAT-DEP) when comparing the MBCT-R+CHA-MW vs CHA-MW arms. Maximum likelihood estimation (MLE) was used to address missingness for all analyses. We also conducted multiple imputation for a sensitivity analysis, with comparable results, and so the former are reported here. We planned to adjust the models to account for baseline covariates that differed between study groups after randomization (P<.10). We accounted for multiple comparisons using the Benjamini-Hochberg false discovery rate procedure [75]. Effect sizes (Cohen d) were calculated. For the number of adverse events, we conducted a negative binomial regression to evaluate between-group differences. All analyses were conducted in Stata (version 18; StataCorp LLC) [76].

From March 19, 2021, to September 2, 2022, a total of 293 people were referred to the study, 113 completed informed consent and enrolled, and 97 were randomized to 1 of the 3 study arms (MBCT-R+CHA-MW: n=37; iCBT+CHA-MW: n=41; and CHA-MW: n=19) (Figure 2). Randomized participants (N=97) were 73% (71/97) female and, on average, aged 40.0 (SD 11.9) years, with 76% (74/97) identifying as White and 28% (27/97) identifying as American Indian or Alaska Native, Asian, Black, Haitian, or African American, or Other. Participants could endorse more than 1 demographic for race. No significant differences in baseline demographics were found between the 3 study arms (Table 2). Assessment completion between study arms was similar in all 3 arms with no difference between groups in the rates of participants who completed all 4 main assessment sessions: 67% (25/37) of the MBCT-R+CHA-MW arm, 68% (28/41) of the iCBT+CHA-MW arm, and 73% (14/19) of the CHA-MW arm.

Baseline differences in primary and secondary outcome variables necessitated running primary and secondary analyses with baseline values as covariates. For the primary analysis (1^o: MBCT-R+CHA-MW vs CHA-MW), baseline values of decentering (secondary outcome) were significantly lower in the CHA-MW arm than in the MBCT-R+CHA-MW arm (P=.04) (Table 3), so subsequent analyses were run with baseline decentering values as a covariate. For the secondary analysis (2^o: iCBT +CHA-MW vs CHA-MW), baseline values of both decentering and depression were lower in the iCBT +CHA-MW arm than in the CHA-MW arm (P=.02, P=.04). Therefore, these secondary analyses were run with both baseline depression and baseline decentering values as covariates alongside baseline levels of the measured outcome.

Interim power analyses indicated that for an anticipated sample size of 100 randomized participants (40:40:20) the study had 80% power to detect a difference of at least d=0.72 between MBCT-R+CHA-MW (n=40) and CHA-MW (n=20) and 80% power to detect a difference of at least d=0.63 between MBCT-R+CHA-MW (n=40) and iCBT (n=40).

Twenty adverse event reports (AERs) were completed during the study (10/37, 27% [MBCT-R] vs 10/41, 24% [iCBT] vs 0/19, 0% [CHA-MW]; P=.045). Analyses indicated significantly fewer AERs in the CHA-MW arm than in both the MBCT-R+CHA-MW and iCBT +CHA-MW arms. One possible explanation for this difference is that participants in the CHA-MW arm had fewer contact points with the study team, during which they might have volunteered or been asked about potentially adverse events. Of the AERs reported, 18 were not serious, with 2 being related to the protocol. One MBCT-R participant experienced back pain–related anxiety during meditation and elected to discontinue the intervention. One iCBT participant reported feeling “worse” when completing the online modules and elected to discontinue the intervention. Regarding the 2 serious adverse events reported, neither were related to the protocol. Two MBCT-R participants were hospitalized for medical concerns, a blood clot and a kidney concern, respectively; however, both elected to continue the intervention.

Contrary to study hypotheses, findings from this randomized controlled clinical trial did not demonstrate that MBCT-R+CHA-MW was more effective than CHA-MW alone in reducing depressive symptoms. Therefore, we did not find support for an additive effect of adding MBCT-R to CHA-MW. Furthermore, the magnitude of participants’ overall engagement with mental health clinician visits beyond their assigned intervention did not differ between groups, suggesting that study assignment did not impact overall mental health service clinical visit utilization significantly during the 24-week study period.

