Skip to content

Community Partners in Care is a Research Project Funded by the National Institutes of Health

CPIC is a Community Partnered Participatory Research (CPPR) Project of Community and Academic Partners Working Together to Learn the Best Way to Reduce Depression in Our Communities.

Status
Completed
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT01699789
Acronym
CPIC
Enrollment
1246
Registered
2012-10-04
Start date
2009-01-31
Completion date
2016-05-31
Last updated
2021-06-24

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Depression, Information Dissemination, Social Determinants of Health

Keywords

Community Partnered Participatory Research, Community Engagement, Implementation, Community Based Participatory Research, Quality Improvement, Evidence Based Practice, Patient Centered Outcomes Research

Brief summary

CPIC is a community initiative and research study funded by the NIH. CPIC was developed and is being run by community and academic partners in Los Angeles underserved communities of color. CPIC compares two ways of supporting diverse health and social programs in under-resourced communities to improve their services to depressed clients. One approach is time-limited expert technical assistance coupled with culturally-competent community outreach to individual programs, on how to use quality improvement toolkits for depression that have already been proven to be effective or helpful in primary care settings, but adapted for this study for use in diverse community-based programs in underserved communities. The other approach brings different types of agencies and members in a community together in a 4 to 6-month planning process, to fit the same depression quality improvement programs to the needs and strengths of the community and to develop a network of programs serving the community to support clients with depression together. The study is designed to determine the added value of community engagement and planning over and above what might be offered through a community-oriented, disease management company. Both intervention models are based on the same quality improvement toolkits that support team leadership, care management, Cognitive Behavioral Therapy, medication management, and patient education and activation. Investigators hypothesized that the community engagement approach would increase agency and clinician participation in evidence-based trainings and improve client mental health-related quality of life. In addition, during the design phase, community participants prioritized adding as outcomes indicators of social determinants of mental health, including physical functioning, risk factors for homelessness and employment. Investigators hypothesized by activating community agencies that can address health and social services needs to engage depressed clients, these outcomes would also be improved more in the collaboration condition. Investigators also hypothesized that the collaboration approach would increase use of services.

Detailed description

Underserved communities of color in low income, largely ethnic-minority neighborhoods face an excessive burden of illness from depression due to higher prevalence of depression and lower access to quality care. Evidence-based quality improvement (QI) programs for depression in primary care settings-where many low-income and minority patients receive their only mental health care-can enhance quality of depression care and improve health outcomes. These programs are under-utilized in community-based health care settings, and have not been adapted for use across diverse agencies (social service, faith based, primary and specialty care) that could partner to support disease management for depression. Partners in Care (PIC)and WE Care are interventions designed to improve access to evidence-based depression treatments (medication management or psychotherapy) for primary care patients and, in WE Care, social service clients. PIC evaluated a services delivery intervention while WE Care was an effectiveness trial with study-provided treatments. Both studies promoted use of the same evidence-based treatments. Both PIC and WE Care programs improved use of evidence-based treatments for depression and health outcomes for African Americans and Latinos. The PIC interventions reduced health outcome disparities evident in usual care in the first follow-up year and at five-year follow-up. While these findings offer hope to underserved communities, such communities have poor resources to support implementation of these programs, and may have historical distrust in research and health care settings. There is no evidence-based approach to support agency networks in underserved communities in implementing QI programs for depression. To address this information gap, investigators created Community Partners in Care (CPIC), a group-level randomized, controlled trial, with randomization at the level of an agency site or unit. The trial is being fielded in two underserved communities, Hollywood and South Los Angeles, and conducted through a community- participatory, partnered research (CPPR) approach. The specific aims of the study are: 1. To engage two underserved communities in improving safety-net care for depression. 2. To examine the effects of a community-engagement approach to implementing evidence-based depression quality improvement toolkits (PIC/WE Care) through a community collaborative network across services sectors, compared to technical assistance to individual programs from the same services sectors coupled with culturally-competent outreach to implement the same toolkits. The outcomes are: a) client access to care, quality of care and health outcomes, with the primary outcome being mental-health related quality of life and additional outcomes reflecting social determinants of mental health of interest to the community (physical health, homelessness risk factors, employment); b) services utilization and costs; c) agency adoption of PIC/WE Care; d) and provider attitudes, knowledge and practice. 3. To describe the process of implementation of the community engagement intervention. CPIC was awarded funds from the Patient Centered Outcomes Research Institute (PCORI) in 2013 to accomplish the following 3 aims: 1. To compare the long-term (3-year) effectiveness of community engagement and planning versus agency technical assistance to implement depression QI and improve depressed clients' health status and risk for homelessness 2. To determine how depressed clients in under-resourced communities prioritize diverse health and social outcomes and identify their preferences for services to address priority outcomes 3. To identify capacities of providers to respond to depressed clients' priorities and to generate recommendations for building capacity to better address clients'priorities. We hypothesize that community engagement and planning will be more effective than technical assistance in improving 3-year outcomes and that clients will prioritize quality of life. We expect to find gaps in provider capacities to address client priorities that network strategies could address. Our primary outcome for the long-term follow-up is mental health related quality of life and secondary outcomes are use of healthcare and community services for depression and physical functioning and homeless risk factors. In 2014, CPIC was awarded funds from the National Institute on Minority Health and Health Disparities (NIMHD) to use existing quantitative CPIC data and collect new qualitative data to describe pathways to reducing disparities. The funding allows us to longitudinally track the implementation of the CEP model in a new county-wide initiative to develop community networks to promote healthy neighborhoods. The aims under this additional funding are: 1. To determine pathways to reducing mental health and social disparities by conducting community-academic partnered analyses of CPIC data by 1) examining intervention effects for disparity subgroups (African Americans, Latinos, gender groups, insurance and housing status groups); 2) identifying predictors and mediators of barriers to access/services and client outcomes; 3) analyzing intervention effects on provider workforce diversity; and 4) generating explanatory models for intervention effects and their sustainability by interviewing CPIC administrators and providers, as well as prior and current clients. 2. To explore the generalizability and replicability of the CPIC partnered model and, more broadly, to inform the process of incorporating science into policy by conducting a longitudinal case study of the CEP model implementation in a county-wide neighborhood health initiative in Los Angeles to reduce mental health and social disparities. As a result of this study, we will be able to explain how community-engaged and participatory models of intervention implementation can reduce health and social disparities and ultimately achieve public health impact. Study findings will be disseminated widely using traditional academic, community-valued, and policy-relevant dissemination channels.

