Background: Community Health Centers care for over 20 million rural, low income and minority Americans every year. Patients often have complex mental health problems such as Posttraumatic Stress Disorder (PTSD) and Bipolar Disorder. However, Community Health Centers located in rural areas face substantial challenges to managing these patients due to lack of onsite mental health specialists, stigma and poor geographic access to specialty mental health services in the community. As a consequence, many rural primary care providers feel obligated, yet unprepared, to manage these disorders, and many patients receive inadequate treatment and continue to struggle with their symptoms. While integrated care models and telepsychiatry referral models are both promising approaches to managing patients with complex mental health problems in rural primary care settings, there have been no studies comparing which approach is more effective for which types of patients. Objectives: The central question examined by this study is whether it is better for offsite mental health specialists to support primary care providers' treatment of patients with PTSD and Bipolar Disorder through an integrated care model or to use telemedicine technology to facilitate referrals to offsite mental health specialists. We hypothesize that patients randomized to integrated care will have better outcomes than patients randomized to referral care. Methods: 1,000 primary care patients screening positive for PTSD or Bipolar Disorder will be recruited from Community Health Centers in three states (Arkansas, Michigan and Washington) and randomized to the integrated care model or the referral model. Patient Outcomes: Telephone surveys will be administered to patients at enrollment and at 6 and 12 month follow-ups. Telephone surveys will measure access to care, therapeutic alliance with providers, patient-centeredness, patient activation, satisfaction with care, appointment attendance, medication adherence, self-reported clinical symptoms, medication side-effects, health related quality of life, and progress towards life goals. A sub-sample of patients will be invited to participate in qualitative interviews to describe their treatment experience using their own words. Likewise, primary care providers will be invited to participate in qualitative interviews to voice their perspective.

Background and Significance: Community Health Centers (CHCs) are the nation's largest and fastest growing network of primary care (PC) clinics. There are 1,200 CHCs that provide clinical services to 21 million Americans. Almost half (49%) of CHC patients live in rural areas, 72% live at or below the Federal Poverty Level (100%), 67% are racial/ethnic minorities, and 36% are uninsured. Nationally, over one million CHC patients are diagnosed with a psychiatric disorder and the need for mental health (MH) services is increasing exponentially, with a 547% increase in CHC patients with a psychiatric diagnosis between 2001 and 2012. CHCs located in rural areas face the greatest challenges to managing psychiatric disorders due to the lack of MH specialists on staff and weak linkages between CHCs and MH specialists in the community. Because rural, minority, low income CHC patients face insurmountable geographical, cultural and financial barriers to specialty MH care, many of their PC providers feel obligated, yet unprepared, to manage complex psychiatric disorders like posttraumatic stress disorder (PTSD) and Bipolar Disorder (BD). PTSD and BD are devastating psychiatric disorders that often go undetected and untreated in PC. Most patients do not receive effective specialty MH care for these problems and the care provided in PC settings is often poor and ineffective. Patients with PTSD and BD have significantly worse educational attainment, lower family, social, and occupational functioning, and significantly lower quality of life. Comparative effectiveness research is needed to guide policy makers about how to best manage the growing demand for MH services in CHCs.

Study Aims: The central question addressed by this mixed-methods pragmatic comparative effectiveness trial is whether it is better to expand the scope of collaborative care programs to treat patients with more complex psychiatric disorders or to facilitate successful referrals to specialty mental health care. The primary objective of this trial is to compare Telepsychiatry Collaborative Care (TCC) and Telepsychiatry Enhanced Referral (TER) from the patient and provider perspective. The secondary objective is to determine whether patients not engaging to TER, improve with Phone-Psychiatry Enhanced Referral (PER). There are four specific aims. Specific Aim #1: To quantitatively compare the treatment experience, engagement, self-reported clinical outcomes, and recovery-oriented outcomes of patients initially randomized to TCC and TER. Specific Aim #2: For the subset of patients randomized to TER who do not engage in treatment and are still symptomatic at 6 months, quantitatively compare treatment experience, treatment engagement, self-reported clinical outcomes and recovery-oriented outcomes of patients randomized to continued-TER or PER. Specific Aim #3: To gain an in-depth understanding of patients' and providers' treatment experience, qualitatively compare those randomized to TCC, TER and PER. Specific Aim #4: To examine treatment heterogeneity among subgroups of patients randomized to TCC and TER based on race/ethnicity, age and clinical severity.

Study Description: The study will be conducted in 15 CHC systems located in the states of Arkansas, Michigan and Washington. These 15 CHC treat 294,645 adult patients living in rural areas; 96.1% live in poverty and 53% are racial/ethnic minorities. Participating clinics will screen patients for PTSD and BD and patients screening positive will be recruited. We will enroll 1,000 patients (500 with PTSD and 500 with BD). A Sequential, Multiple Assignment, Randomized Trial (SMART) design will be used to compare TCC and TER, and to determine whether patients not engaging to TER improve with PER. Specifically, patients not engaging to TER by six months will be randomized a second time to either continued-TER or PER. Patients randomized to TCC will meet with an offsite telepsychiatrist consultant via interactive video at the beginning of treatment who will assign an accurate diagnosis and provide treatment recommendations for the PC providers who will retain primary responsibility for treatment. In addition, PC providers will be supported by onsite care managers who will conduct patient outreach to foster proactive communications between an activated informed patient and a coordinated care team. Patient randomize to TER will remain in the PC setting, but receive ongoing pharmacotherapy and psychotherapy from offsite MH specialists via interactive video. Patients not engaging and responding to TER who are randomized to PER will receive ongoing treatment from offsite MH specialists via phone in the comfort of their own home. We will use a pragmatic trial design, with broad inclusion criteria (screening positive for PTSD or BD) and limited exclusion criteria (already engaged in specialty MH care). Intervention fidelity will be measured, but not controlled. Patient engagement will also be measured, but not required, and intent to treat analysis will be conducted. Patients will be the unit of randomization. Mixed quantitative and qualitative methods will be used to assess self-reported outcomes. All patients will be administered surveys at baseline, 6 and 12 months by telephone to minimize patient burden and attrition. A sub-sample of patients will be invited to participate in qualitative interviews to describe their treatment experience using their own words. Likewise, PC providers will be invited to participate in qualitative interviews to voice their perspective. The primary outcome will be patient self-reported health related quality of life. Secondary outcomes include access to care, therapeutic alliance with providers, patient-centeredness, patient activation, satisfaction with care, appointment attendance, medication adherence, self-reported clinical symptoms, medication side-effects, and progress towards life goals.

Inclusion Criteria:

• Enrolled as a patient at a participating Federally Qualified Health Center
• Screen positive for Bipolar Disorder on the Composite International Diagnostic Interview (CIDI) AND/OR screen positive for PTSD on the PTSD Check List (PCL)-6

Exclusion Criteria:

• Currently prescribed a psychotropic medication by a mental health specialist.
• Lacks capacity to provide informed consent
• Does not speak English or Spanish

If requested by the funding agency (the Patient Centered Outcomes Research Institute), a complete, cleaned and de-identified copy of the final quantitative dataset used to test the stated hypotheses will be made available to other researchers within one year of the study completion date. The final data set will include de-identified demographic and clinical data obtained from the telephone survey for all patients participating in the comparative effectiveness trial. Along with the data set, we will create a code book documenting all variables (e.g., common names for single questionnaire items, and scoring algorithms for derived variables).