Low levels of intervention completion and study attrition during the COVID-19 pandemic may have limited the study’s ability to detect between-group differences. Most importantly, intervention arm participants had low rates of intervention completion (6+ sessions) in the completer analysis both for MBCT-R (30%) and for iCBT (24%), with an approximately 8-point difference in CAT-DEP score reduction between evidence-based intervention completers and noncompleters. This is a potential contributing factor for lack of superiority between either of the 2 additive evidence-based intervention arms versus the CHA-MW arm. The rate of completion was lower than expected for MBCT-R. As the live online, synchronous, and group-based option providing social connection, the MBCT-R+CHA-MW arm was anticipated to have a higher level of engagement with lower attrition rates than what occurred in the study [77]. While videoconference delivery of MBCT-R was anticipated to have greater attrition than what is typically seen in-person MBCT delivery (19%‐26%), MBCT-R was still expected only to have a noncompletion rate similar to the 50% seen for other digital mindfulness-based group interventions [77,78] and not the 70% noncompletion rate found in this study. Notably, in comparison, the 24% completion rate for MoodGYM was not dissimilar to published trials for iCBT [58]. Given the significant difference in depression reduction between completers and noncompleters, a larger future study with higher evidence-based intervention completion rates may have the potential to demonstrate an additive effect that was not seen in this study. Notably, during the COVID-19 pandemic, digital and mental health programs experienced high rates of dropout and nonadherence, which was due to systemic and technological barriers [79,80], personal (eg, work or time conflicts, infection with COVID-19, etc) and psychological constraints (eg, pandemic distress, depression, and anxiety) [79,81], and baseline high attrition among diverse populations [82]. Both MBCT-R+CHA-MW and iCBT+CHA-MW were higher-intensity interventions than CHA-MW alone, which may have put greater demands on participant resources (eg, time willing to be online and difficulty having privacy to engage in the program), that were already strained for many people during the pandemic [83,84]. Finally, only approximately 69% (67/97) of participants completed all 4 CAT-DEP assessment timepoints, resulting in an acceptable, but not ideal, level of missing data that increased standard errors for estimating the primary outcome in both the intention-to-treat and completer analyses.

Several other additional crucial factors impacted the study’s ability to detect group differences. First, despite adequate randomization for the primary comparison, there were differing levels of depression symptom severity at baseline in the secondary analysis for iCBT+CHA-MW versus CHA-MW. Although baseline depression was added as a covariate in the mixed-effects regression, inadequate randomization is still a limitation of the secondary outcome analysis. Second, floor and ceiling effects may have had an impact, especially for iCBT in the secondary analysis, since this arm had a lower level of depression severity at baseline. Third, these differences also created the potential for bidirectional interactions between higher symptom severity or longer duration of depressive symptoms and treatment-seeking behaviors [85,86]. Fourth, on average, the study sample endorsed relatively lower baseline depression scores (mean 55.3, SD 11.9) that fell within the mild-moderate range for symptom classification (CAT-DEP scores=50‐65). This limited range may have made it difficult to detect significant between-group differences with this sample size. Finally, referral to psychopharmacology was part of CHA-MW’s stratified care algorithm, so those with new or worsening CAT-MH symptoms were referred to medication management. Since the CHA-MW arm had the largest increase in new psychopharmacology visits during the 24-week study period, this could have contributed to its effect; the low-dose comparator CHA-MW may have had more new referrals than the other arms to psychopharmacology to start an antidepressant medication trial, which is another well-proven evidence-based treatment for depression, and may account for some of the benefit of the CHA-MW approach.

When considered as a whole, results suggest the resource efficiency and relative effectiveness of CHA-MW as a stand-alone population mental health intervention, which pairs a low-intensity online psychoeducation community newsletter with a platform for remotely delivered, automated mental wellness monitoring and a responsive semiautomated, stratified care algorithm for referring people with worsening mental health symptoms to standard telemental health care. CHA-MW could be reasonably effective with or without additional remotely delivered mindfulness or CBT modules, which may be especially true during times with high-stress events (eg, COVID-19) when intervention adherence for both live online, synchronous, group-based interventions and asynchronous, individual, digital interventions may be more challenging for participants to complete.

For the primary outcome, results indicated significant reductions in depression across all treatment arms. All CAT-MH modules receive a T-score on a 100-point scale. Therefore, the posterior standard deviation (uncertainty) of the severity estimates for all CAT-MH modules (eg, CAT-DEP) is 5 points on a 100-point scale, so the clinically meaningful difference for a single person is 5 points. As such, a 5-point change is beyond what can be expected based on measurement error alone, and the within-group point reductions ranging from 12.7 to 15.2 across all interventions represented clinically meaningful reductions in depression symptoms. However, no significant between-group differences were found. MBCT-R+CHA-MW, iCBT+CHA-MW, and CHA-MW arms all demonstrated similar patterns of effect. This finding aligns with previous comparative research findings. MBCT is generally effective for reducing current depressive symptoms with comparable effect sizes as specific, active controls, such as CBT, depression education programs, or psychoeducation, at posttreatment and 6-month follow-up, but it is more effective than nonspecific controls, such as waitlist or treatment as usual, at posttreatment but not follow-up [36]. Consistent with this literature, this study found that MBCT-R+CHA-MW demonstrated similar effectiveness to iCBT+CHA-MW (the specific, active comparator); however, it also was similar in effectiveness to CHA-MW (the nonspecific, lower intensity control) at 24-week follow-up. While these findings ran contrary to study hypotheses, which predicted that MBCT-R+CHA-MW would be more effective than CHA-MW alone, it is notable that the nonspecific CHA-MW intervention had previously demonstrated a preliminary effect of moderate size on depression among people with clinically significant symptoms (d=−0.42) [21]. While this study was not designed as a noninferiority trial, present findings suggest a potential for comparative effectiveness of CHA-MW as a stand-alone intervention for depressive symptoms when compared with 2 more resource-intensive, active, and evidence-based treatments during stressful periods, such as the COVID-19 pandemic, when adherence barriers for completing additional interventions may be common.