Interventions

The quality improvement program is an evidence-based toolkit from prior studies (see Names above) that supported team leadership, case and care management, medication management, and Cognitive Behavioral Therapy for Depression. The Case management manual supported depression screening and monitoring/tracking of outcomes; patient education and activation, care coordination, and behavioral activation and problem solving. The toolkit includes education on depression and a community health worker manual.

BEHAVIORALResources for Services Expert Team

The expert team consisted for RS consisted of 3 psychiatrists, a psychologist expert in Cognitive Behavioral Therapy, a nurse care manager, a community engagement specialist, a quality improvement expert, and staff support. They team offered 12 web-based seminars to each community on components of collaborative care as well as site visits to primary care clinics on clinical assessment and medication management.

BEHAVIORALCommunity Engagement and Planning Council

The CEP Council was supported by a workbook developed by the overall CPIC Council that provided principles, approach, agendas, and resources for the multi-sector planning meetings. The CEP Councils met twice a month for 4-6 months to develop their plan and met monthly during implementation of trainings. The study Council supported CEP meetings. Community leaders co-led trainings with study experts to help assure sustainability. Each CEP council had $15K to defray costs of venues, materials, and consultations, while the study provided that for RS.

Sponsors

National Institute of Mental Health (NIMH)
CollaboratorNIH
Robert Wood Johnson Foundation
CollaboratorOTHER
National Library of Medicine (NLM)
CollaboratorNIH
Patient-Centered Outcomes Research Institute
CollaboratorOTHER
National Institute on Minority Health and Health Disparities (NIMHD)
CollaboratorNIH
RAND
Lead SponsorOTHER

Study design

Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Outcomes Assessor)

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No

Inclusion criteria

Administrators * Age 18 and above * Work or volunteer for an enrolled program in the study and be designated as a liaison by the program Providers * Age 18 and above * Have direct contact with patients/clients Clients * Age 18 and above * Score 10 or greater on modified Patient Health Questionnaire (PHQ-8)

Exclusion criteria

grossly disorganized by screener staff assessment Not providing personal contact information Administrators - Under age 18 Providers \- Under age 18 Clients * Under age 18 * Gross cognitive disorganization by screener staff assessment * Providing no contact information

Design outcomes

Primary

MeasureTime frameDescription
Percent of Participants With Poor Mental Health Quality of Life, MCS12≤ 406 months follow-upFrom the Short Form, 12-item quality of life measure, mental health-related quality of life is the primary client outcome. Poor mental health related quality of life is defined as MCS12≤ 40 (one standard deviation below population mean).
Percent of Participants With PHQ-9 Score ≥ 106 months follow-upPatient Health Questionnaire 9-item version (PHQ-9) at least mild depression (score ≥ 10)
Percent of Participants With PHQ-8 Score ≥ 1036 months follow-upPatient Health Questionnaire 8-item version (PHQ-8) at least mild depression (score ≥ 10)

