A doctor is in the house: Stakeholder focus groups about expanded scope of practice of community psychiatrists

Abstract

Objective:

We sought to understand stakeholder perspectives on barriers to metabolic screening for people with severe mental illness. We additionally assessed the feasibility of expanding psychiatrists’ scope of practice to include treatment of cardiometabolic abnormalities.

Methods:

We conducted four focus groups among patients with severe mental illness, community psychiatrists, primary care providers, and public health administrators. Focus group transcripts were thematically analyzed.

Results:

Three domains emerged: challenges with patient navigation of the complex health care system, problem list prioritization difficulties, and concern that treatment of cardiometabolic abnormalities were beyond the scope of practice of psychiatrists.

Conclusions:

Stakeholders agreed that navigating the health care system was challenging for this population and led to undertreatment of cardiometabolic risk factors. Expansion of psychiatrists’ scope of practice within community mental health appears acceptable to patients and may be a mechanism to improve cardiometabolic care among people with severe mental illness.

INTRODUCTION

People with severe mental illness (e.g., schizophrenia, bipolar disorder) die on average 25 years earlier than the general population, most commonly from premature cardiovascular disease (Colton CW, 2006; Olfson, Gerhard, Huang, Crystal, & Stroup, 2015) . This mortality gap is likely due to several risk factors, including smoking, sedentary lifestyle, poor eating habits, and substance abuse (Compton, Daumit, & Druss, 2006; B. G. Druss, 2007; B. G. Druss & von Esenwein, 2006). However, another significant contributing factor is expanded use of antipsychotic medications, many of which are associated with metabolic complications (e.g., obesity, diabetes, and dyslipidemia) that increase cardiometabolic risk (Daumit, et al., 2008; Davidson, et al., 2001; Lambert, Velakoulis, & Pantelis, 2003; McEvoy, et al., 2005; Newcomer, 2005; Newcomer & Hennekens, 2007). Despite national metabolic screening guidelines (Clark, 2004), most people with severe mental illness who take antipsychotic medications are not screened for metabolic abnormalities (Essock, et al., 2009; Haupt, et al., 2009; Mangurian, et al., 2015; Morrato, et al., 2009; Morrato, et al., 2010). Furthermore, if metabolic abnormalities are identified, this population is often not treated (Nasrallah, et al., 2006; Newcomer & Hennekens, 2007).

Integration of primary care and behavioral health services represents a national goal, as evidenced by legislation passed by the US House of Representatives that would allow Medicaid to reimburse for physical and mental health services received the same day (“HR2646: Helping Families in Mental Health Crisis Act,” 2016). Most attention currently focuses on integrating mental health into primary care systems, particularly through evidence-based collaborative care (Unutzer, Katon, Callahan, Williams, Hunkeler, Harpole, Hoffing, Della Penna, Noel, Lin, Arean, Hegel, Tang, Belin, Oishi, Langston, et al., 2002). However, systemic solutions to improve primary care for people with severe mental illness served in specialty mental health settings do not yet exist (Reilly, et al., 2013). This lack of system-level models is concerning given the early morbidity in this population, and because the mental and physical health systems in the US often operate separately (geographically, culturally, electronically, and financially) (B. G. Druss, et al., 2008; B. G. Druss & von Esenwein, 2006; B. G. Druss & Walker, 2011; Lambert & Newcomer, 2009). In addition to problems inherent with parallel systems of care, our prior work has identified disagreement among primary care and psychiatry clinicians regarding which provider is expected to conduct guideline-recommended medical screening and treatment for people taking antipsychotic medications (Mangurian, et al., 2013; Parameswaran, et al., 2013).

Since people with severe mental illness have low rates of primary care utilization (CDC, 2015), but often use community mental health services (Alakeson, Frank, & Katz, 2010; B. G. Druss, et al., 2008; Steinø, Jørgensen, & Christoffersen, 2013; Wang, et al., 2005), these clinics have become de facto medical homes for this population (Amiel & Pincus, 2011). Co-located primary care providers integrated within the behavioral health team has been identified as a potential opportunity, but this is both costly and logistically challenging (Horvitz-Lennon, Kilbourne, & Pincus, 2006; Scharf, et al., 2013). Alternatively, expanding the scope of practice of community psychiatrists to take more responsibility for cardiometabolic care is reasonable, since these psychiatrists serve as the first—and often only—line of care for this vulnerable population. Though the medical home model may represent a major paradigm shift, psychiatrists prescribing statins for dyslipidemia is akin to primary care providers treating depression with selective serotonin reuptake inhibitors.

Vanderlip et al. (2016) recently proposed a framework for extending psychiatrists’ roles in treating general medical conditions, but did not assess opinions of stakeholders regarding this theoretical expansion (Vanderlip, Raney, & Druss, 2016). Our study elicited perspectives from four stakeholder groups (patients, psychiatrists, primary care clinicians, and clinic administrators) regarding barriers to care and expansion of the psychiatrist scope of practice. By conducting focus groups, we aimed to gauge acceptability of designating specialty mental health clinics as medical homes and identify barriers and facilitators to metabolic screening previously unidentified through quantitative techniques.

METHODS

Study Design

We conducted focus groups between January 1 and December 31, 2013. Focus groups with patients with severe mental illness occurred at a community mental health clinic in San Francisco. Provider focus groups were convened at a San Francisco Department of Public Health (SFDPH)-supported housing conference room. Clinic administrator focus groups were convened at SFDPH administrative offices.