Importantly, from a methodological perspective, the iCBT+CHA-MW versus CHA-MW comparison had a large difference in mean depression at baseline that was greater than the 5-point clinically meaningful difference. While analytic methods addressed this by including CAT-DEP baseline levels as a covariate in the secondary analysis, the difference at baseline in the iCBT arm limited the ability to draw firm conclusions from the secondary analysis.

Overall, no significant differences in overall mental health clinician visit use over the 24-week study period were found. Contrary to hypotheses, which predicted that mental health clinician visit use would decrease as intervention intensity increased, participants tended to have similar engagement with mental health clinician visits irrespective of their assigned intervention. This finding suggests that the interventions did not differentially impact visit use during the 24-week study period. Analysis of disaggregated visits (ie, telephone encounter, individual psychotherapy, individual psychopharmacology, and group psychotherapy) found a significant difference related to psychopharmacology visits, such that participants in the MBCT-R+CHA-MW arm were more likely to be engaged in psychopharmacology visits than those in the iCBT+CHA-MW arm, yet the majority of these participants in the MBCT-R+CHA-MW arm were already engaged with individual psychopharmacology visits prior to the study. Additionally, the largest increase in new psychopharmacology visits was in the CHA-MW arm (20% increase), which may have contributed to the depression reduction effect of CHA-MW, as previously noted. This may have happened in CHA-MW because the stratified care approach facilitated timely entrance to medication management for depression when appropriate. The brief 20-minute online video or phone visits, using a motivational interviewing approach, focused on providing a broad overview of mental wellness options available while also enhancing motivation for treatment and reducing stigma for psychopharmacologic options when appropriate. Notably, a subsequent sensitivity moderation analysis did not find that mental health visit use moderated primary or exploratory mental health outcomes. Regardless of their engagement with psychopharmacology visits, participants in both MBCT-R+CHA-MW and iCBT+CHA-MW arms demonstrated similar improvements in the measured mental health outcomes. This is consistent with past studies that found that preferences for medication did not reduce the effectiveness of MBCT [87].

Taken together, these findings suggest the relative usefulness of CHA-MW, particularly as it was the least demanding intervention of the 3 investigated in this study. However, it is also possible that intervention effects may not have translated into changes in mental health visit use during the 24-week study period. For instance, data for this study included only visits within our health care system; it is possible that participants sought psychotherapy elsewhere, or even that they may have been in therapy elsewhere and decided to stop due to improvement during the study. Follow-up studies in a larger sample with greater power, higher levels of intervention completion, and a longer follow-up period with collateral service data would help to render a clearer picture of how these interventions may differentially impact mental health visit use, which in turn would inform a greater understanding of their respective costs and benefits.

MBCT-R+CHA-MW did not demonstrate superiority compared with iCBT+CHA-MW or CHA-MW for reducing exploratory clinical outcomes of anxiety or stress. Significant within-arm improvements were found for 6 of the 7 exploratory mental health outcomes within each of the 3 study arms, including reduced anxiety, perceived stress, experiential avoidance, and emotional dysregulation, as well as increased interoceptive awareness and self-compassion. Additionally, for each of these mechanistic outcomes, no significant between-group differences were found in the magnitude of pre-post change or in the rate of change over the 24-week study period. MBCT-R+CHA-MW, iCBT+CHA-MW, and CHA-MW all demonstrated similar patterns of effect.

By contrast, decentering only had significant within-group differences from baseline to week 8 in the CHA-MW arm. Decentering, defined as the ability to observe one’s thoughts and feelings as transitory, objective events in the mind rather than as true reflections of the self [73], is a skill central to both mindfulness and cognitive-based therapies. Based on prior studies, which have demonstrated that decentering partially mediated improvements during MBCT [26,88,89], we had anticipated that MBCT-R would increase decentering. Unexpectedly, although decentering was cultivated in both active study arms, neither the MBCT-R+CHA-MW nor the iCBT+CHA-MW arms significantly increased decentering. However, given that decentering scores were significantly lower in the CHA-MW arm at baseline, interpretation of these findings may also be confounded by regression to the mean. Further research, including dismantling studies, is needed to better understand the differential impacts of CHA-MW, MBCT-R, and iCBT on decentering.