Secondary

MeasureTime frameDescription
Percent of Participants With Homeless or ≥ 2 Risk Factors for Homelessness6 months follow-upDefined as current homelessness or living in a shelter or having at least 2 risk factors (e.g., no place to stay for at least 2 nights or eviction from a primary residence, financial crisis, or food insecurity in the past 6 months)
Percent of Participants With Working for Pay6 months follow-up
Percent of Participants With Any Missed Work Day in Last 30 Days, if Working6 months follow-up
Percent of Participants With Hospitalization for Behavioral Health in the Past 6 Months6 months follow-upself-reported services use in the past 6 months for overnight hospital stays for mental health or substance abuse
Percent of Participants With >=4 Hospital Nights for Behavioral Health in the Past 6 Months6 months follow-upself-reported services use in the past 6 months with \>=4 overnight hospital stays for any emotional, mental, alcohol, or drug problem, median cut point for baseline variable
Percent of Participants With >=2 Emergency Room Visits in the Past 6 Months6 months follow-upself-reported services use in the past 6 months with \>=2 emergency room visits in past 6 months, median cut point for baseline variable
Percent of Participants With Any MHS Outpatient Visit in the Past 6 Months6 months follow-upself-reported mental health outpatient visit from mental health provider, including psychiatrists, psychologists, social workers, psychiatric nurses, or counselors in the past 6 months
Percent of Participants With Any PCP Visit With Depression Service in the Past 6 Months6 months follow-upself-reported services use in the past 6 months with any primary care visit for depression
Percent of Participants With >= 2 PCP Visits With Depression Services, if Any6 months follow-up
Percent of Participants With Faith-based Program Participation in the Past 6 Months6 months follow-upWent to any religious or spiritual places such as a church, mosque, temple, or synagogue in the past 6 months
Percent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months6 months follow-up
Percent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months6 months follow-up
Medication Visits Among MHS Users in the Past 6 Months6 months follow-up
Faith-based Visits With Depression Service if Faith Participation in the Past 6 Months6 months follow-upFor this sector, depression/mental health service is defined by client report of having assessment, counseling, education, medication discussion or referral for depression or emotional or mental health problems.
Park or Community Center Visits With Depression Service if Went to Park or Community Center in Past 6 Months6 months follow-upFor this sector, depression/mental health service is defined by client report of having assessment, counseling, education, medication discussion or referral for depression or emotional or mental health problems.
Total Mental Health Related Outpatient Visits in the Past 6 Months6 months follow-upTotal outpatient visits for depression, mental health or substance abuse from emergency rooms, primary care or public health, mental health, substance abuse, or social-community services sectors in the past 6 months
PCS-12 Scores on 12-Item Physical Health Summary Measure, Comparison Between CEP and RS Groups36 months follow-up12-item physical composite score (PCS-12). Possible scores on range from 0 to 100, with higher scores indicating better physical health
Nights Hospitalized for Behavioral Health Reason in the Past 6 Months36 months follow-upself-reported number of overnight hospital stays for any emotional, mental, alcohol, or drug problem in past 6 months
N of Emergency Room or Urgent Care Visits in the Past 6 Months36 months follow-up
N of Visits to Primary Care in Past 6 Months36 months follow-up
N of Outpatient Visits to Primary Care for Depression Services in the Past 6 Months36 months follow-up
N of Outpatient Mental Health Visits in Past 6 Months36 months follow-up
N of Outpatient Visits to a Substance Abuse Treatment Agency or Self Help Group in the Past 6 Months36 months follow-up
N of Social Services for Depression Visits in the Past 6 Months36 months follow-up
Number of Calls to Hotline for Substance Use or Mental Health Problem in the Past 6 Months36 months follow-up
N of Days on Which a Self-help Visit for Mental Health Was Made in the Past 6 Months36 months follow-up
Percent of Participants With Any Faith-based Services for Depression in the Past 6 Months36 months follow-up
Percent of Participants With Use of Any Antidepressant in the Past 6 Months36 months follow-up
Percent of Participants With Use of Any Mood Stabilizer in the Past 6 Months36 months follow-up
Percent of Participants With Use of Any Antipsychotic in the Past 6 Months36 months follow-up
Percent of Participants With Any Visit in Health Care Sector in the Past 6 Months36 months follow-up
Percent of Participants With Any Community-sector Visit for Depression in the Past 6 Months36 months follow-up
Percent of Participants With Any Depression Treatment in the Past 6 Months36 months follow-upAntidepressant use for at least two months or at least four outpatient visits to mental health or primary care setting for depression services
Survival Analysis for Time to the First Clinical Remissionfrom baseline to 3 yearsclinical remission: Patient Health Questionnaire, PHQ-8 score \<10. Cox Proportional Hazard model was used to examine the impact of the intervention on speed of clinical remission over the 3 years follow-up period, defined as the first assessment with clinical remission (PHQ-8\<10).
Percent of Participants With Mental Wellness6 months follow-upMental wellness is defined as at least a good bit of time in the prior 4 weeks on any of three items: feeling peaceful or calm, being a happy person, having energy
Percent of Participants With Clinical Remission4 years follow-upClinical remission defined as Patient Health Questionnaire-2 (PHQ-2) score \< 3.
Percent of Participants With Community-Defined Remission4 years follow-upCommunity-Defined Remission defined as PHQ-2\<3, MCS-12\>40, or mental wellness
Survival Analysis for Time to the First Community-Defined Remissionfrom baseline to 3 yearsCommunity-Defined Remission: PHQ-8\<10 or MCS-12\>40 or any mental wellness. Cox Proportional Hazard model was used to examine the impact of the intervention on speed of community-defined remission over the 3 years follow-up period, defined as the first assessment with community-defined (PHQ-8\<10 or MCS-12\>40 or any mental wellness)
Percent of Participants Reported Organized Life6 months follow-upA response of somewhat or definitely true to my life is organized versus unsure or somewhat false or definitely false
Percent of Participants With Physically Active6 months follow-upPhysically Active is defined as at least active to How physically active you are?

