CAMH PARTNERs Integrated Care Study (PARTNERs)

This research study will evaluate an integrated care model of telephone-based, computer-aided care management using a new role of Mental Health Technician and specialized software to support primary care providers in providing mental health care. The study will compare the effectiveness of this model vs. enhanced usual care in improving initiation of specific treatment by the primary care provider, reduction in severity of symptoms, and improvement in quality of life or functioning.

This is a randomized clinical trial of an intervention to improve the outcomes of depressive disorders, generalized anxiety, panic disorder, and at-risk drinking or alcohol use disorders in primary care. Two groups will be included: (1) Adult patients who self-refer or are referred to the study by their Primary Care Provider; or (2) patients or are referred by their Primary Care Provider to the tele-mental health consultation service at the Centre for Addiction and Mental Health (Toronto, Ontario). Patients who consent and meet the eligibility criteria will be randomized to either: (1) usual care provided by their Primary Care Provider enhanced by a baseline assessment and follow-up assessments 4, 8, and 12 months later; or (2) an intervention consisting of care provided by their Primary Care Provider enhanced by a baseline assessment, follow-up assessments 4, 8, and 12 months later plus psychoeducation, monitoring, and support offered by telephone by a Mental Health Technician receiving supervision from a project psychiatrist. At baseline and 4, 8, and 12 months later, symptoms, quality of life, and function of all participants will be assessed by telephone by Research Analysts (RAs) blind to the randomization. In addition, at 12 months, patient satisfaction will be assessed. Results of these assessments will be shared with the participant's Primary Care Provider, regardless of randomization. The Primary Care Provider will receive a document summarizing the results of the baseline assessment and the three follow-up assessments for the participants of both the usual care and intervention groups. For participants in the intervention group the Primary Care Provider will also receive treatment recommendations and updates as needed. In the enhanced usual care group, the Primary Care Provider will treat the participant as they deem appropriate. In the intervention group, participants will be assigned to a team constituted by a Mental Health Technician and a psychiatrist. The team will communicate evidence-based treatment recommendations to the Primary Care Provider, for example initiating the use of an antidepressant or referring the participant for therapy or counselling. The Mental Health Technician will offer participants telephone-based psychoeducation about their diagnoses made by their Primary Care Provider or other consulting physicians, their symptoms, and the treatment selected by the Primary Care Provider. The Mental Health Technician will also monitor by phone the progress of treatment and communicate updates to the Primary Care Provider with additional recommendations as required. Thus, the Mental Health Technician will contact the Primary Care Provider on an as needed basis. In addition, for participants who are doing well and for whom no new recommendations are needed, the Mental Health Technician will provide a brief "progress report" to the Primary Care Provider on a monthly basis. In addition, at the completion of the one-year period, the RA will review all the participants' charts to abstract the intervention (e.g., medications, doses, refills) and other health services (e.g., diagnostic tests, referrals, mental health evaluations) they have received during their participation in the study. The chart audit will be conducted using REDCap, a secure application supported by the CAMH Information Management Group (IMG). IMG has all the necessary physical and operational securities in place to meet security and privacy regulations for data transmission and storage using REDCap. Investigators will follow the best practices outlined by IMG to ensure that all data and information is kept private and confidential.

This group will receive Enhanced Usual Care. The subject's Primary Care Provider will receive results of baseline, four, eight, and twelve month assessments examining the subject's symptoms of depression, anxiety, and other mental health problems, use of alcohol and illicit substances, physical pain, and quality of life . If the subjects primary care provider chooses, they can use this information to construct the subject's treatment plan.

The subject's Primary Care Provider will receive results of baseline, four, eight, and twelve month assessments examining the subject's symptoms of depression, anxiety, and other mental health problems, use of alcohol and illicit substances, physical pain, and quality of life and recommendations for treatment. Over a 3-12 months period, the Intervention group will receive a brief, telephone-based intervention by a Mental Health Technician that will focus on monitoring symptoms, treatment adherence, and providing psychoeducation.

In addition to structured assessments at baseline, four, eight and twelve months (the results will be shared with their Primary Care Provider), subjects will receive psychoeducation regarding their depression, anxiety, and alcohol use disorder or at-risk drinking; symptom monitoring; and medication adherence support. Their Primary Care Provider will also be provided evidence-based recommendations regarding their treatment.

Subjects will receive structured, quantitative assessments at baseline, four, eight and twelve months. The results will be shared with their Primary Care Provider.

Severity of depressive symptoms, as measured with PHQ-9 during phone interviews at 4, 8, and 12 months

Mixed models will be used to assess changes in depressive symptoms among depressed participants (i.e., those with a PHQ-9 score of 10 or higher at baseline) adjusting for relevant covariates and clustering within subjects over time and within sites. Change in PHQ-9 scores (follow-up minus baseline scores) will be compared between treatment groups with a linear contrast model including the main effects of time, treatment group, and their interaction with the expectation to detect a group effect and a group x time interaction. Time will have 4 levels (Baseline, 4, 8 and 12 months); treatment group will have two levels (EUC and Intervention). We will test the bivariate association between baseline socio-demographics and baseline PHQ-9 scores with treatment group. Variables associated at a significance level of 0.05, will be entered in the model as covariates. Comorbidities will be also entered as control variables.