In addition to limitations already discussed, the following should be noted to contextualize our findings. First, while the intention-to-treat analysis, using MLE methods, allowed all 97 participants to be included in the final analyses, the study was underpowered to support additional subgroup analyses. For instance, the absence of detectable between-group differences may reflect limited statistical power. It remains possible that subgroups of participants derive greater benefit from higher-intensity intervention; however, a larger trial is needed to evaluate moderators. Second, the sociodemographic characteristics of our study sample, which was predominantly White, female, educated, and heterosexual, limit the generalizability of study findings to aspects of identity and experience represented by the sample. Additionally, all study interventions took place online, which required technology access and literacy, resources that are not equally distributed across the population, nor immune to self-selection biases. The study also included rigorous exclusion criteria, which may further limit the generalizability of findings. Third, the study involved multiple exploratory outcome measures, which may have felt burdensome to some participants and could have negatively affected assessment completion rates. Fourth, study measures were exclusively self-report and, as such, vulnerable to biases of mood, memory, and present perceived stress. They may also reflect common method variance, or overlapping variability that is attributable to factors related to the common assessment method used (eg, positive or negative affect and social desirability), rather than reflect true relationships among the outcomes measured. Fifth, regarding missing data, results could be biased because of the use of the missing completely at random and missing at random assumptions used by the MLE approach. Sixth, the study was not designed as a noninferiority trial and, as such, is not positioned to assert the inferiority of one intervention or another. Furthermore, while CHA-MW monitoring and response served as a low-intensity, nonspecific, control condition in this specific integrated primary care setting, the study did not involve a control in the sense of treatment as usual or a waitlist. Considering evidence supporting the spontaneous remission of depression over time [90], care should be taken in attributing observed changes to the effectiveness of the various study arms, particularly given the study’s focus on individuals with mild to moderate depression and its relatively brief, 24-month follow-up period. Seventh, while the content of the MBCT-R intervention directly aligned with that of evidence-based MBCT, its method of delivery was modestly adapted to a videoconference platform to allow for remote implementation to larger groups. Adaptation may limit the interpretability of the findings. Finally, the COVID-19 pandemic has had various temporal waves and cycles of high-intensity stress, which could have impacted engagement, intervention participation and adherence, and the overall effectiveness for some cohorts more than others.

Considering the relative resource efficiency of CHA-MW, replication of our findings with a cost-benefit analysis may reveal significant savings on a per person basis for CHA-MW over MBCT-R+CHA-MW or iCBT+CHA-MW. Future research could compare the effectiveness and resource efficiency of CHA-MW with other existing population mental health triaging systems used within community health care systems (eg, collaborative care [91,92]). MBCT or iCBT assignments were randomized in this study, yet treatment matching might have improved the impacts of the mindfulness and iCBT interventions, although studies have not always shown this to be the case [93]. Any future iterations of MBCT-R could include brief individual interviews, based upon evidence in a recent study that found that adding brief individual interviews (which often happen formally or informally during in-person mindfulness-based interventions) can have a measurable difference on adherence and completion rates in online mindfulness-based interventions [94]. Further research may be able to determine whether CHA-MW reached a ceiling that resulted in the other interventions having, on average, negligible additional benefit, and whether low completion rates during the COVID-19 pandemic and/or MBCT-R–specific adaptations (eg, larger group sizes online) contributed to current findings.

Implementing programs similar to CHA-MW, which pair online psychoeducation newsletters with a resource-efficient, high-tech, and soft-touch digital population mental health monitoring and semiautomated stratified care algorithm that optimizes patient agency and engagement with responsive, precise, and evidence-based care, could help to prevent trajectories of progressive impairment in the context of large-scale surges in the demand for individual and community mental health services during periods of high population stress.

Findings from this randomized controlled clinical trial conducted during the COVID-19 pandemic suggest that MBCT-R+CHA-MW, iCBT+CHA-MW, and CHA-MW were each effective in treating depression. However, this study failed to demonstrate significant differences in the effectiveness of MBCT-R+CHA-MW as compared with CHA-MW in treating depression. Intervention completion during the COVID-19 pandemic was low, which may have impacted the extent of the effect that these evidence-based, remotely delivered interventions could have on depression symptom severity. These findings advance current research and clinical paradigms by spotlighting the relative cost- and care-effectiveness of CHA-MW when online intervention engagement and adherence may be low, with the finding being consistent whether the online remotely delivered intervention is synchronous group or asynchronous individual. Given the lack of evidence supporting any additive benefits of MCBT or iCBT to CHA-MW, more definitive research, with a highly powered trial, is needed to examine whether CHA-MW, with its effectiveness at mental wellness monitoring and referring participants to existing telemental health options, may be sufficient for tackling rapidly increasing mental health care needs during large-scale stressor events.