Countries

United States

Participant flow

Recruitment details

From March 2010 to November 2010, the study screened 4,440 clients from 93 programs in 50 agencies. The ninety-three programs, included 17 primary care/public health, 18 mental health, 20 substance abuse, ten homeless services, and 28 social/other community services.

Participants by arm

ArmCount
Resources for Services RS
RS offers time-limited technical assistance to individual agencies and outreach from a community engagement specialty, to review components of the QI Program Intervention as implemented by the RS Expert Team. QI Program: The QI program is an evidence-based toolkit that supported team leadership, case and care management, medication management, and CBT for depression. The Case management manual supported depression screening and tracking of outcomes; patient education and activation, care coordination, and behavioral activation and problem solving. The toolkit includes education on depression and a community health worker manual. The expert team for RS consisted of 3 psychiatrists, a psychologist expert in CBT, a nurse care manager, a community engagement specialist, a QI expert, and staff support. The team offered 12 webinars to each community on components of collaborative care as well as site visits to primary care clinics on clinical assessment and medication management.
504
Community Engagement and Planning CEP
CEP supports 4 months of planning for the CEP Council of representatives from assigned programs in biweekly 2-hr meetings to fit trainings in the QI Program to the community and develop strategies across programs to collaborate as a network. The CEP Council developed a written plan for training and monitoring and supported implementation of the training plan. CEP sites received enrolled client lists. The toolkit is the same as RS. The CEP Council was supported by a workbook developed by the overall CPIC Council that provided principles, approach, agendas, and resources for the multi-sector planning meetings. The CEP Councils met twice a month for 4-6 months to develop their plan and met monthly during implementation of trainings. The study Council supported CEP meetings. Community leaders co-led trainings with study experts to help assure sustainability. Each CEP council had $15K to defray costs of venues, materials, and consultations, while the study provided that for RS.
514
Total1,018

Withdrawals & dropouts

PeriodReasonFG000FG001
Overall StudyDeath12
Overall StudyLost to Follow-up83101
Overall StudyWithdrawal by Subject1823

Baseline characteristics

CharacteristicCommunity Engagement and Planning CEPTotalResources for Services RS
Age, Continuous46.6 years
STANDARD_DEVIATION 13.2
45.8 years
STANDARD_DEVIATION 12.9
44.9 years
STANDARD_DEVIATION 12.4
Education
High school or above
290 participants573 participants283 participants
Education
Less than high school education
224 participants445 participants221 participants
Health Insurance Status
Had health insurance
255 participants473 participants218 participants
Health Insurance Status
No health insurance
259 participants545 participants286 participants
Race/Ethnicity, Customized
African American
249 participants488 participants239 participants
Race/Ethnicity, Customized
Latino
215 participants409 participants194 participants
Race/Ethnicity, Customized
Non-Hispanic white
41 participants86 participants45 participants
Race/Ethnicity, Customized
Other (Asian, Native American etc)
9 participants35 participants26 participants
Sex: Female, Male
Female
309 Participants595 Participants286 Participants
Sex: Female, Male
Male
205 Participants423 Participants218 Participants

Adverse events

Event typeEG000
affected / at risk
EG001
affected / at risk
deaths
Total, all-cause mortality
4 / 6064 / 640
other
Total, other adverse events
126 / 606126 / 640
serious
Total, serious adverse events
0 / 6060 / 640

Outcome results

Primary

Percent of Participants With PHQ-8 Score ≥ 10

Patient Health Questionnaire 8-item version (PHQ-8) at least mild depression (score ≥ 10)

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With PHQ-8 Score ≥ 1065.8 percentage of participants
Community Engagement and Planning CEPPercent of Participants With PHQ-8 Score ≥ 1066.0 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients at 3 years to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.6, 1.7]
Primary

Percent of Participants With PHQ-9 Score ≥ 10

Patient Health Questionnaire 9-item version (PHQ-9) at least mild depression (score ≥ 10)

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With PHQ-9 Score ≥ 1067.0 percentage of participants
Community Engagement and Planning CEPPercent of Participants With PHQ-9 Score ≥ 1061.7 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.48, 1.26]
Primary

Percent of Participants With Poor Mental Health Quality of Life, MCS12≤ 40

From the Short Form, 12-item quality of life measure, mental health-related quality of life is the primary client outcome. Poor mental health related quality of life is defined as MCS12≤ 40 (one standard deviation below population mean).

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4050.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4044.8 percentage of participants
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a logit link function was used with adjustment for covariates.95% CI: [0.61, 0.97]
Primary

Percent of Participants With Poor Mental Health Quality of Life, MCS12≤ 40

From the Short Form, 12-item quality of life measure, mental health-related quality of life is the primary client outcome. Poor mental health related quality of life is defined as MCS12≤ 40 (one standard deviation below population mean).