Initiation or change in antidepressant medication documented in the health record at the 4-, 8-, and 12-month follow up

Generalized Estimating Equation (GEE) will be used to compare the proportion of subjects that started the treatment for each specific comorbidity in the first 12 months after start of the study. The model will use logit link and consider sites and PCPs as clustering structures, and besides the simple test of study group effect the model will also control for relevant covariates detected in the initial descriptive group comparison (see Primary Outcome). Time will also enter the model as a categorical predictor with 4 levels. Separate models will be fitted for each condition (depression, anxiety, and alcohol use disorder) and each model will control for the existence or not of the other two conditions. The outcome variable for each analysis will be a simple binary indicator of whether the treatment started or not during the interval analyzed.

Participants will be interviewed over the phone with the Generalized Anxiety Disorder-7 (GAD-7) and the Alcohol Use 7 Day Timeline Followback at baseline, 4, 8, and 12 months. Mixed models will be used to assess changes in anxiety symptoms among anxious participants (i.e., those with a GAD-7 score of 10 or higher at baseline) or changes in the number of drinks during the past week among those with at-risk drinking (defined as drinking 10 or more drinks per week for a female, 15 or more drinks per week for a male, or having 4 or more binges during the past 3 months - with a binge defined as 4 or more drinks in a single day for a female and 5 or more drinks in a single day for a male). Mixed models adjust for all relevant covariates and clustering within subjects over time, within sites, and within Primary Care Providers. These changes will be defined and analyzed as the changes in PHQ-9 scores described above.

Time to initiation of first treatment will be tested using Cox regression, and logistic regression analysis will be used to determine whether the proportion of cases receiving treatment differs across study arms after controlling for all relevant covariates

Proportion of subjects receiving treatment for depression, anxiety, an alcohol use disorder or at-risk drinking

Generalized linear models will be used to determine whether the proportion of subjects receiving treatment differs across study arms after controlling for clustering across sites and within Primary Care Providers in addition to all relevant covariates.

Rates of depression remission, depression response, anxiety remission, anxiety response, and decrease in alcohol use to the point where it meets the guidelines for safe drinking, remaining stable, or worsening.

Cox regression (for time to remission or response) and GEE (investigating rates of remission or response for each condition as a binary outcome). The following operational definitions will be used: depression response: Patient Health Questionnaire-9 score<10; depression remission: Patient Health Questionnaire-9 score< 5; anxiety response: Generalized Anxiety Disorder-7< 10; anxiety remission: Generalized Anxiety Disorder-7< 5; at risk-drinking threshold: no more than 15 drinks per week for men with no more than 3 drinks a day most days or 10 drinks a week for women with no more than 2 drinks per day most days. A GEE model will be adjusted for each of the outcomes, using a logit link. Study site and PCP will be used to control for clustering. The models will include and explore a categorical Time effect with 4 levels (baseline, 4, 8 and 12months), study group and their interaction, as well as relevant covariates found with approaches described in the Primary and Secondary Outcomes.

Participants will be interviewed over the phone with the Veterans Rand-12 (VR-12) questionnaire at baseline, 4, 8, and 12 months. Mixed models will be used to assess changes in scores adjusting for all relevant covariates and clustering within subjects over time, within sites, and within PCP. These changes will be defined and analyzed as the changes in PHQ-9 scores described above with thee scores for the three main conditions (PHQ-9, GAD-7 and 7 Day Timeline Followback) entered as control variables.

Inclusion Criteria: Age 18 and older Receiving care from a Primary Care Provider at one of the participating health care organizations Self-referred or referred by a participating primary care provider because of suspected depression, anxiety, alcohol disorder or at risk drinking Having access to a telephone Willingness and ability to converse in English by telephone Corrected auditory acuity that enables to converse in English by telephone Willingness and ability to provide informed consent Exclusion Criteria: Not meeting all inclusion criteria Lifetime primary psychotic illness, bipolar disorder, obsessive-compulsive disorder, or post-traumatic disorder Current substance abuse or dependence (not including alcohol use disorders) Clinically significant cognitive impairment as indicated by a score of 16 or above on the Blessed Orientation Memory Concentration Test High risk for suicide as indicated by the 5-item Paykel Scale Being physically unstable as evidenced by needing to be hospitalized Being expected by one's Primary Care Provider to die during the next 6 months

Rodie DJ, Fitzgibbon K, Perivolaris A, Crawford A, Geist R, Levinson A, Mitchell B, Oslin D, Sunderji N, Mulsant BH; PARTNERs Study Group. The primary care assessment and research of a telephone intervention for neuropsychiatric conditions with education and resources study: Design, rationale, and sample of the PARTNERs randomized controlled trial. Contemp Clin Trials. 2021 Apr;103:106284. doi: 10.1016/j.cct.2021.106284. Epub 2021 Jan 19.

Zaheer S, Garofalo V, Rodie D, Perivolaris A, Chum J, Crawford A, Geist R, Levinson A, Mitchell B, Oslin D, Sunderji N, Mulsant BH; PARTNERs Study Group. Computer-Aided Telephone Support for Primary Care Patients with Common Mental Health Conditions: Randomized Controlled Trial. JMIR Ment Health. 2018 Dec 10;5(4):e10224. doi: 10.2196/10224.

The Centre for Addiction and Mental Health (CAMH) is the leading mental health and addictions research facility in Canada, and one of the largest in the world.