Integrating Behavioral Health and Primary Care for Comorbid Behavioral and Medical Problems

Arthritis, Asthma, Chronic Obstructive Lung Disease, Diabetes, Heart Failure, Hypertension, Anxiety, Chronic Pain, Depression, Fibromyalgia, Insomnia, Irritable Bowel Syndrome, Problem Drinking, Substance Use Disorder

Behavioral problems are part of many of the chronic diseases that cause the majority of illness, disability and death. Tobacco, diet, physical inactivity, alcohol, drug abuse, failure to take treatment, sleep problems, anxiety, depression, and stress are major issues, especially when chronic medical problems such as heart disease, lung disease, diabetes, or kidney disease are also present. These behavioral problems can often be helped, but the current health care system doesn't do a good job of getting the right care to these patients. Behavioral health includes mental health care, substance abuse care, health behavior change, and attention to family and other psychological and social factors. Many people with behavioral health needs present to primary care and may be referred to mental health or substance abuse specialists, but this method is often unacceptable to patients. Two newer ways have been proposed for helping these patients. In co-location, a behavioral health clinician (such as a Psychologist or Social Worker) is located in or near the primary practice to increase the chance that the patient will make it to treatment. In Integrated Behavioral Health (IBH), a Behavioral Health Clinician is specially trained to work closely with the medical provider as a full member of the primary treatment team. The research question is: Does increased integration of evidence-supported behavioral health and primary care services, compared to simple co-location of providers, improve outcomes? The key decision affected by the research is at the practice level: whether and how to use behavioral health services. The investigators plan to do a randomized, parallel group clustered study of 3,000 subjects in 40 practices with co-located behavioral health services. Practices randomized to the active intervention will convert to IBH using a practice improvement method that has helped in other settings. The investigators will measure the health status of patients in each practice before and after they start using IBH. The investigators will compare the change in those outcomes to health status changes of patients in practices who have not yet started using IBH. The investigators plan to study adults who have both medical and behavioral problems, and get their care in Family Medicine clinics, General Internal Medicine practices, and Community Health Centers.

The chronic diseases that drive the majority of mortality, morbidity and cost in America and around the globe are largely behavioral in origin or management. Tobacco, diet, physical inactivity, alcohol, substance abuse, non-adherence to treatment, insomnia, anxiety, depression, and stress are major causes of morbidity, mortality and expense, especially when chronic medical problems such as heart disease, lung disease, diabetes, or arthritis are also present. Behavioral problems can often be effectively managed with improved outcomes for patients, their families and the health care system, but the current health care system is often unable to provide such care. Behavioral Health includes mental health care, substance abuse care, health behavior change, and attention to family and other psychosocial factors. Many people with behavioral health needs present to primary care and may be referred to mental health or substance abuse specialists, but this method is often unacceptable to patients. Two newer models have been proposed for helping these patients. In co-location, a behavioral health provider is located in or near the primary practice to increase the likelihood of successful referral and treatment initiation. An alternative is Integrated Behavioral Health (IBH) in which a Behavioral Health Clinician is specially trained to work closely with the medical provider as a full member of the primary treatment team. Although it is clear that the status quo of under-diagnosis or inadequate referral and treatment is not acceptable, it is not known which of the alternative models is best. The research question is: Does increased integration of evidence-supported behavioral health and primary care services, compared to simple co-location of providers, improve patient-centered outcomes in patients with multiple morbidities? The key health decisions affected by the research are those made at the practice level: whether and how best to incorporate behavioral health (BH) services. At the patient level, the decision of whether to seek out or accept offered BH services will be influenced by the manner they are made available. Aim 1: Determine if increased integration of evidence-supported behavioral health and primary care results in better patient-centered outcomes than simple co-location of behavioral providers without systematic integration. Aim 2: Determine if structured improvement process techniques are effective in increasing BH integration. Aim 3: Explore how contextual factors affect the implementation and patient centeredness of integrated BH care. Aim 4: Assess the costs of implementing integration in this setting. Aim 5: Covid Enhancement: What is the effect of the pandemic upon the effectiveness of IBH? This is a prospective, cluster-randomized, mixed methods comparison of co-location of BH services vs. IBH in 3,000 subjects in 40 primary care practices around the US. Usual care (the control comparator) for practices attempting to deliver BH services is co-location of a BH clinician within or adjacent to the primary care facility, without increased integration. The active comparator (the intervention) is Integrated Behavioral Health to support the delivery of protocol supported, stepped, data-driven, evidence-supported, BH care. In both cases, the expenses (such as salaries for the Behavioral Health Clinicians) will be paid by the practices. The intervention consists of training for practice leaders, Behavioral Health Clinicians, primary care providers, and office staff, a Structured Improvement Process support for practice redesign, and a toolkit of suggested tactics for implementing BH. The target patient population is adults with multiple comorbid medical and behavioral problems receiving services in the target practice settings: Family Medicine clinics, General Internal Medicine practices, and Community Health Centers. The investigators will enroll 40 practices from around the country to represent a broad spectrum of US primary care sites including those serving racial and ethnic minority groups, low-income groups, women, seniors, residents of rural areas, and patients with special health needs, disabilities, multiple chronic diseases, low health literacy or numeracy and/or limited English proficiency. The intervention will be directed at the practices and its impact measured in a randomly selected sample of 75 patients with behavioral health needs from each practice for a total of 3000 patients followed for 2 years. The primary outcome is the PROMIS-29, a patient-centered measure of global health and functioning. Secondary analyses will assess other outcomes important to patients as well subgroup analyses to allow exploration of what types of patients and practices benefit most from Integrated Behavioral Health. Aim 2 will study the effect of the intervention on practice structure and processes. Aim 3 will identify barriers and supports for successful integration. Aim 4 will assess costs of implementation. The analyses for Aims 1 and 2 will use generalized linear mixed models of patient health status to perform intention-to-treat analyses as a function of experimental condition (co-location vs. integration), patient characteristics, and time of measurement, with multiple measures clustered within patients and patients clustered within practices. The parameters of interest are the central tendency (mean), statistical significance (P values) and 95% confidence intervals (CI) of the adjusted change in PROMIS-29 domain score since before the intervention. Each of the 8 outcome domains in the PROMIS-29 will be modeled individually as 8 separate hypotheses with adjustment for multiple comparisons. Secondary outcomes (Communication, Empathy, Adherence, etc.) will use similar models. Aim 3 will use mixed methods analysis of surveys, focus groups, key informant interviews and other data sources to explore the relationship between the context of the intervention and the patient-centeredness of the resultant care. Aim 4 will use a survey of practice managers in a subset of practices to collect information on staffing changes,staff time, supplies and capital expenses incurred as part of implementation.

United States