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4039.4 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4045.0 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients at 3 years to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.7, 2.3]
Primary

Percent of Participants With Poor Mental Health Quality of Life, MCS12≤ 40

From the Short Form, 12-item quality of life measure, mental health-related quality of life is the primary client outcome. Poor mental health related quality of life is defined as MCS12≤ 40 (one standard deviation below population mean).

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4051.4 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Poor Mental Health Quality of Life, MCS12≤ 4044.1 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.57, 0.95]
Secondary

Faith-based Visits With Depression Service if Faith Participation in the Past 6 Months

For this sector, depression/mental health service is defined by client report of having assessment, counseling, education, medication discussion or referral for depression or emotional or mental health problems.

Time frame: 6 months follow-up

Population: Individuals who reported any faith based participation in past 6 months

ArmMeasureValue (MEAN)
Resources for Services RSFaith-based Visits With Depression Service if Faith Participation in the Past 6 Months0.7 visits
Community Engagement and Planning CEPFaith-based Visits With Depression Service if Faith Participation in the Past 6 Months1.9 visits
Comparison: Adjusted analyses used multiply imputed data (N=588), weighted for eligible sample for enrollment; Poisson regression model adjusted for baseline status of the dependent variable and covariates and accounted for the design effect of the cluster randomization.95% CI: [1.39, 5.8]
Secondary

Medication Visits Among MHS Users in the Past 6 Months

Time frame: 6 months follow-up

Population: Individuals who reported any mental health specialty outpatient visit in past 6 months

ArmMeasureValue (MEAN)
Resources for Services RSMedication Visits Among MHS Users in the Past 6 Months10.9 visits
Community Engagement and Planning CEPMedication Visits Among MHS Users in the Past 6 Months5.3 visits
Comparison: Adjusted analyses used multiply imputed data (N=553), weighted for eligible sample for enrollment; Poisson regression model adjusted for baseline and covariates and accounted for the design effect of the cluster randomization.95% CI: [0.3, 0.82]
Secondary

Nights Hospitalized for Behavioral Health Reason in the Past 6 Months

self-reported number of overnight hospital stays for any emotional, mental, alcohol, or drug problem in past 6 months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSNights Hospitalized for Behavioral Health Reason in the Past 6 Months1.2 nights
Community Engagement and Planning CEPNights Hospitalized for Behavioral Health Reason in the Past 6 Months0.2 nights
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.1, 0.8]
Secondary

N of Days on Which a Self-help Visit for Mental Health Was Made in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Days on Which a Self-help Visit for Mental Health Was Made in the Past 6 Months6.3 days
Community Engagement and Planning CEPN of Days on Which a Self-help Visit for Mental Health Was Made in the Past 6 Months5.6 days
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.4, 1.8]
Secondary

N of Emergency Room or Urgent Care Visits in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Emergency Room or Urgent Care Visits in the Past 6 Months1.5 visits
Community Engagement and Planning CEPN of Emergency Room or Urgent Care Visits in the Past 6 Months1.9 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.4, 3.7]
Secondary

N of Outpatient Mental Health Visits in Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Outpatient Mental Health Visits in Past 6 Months5.5 visits
Community Engagement and Planning CEPN of Outpatient Mental Health Visits in Past 6 Months5.6 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.7, 1.6]
Secondary

N of Outpatient Visits to a Substance Abuse Treatment Agency or Self Help Group in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Outpatient Visits to a Substance Abuse Treatment Agency or Self Help Group in the Past 6 Months11.1 visits
Community Engagement and Planning CEPN of Outpatient Visits to a Substance Abuse Treatment Agency or Self Help Group in the Past 6 Months12.3 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.3, 4]
Secondary

N of Outpatient Visits to Primary Care for Depression Services in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Outpatient Visits to Primary Care for Depression Services in the Past 6 Months1.1 visits
Community Engagement and Planning CEPN of Outpatient Visits to Primary Care for Depression Services in the Past 6 Months1.1 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.5, 2.1]
Secondary

N of Social Services for Depression Visits in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Social Services for Depression Visits in the Past 6 Months0.6 visits
Community Engagement and Planning CEPN of Social Services for Depression Visits in the Past 6 Months0.6 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.4, 2.7]
Secondary

N of Visits to Primary Care in Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSN of Visits to Primary Care in Past 6 Months3.9 visits
Community Engagement and Planning CEPN of Visits to Primary Care in Past 6 Months4.1 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.8, 1.5]
Secondary

Number of Calls to Hotline for Substance Use or Mental Health Problem in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSNumber of Calls to Hotline for Substance Use or Mental Health Problem in the Past 6 Months0.2 calls
Community Engagement and Planning CEPNumber of Calls to Hotline for Substance Use or Mental Health Problem in the Past 6 Months0.3 calls
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.2, 8.6]
Secondary

Park or Community Center Visits With Depression Service if Went to Park or Community Center in Past 6 Months

For this sector, depression/mental health service is defined by client report of having assessment, counseling, education, medication discussion or referral for depression or emotional or mental health problems.

Time frame: 6 months follow-up

Population: Individuals who reported any park or community center visit in past 6 months

ArmMeasureValue (MEAN)
Resources for Services RSPark or Community Center Visits With Depression Service if Went to Park or Community Center in Past 6 Months0.3 visits
Community Engagement and Planning CEPPark or Community Center Visits With Depression Service if Went to Park or Community Center in Past 6 Months1.6 visits
Comparison: Adjusted analyses used multiply imputed data (N=410), weighted for eligible sample for enrollment; Poisson regression model adjusted for baseline and covariates and accounted for the design effect of the cluster randomization.95% CI: [1.5, 24.9]
Secondary

PCS-12 Scores on 12-Item Physical Health Summary Measure, Comparison Between CEP and RS Groups

12-item physical composite score (PCS-12). Possible scores on range from 0 to 100, with higher scores indicating better physical health

Time frame: 36 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSPCS-12 Scores on 12-Item Physical Health Summary Measure, Comparison Between CEP and RS Groups38.7 units on a scale
Community Engagement and Planning CEPPCS-12 Scores on 12-Item Physical Health Summary Measure, Comparison Between CEP and RS Groups39.9 units on a scale
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a linear regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.2, 2]
Secondary

Percent of Participants Reported Organized Life

A response of somewhat or definitely true to my life is organized versus unsure or somewhat false or definitely false

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants Reported Organized Life42.7 percentage of participants
Community Engagement and Planning CEPPercent of Participants Reported Organized Life51.7 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1.03, 2.04]
Secondary

Percent of Participants With >=2 Emergency Room Visits in the Past 6 Months

self-reported services use in the past 6 months with \>=2 emergency room visits in past 6 months, median cut point for baseline variable

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With >=2 Emergency Room Visits in the Past 6 Months28.3 percentage of participants
Community Engagement and Planning CEPPercent of Participants With >=2 Emergency Room Visits in the Past 6 Months24.5 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.52, 1.25]
Secondary

Percent of Participants With >= 2 PCP Visits With Depression Services, if Any

Time frame: 6 months follow-up

Population: Individuals who reported any PCP visit in past 6 months

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With >= 2 PCP Visits With Depression Services, if Any61.9 percentage of participants
Community Engagement and Planning CEPPercent of Participants With >= 2 PCP Visits With Depression Services, if Any79.8 percentage of participants
Comparison: Adjusted analyses used multiply imputed data (N=298), weighted for eligible sample for enrollment; logistic regression model adjusted for baseline and covariates and accounted for the design effect of the cluster randomization.p-value: <0.0195% CI: [1.4, 4.94]Regression, Logistic
Secondary

Percent of Participants With >=4 Hospital Nights for Behavioral Health in the Past 6 Months

self-reported services use in the past 6 months with \>=4 overnight hospital stays for any emotional, mental, alcohol, or drug problem, median cut point for baseline variable

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With >=4 Hospital Nights for Behavioral Health in the Past 6 Months5.8 percentage of participants
Community Engagement and Planning CEPPercent of Participants With >=4 Hospital Nights for Behavioral Health in the Past 6 Months2.1 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.14, 0.88]
Secondary

Percent of Participants With Any Community-sector Visit for Depression in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Community-sector Visit for Depression in the Past 6 Months28.3 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Community-sector Visit for Depression in the Past 6 Months35.6 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1, 2]
Secondary

Percent of Participants With Any Depression Treatment in the Past 6 Months

Antidepressant use for at least two months or at least four outpatient visits to mental health or primary care setting for depression services

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Depression Treatment in the Past 6 Months43.2 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Depression Treatment in the Past 6 Months43.5 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.6, 1.7]
Secondary

Percent of Participants With Any Faith-based Services for Depression in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Faith-based Services for Depression in the Past 6 Months9.4 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Faith-based Services for Depression in the Past 6 Months15.2 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1.2, 2.6]
Secondary

Percent of Participants With Any MHS Outpatient Visit in the Past 6 Months

self-reported mental health outpatient visit from mental health provider, including psychiatrists, psychologists, social workers, psychiatric nurses, or counselors in the past 6 months

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any MHS Outpatient Visit in the Past 6 Months53.9 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any MHS Outpatient Visit in the Past 6 Months53.6 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition95% CI: [0.69, 1.41]
Secondary

Percent of Participants With Any MHS Outpatient Visit in the Past 6 Months

self-reported mental health outpatient visit from mental health provider, including psychiatrists, psychologists, social workers, psychiatric nurses, or counselors in the past 6 months

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any MHS Outpatient Visit in the Past 6 Months44.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any MHS Outpatient Visit in the Past 6 Months42.6 percentage of participants
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a logit link function was used with adjustment for covariates.95% CI: [0.66, 1.66]
Secondary

Percent of Participants With Any Missed Work Day in Last 30 Days, if Working

Time frame: 6 months follow-up

Population: Population of individuals who are working

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Missed Work Day in Last 30 Days, if Working63.1 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Missed Work Day in Last 30 Days, if Working51.5 percentage of participants
Comparison: Adjusted analyses used multiply imputed data (N= 249), weighted for eligible sample for enrollment; logistic regression model adjusted for baseline and covariates and accounted for the design effect of the cluster randomization.95% CI: [0.32, 1.09]
Secondary

Percent of Participants With Any PCP Visit With Depression Service in the Past 6 Months

self-reported services use in the past 6 months with any primary care visit for depression

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any PCP Visit With Depression Service in the Past 6 Months25.1 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any PCP Visit With Depression Service in the Past 6 Months28.4 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a generalized estimating equation logistic regression model adjusted for covariates, accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.74, 1.42]
Secondary

Percent of Participants With Any PCP Visit With Depression Service in the Past 6 Months

self-reported services use in the past 6 months with any primary care visit for depression

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any PCP Visit With Depression Service in the Past 6 Months29.2 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any PCP Visit With Depression Service in the Past 6 Months29.4 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.7, 1.46]
Secondary

Percent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months41.1 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months39.4 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.61, 1.4]
Secondary

Percent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months34.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Use of Park and Recreation or Community Centers in the Past 6 Months36.6 percentage of participants
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a logit link function was used with adjustment for covariates.95% CI: [0.72, 1.32]
Secondary

Percent of Participants With Any Visit in Health Care Sector in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Any Visit in Health Care Sector in the Past 6 Months84.2 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Any Visit in Health Care Sector in the Past 6 Months84.3 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.5, 2]
Secondary

Percent of Participants With Clinical Remission

Clinical remission defined as Patient Health Questionnaire-2 (PHQ-2) score \< 3.

Time frame: 4 years follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Clinical Remission39.7 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Clinical Remission51.7 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline covariates. Data were multiply imputed and weighted for eligible sample for enrollment, accounted for the design effect of the cluster randomization.95% CI: [1, 2.99]
Secondary

Percent of Participants With Community-Defined Remission

Community-Defined Remission defined as PHQ-2\<3, MCS-12\>40, or mental wellness

Time frame: 4 years follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Community-Defined Remission69.0 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Community-Defined Remission84.2 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline covariates. Data were multiply imputed and weighted for eligible sample for enrollment, accounted for the design effect of the cluster randomization.95% CI: [1.24, 5.54]
Secondary

Percent of Participants With Faith-based Program Participation in the Past 6 Months

Went to any religious or spiritual places such as a church, mosque, temple, or synagogue in the past 6 months

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Faith-based Program Participation in the Past 6 Months59.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Faith-based Program Participation in the Past 6 Months57.1 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a generalized estimating equation logistic regression model adjusted for covariates, accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.66, 1.21]
Secondary

Percent of Participants With Faith-based Program Participation in the Past 6 Months

Went to any religious or spiritual places such as a church, mosque, temple, or synagogue in the past 6 months

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Faith-based Program Participation in the Past 6 Months57.0 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Faith-based Program Participation in the Past 6 Months53.9 percentage of participants
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a logit link function was used with adjustment for covariates.95% CI: [0.6, 1.05]
Secondary

Percent of Participants With Homeless or ≥ 2 Risk Factors for Homelessness

Defined as current homelessness or living in a shelter or having at least 2 risk factors (e.g., no place to stay for at least 2 nights or eviction from a primary residence, financial crisis, or food insecurity in the past 6 months)

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Homeless or ≥ 2 Risk Factors for Homelessness39.8 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Homeless or ≥ 2 Risk Factors for Homelessness29.7 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.38, 0.96]
Secondary

Percent of Participants With Hospitalization for Behavioral Health in the Past 6 Months

self-reported services use in the past 6 months for overnight hospital stays for mental health or substance abuse

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Hospitalization for Behavioral Health in the Past 6 Months5.0 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Hospitalization for Behavioral Health in the Past 6 Months4.3 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a generalized estimating equation logistic regression model adjusted for covariates, accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.4, 1.22]
Secondary

Percent of Participants With Hospitalization for Behavioral Health in the Past 6 Months

self-reported services use in the past 6 months for overnight hospital stays for mental health or substance abuse

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Hospitalization for Behavioral Health in the Past 6 Months10.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Hospitalization for Behavioral Health in the Past 6 Months5.8 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.28, 0.95]
Secondary

Percent of Participants With Mental Wellness

Mental wellness is defined as at least a good bit of time in the prior 4 weeks on any of three items: feeling peaceful or calm, being a happy person, having energy

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Mental Wellness33.6 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Mental Wellness45.9 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1.19, 2.59]
Secondary

Percent of Participants With Physically Active

Physically Active is defined as at least active to How physically active you are?

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Physically Active40.3 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Physically Active49.6 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1.14, 1.98]
Secondary

Percent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months39.2 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months31.5 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.34, 1.25]
Secondary

Percent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months

Time frame: 12 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months34.0 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Use of an Antidepressant Medication for 2 Months or More in the Past 6 Months28.7 percentage of participants
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a logit link function was used with adjustment for covariates.95% CI: [0.55, 1.39]
Secondary

Percent of Participants With Use of Any Antidepressant in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Use of Any Antidepressant in the Past 6 Months28.7 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Use of Any Antidepressant in the Past 6 Months26.9 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.5, 1.5]
Secondary

Percent of Participants With Use of Any Antipsychotic in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Use of Any Antipsychotic in the Past 6 Months21.7 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Use of Any Antipsychotic in the Past 6 Months23.4 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.7, 1.7]
Secondary

Percent of Participants With Use of Any Mood Stabilizer in the Past 6 Months

Time frame: 36 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Use of Any Mood Stabilizer in the Past 6 Months2.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Use of Any Mood Stabilizer in the Past 6 Months6.4 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 980 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [1, 8.3]
Secondary

Percent of Participants With Working for Pay

Time frame: 6 months follow-up

ArmMeasureValue (NUMBER)
Resources for Services RSPercent of Participants With Working for Pay23.5 percentage of participants
Community Engagement and Planning CEPPercent of Participants With Working for Pay24.7 percentage of participants
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a logistic regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.69, 1.7]
Secondary

Survival Analysis for Time to the First Clinical Remission

clinical remission: Patient Health Questionnaire, PHQ-8 score \<10. Cox Proportional Hazard model was used to examine the impact of the intervention on speed of clinical remission over the 3 years follow-up period, defined as the first assessment with clinical remission (PHQ-8\<10).

Time frame: from baseline to 3 years

Population: Sample does not include persons in clinical remission at baseline

ArmMeasureValue (MEAN)Dispersion
Resources for Services RSSurvival Analysis for Time to the First Clinical Remission21.14 months to remissionStandard Deviation 13.98
Community Engagement and Planning CEPSurvival Analysis for Time to the First Clinical Remission20.05 months to remissionStandard Deviation 14.14
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Cox proportional-hazards model adjusted for baseline covariates. Data were multiply imputed and weighted for eligible sample for enrollment, accounted for the design effect of the cluster randomization.p-value: >0.0595% CI: [0.83, 1.5]Regression, Cox
Secondary

Survival Analysis for Time to the First Community-Defined Remission

Community-Defined Remission: PHQ-8\<10 or MCS-12\>40 or any mental wellness. Cox Proportional Hazard model was used to examine the impact of the intervention on speed of community-defined remission over the 3 years follow-up period, defined as the first assessment with community-defined (PHQ-8\<10 or MCS-12\>40 or any mental wellness)

Time frame: from baseline to 3 years

Population: Sample does not include persons in community-defined remission at baseline

ArmMeasureValue (MEAN)Dispersion
Resources for Services RSSurvival Analysis for Time to the First Community-Defined Remission14.05 months to remissionStandard Deviation 11.97
Community Engagement and Planning CEPSurvival Analysis for Time to the First Community-Defined Remission12.14 months to remissionStandard Deviation 10.93
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Cox proportional-hazards model adjusted for baseline covariates. Data were multiply imputed and weighted for eligible sample for enrollment, accounted for the design effect of the cluster randomization.p-value: >0.0595% CI: [0.99, 1.52]Regression, Cox
Secondary

Total Mental Health Related Outpatient Visits in the Past 6 Months

Total outpatient visits for depression, mental health or substance abuse from emergency rooms, primary care or public health, mental health, substance abuse, or social-community services sectors in the past 6 months

Time frame: 12 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSTotal Mental Health Related Outpatient Visits in the Past 6 Months18.7 visits
Community Engagement and Planning CEPTotal Mental Health Related Outpatient Visits in the Past 6 Months17.0 visits
Comparison: Intent-to-treat analyses of repeated measures were developed including all participants with data at baseline, 6-month, or 12-month. Missing data were imputed. A generalized estimating equation (GEE) with a log link function was used with adjustment for covariates.95% CI: [0.65, 1.29]
Secondary

Total Mental Health Related Outpatient Visits in the Past 6 Months

Total outpatient visits for depression, mental health or substance abuse from emergency rooms, primary care or public health, mental health, substance abuse, or social-community services sectors in the past 6 months

Time frame: 6 months follow-up

ArmMeasureValue (MEAN)
Resources for Services RSTotal Mental Health Related Outpatient Visits in the Past 6 Months22.9 visits
Community Engagement and Planning CEPTotal Mental Health Related Outpatient Visits in the Past 6 Months21.9 visits
Comparison: Intent-to-treat, comparative-effectiveness analyses with intervention status as the independent variable, using a Poisson regression model adjusted for baseline and covariates, and accounted for the design effect of the cluster randomization. We weighted data for 1,018 clients to characteristics of the eligible sample, with item-level imputation for missing data and wave-level imputation for missing surveys. Weights account for non-enrollment among eligible clients and attrition.95% CI: [0.59, 1.57]

Source: ClinicalTrials.gov · Data processed: Mar 19, 2026