Participants

Patients with severe mental illness (SMI):

We recruited adult patients attending a large SFDPH-affiliated community mental health clinic that serves some of the county’s most psychiatrically ill patients. Research staff described the study to potential participants during a regularly-scheduled clinic meeting. Interested individuals were then encouraged to approach research staff to be considered for participation, and only those who could not provide informed consent were excluded. The clinically-validated MacArthur Competence Assessment Tool for Clinical Research (MacCAT-CR) was used to ensure that each person had adequate understanding of the research and sufficient capacity to consent (Appelbaum, 2001; Palmer, et al., 2005). Those who expressed interest and had capacity to consent were invited to participate in the focus group that afternoon. Potential participants were assessed again for capacity for informed consent immediately prior to their participation.

Providers (community psychiatrists and primary care physicians):

Email addresses for all community outpatient psychiatrists and primary care providers in San Francisco County were acquired with permission from SFDPH and supplemented with information from medical directors of individual clinics. We stratified by clinic and then randomized assigned numbers to potential subjects. Primary care providers at each clinic were invited in numerical order until one accepted or until all providers from that clinic had been invited to participate. Physicians were included if they worked for the SFDPH or an SFDPH-affiliated clinic, spent at least 5% of their time working with people who had SMI, and primarily worked with adults (≥18y/o). Fellows and residents, those who primarily worked in substance abuse treatment programs or children’s programs, and psychiatrists who worked in a behavioral and mental health co-located clinics were excluded, as these groups went beyond the study’s scope.

Administrators:

SFDPH Administrative Directors of primary care outpatient services, behavioral health outpatient services, integration, pharmacy, and information technology were invited via email, and all agreed to participate.

All participants provided written informed consent. Participants were compensated with refreshments and either $20 Walgreens® or Starbucks® gift cards. All activities were approved by UCSF’s Committee on Human Research (12–09789 and 13–10557).

Measures

We developed focus group questions based on results of our previous survey studies of psychiatrists and primary care clinicians (Mangurian, et al., 2013; Parameswaran, et al., 2013). The focus group topics emphasized barriers to cardiometabolic screening and acceptability of expansion of psychiatrists’ scope of practice. Focus group topics were divided into three sections: cardiometabolic screening, treatment of metabolic abnormalities, and expanded scope of practice. Additionally, we included questions on addressing needs of vulnerable subpopulations (e.g., racial or ethnic minorities, people with limited English proficiency). Facilitators used the s focus group guide to organize the discussion, which varied slightly depending on respondent group (Supplemental eTable 1).

Procedures

We conducted four focus groups following standard methods (Morgan, 1988). All were conducted in English and moderated by the first author. For consenting patients, the clinic’s medical director provided information on whether or not they had a primary care clinician identified in the electronic medical record. The sessions lasted 90 minutes and were audio-recorded and transcribed.

Data Analysis

The research team summarized focus group data for emerging themes in Dedoose® (SocioCultural Research Consultants LLC, 2015). Analytic emphasis was placed upon description and comparison within and across groups (Stewart & Shamdasani, 1990). After deductive and inductive coding and codebook development, an “overview grid” (Knodel, 1993) was created, summarizing responses by group and topic. This grid was used to identify patterns in responses and to characterize frequency, extensiveness, intensity and consistency of responses (Kreuger, 1998). Conclusions about each theme were summarized for each group and compared between sessions. Finally, we noted opinions and attitudes that were not universally agreed upon within each group. We present shared views about barriers to screening and ways in which screening and treatment interventions could be tailored to reduce those barriers, as well as discrepant views that required further discussion.

RESULTS

Nearly half (18/40) of patients queried were interested in participating. Of those interested, 44% (8/18) did not have the capacity to provide informed consent. Of those with capacity, 80% (8/10) were able to participate in the focus group. Ninety-one percent (10/11) of those with capacity and 57% (4/7) of those without capacity had a primary care clinician identified in their medical record.

Of potentially eligible community psychiatrists, 58% (80/138) met inclusion criteria and were invited to participate. Eleven percent (9/80) of those meeting criteria expressed interest in attending the focus group, with 89% (8/9) of those ultimately consenting and participating. Of all primary care clinicians, 96% (161/168) met inclusion criteria and were initially invited to participate. Sixteen percent (26/161) responded that they were interested in participating, schedules permitting. After being emailed a schedule, 54% (14/26) of interested physicians responded. Half (7/14) of those providers could participate at the time convenient to most responders. Six (86%) were consented and participated. Among administrators, 56% (9/16) initially agreed to attend and 78% (7/9) were consented and participated (Supplemental eTable 2 provides demographic information).

Barriers to Screening

Healthcare System Navigation Difficulty

A recurring theme in each focus group was the difficulty that some people with SMI have in navigating the health care system. Whereas some patients had already established primary care clinicians, others found the complexity of identifying providers problematic. Patients reported being overwhelmed with complex healthcare options, and stated that any assistance in selecting doctors and making appointments would be extremely helpful (“they sent me a big, big book [of doctors] and I just got confused”). Even among people with SMI who had primary care providers and were navigating the health care system, they reported a preference for a behavioral health medical home. Some psychiatrists also recognized the utility of behavioral health homes to help start necessary medication early, especially given long waiting lists for primary care providers. All provider groups acknowledged the problem of navigating the system of care, with a primary care provider recognizing that “…it’s difficult for our patients to go to the lab even though it’s just across the way.” An administrator said plainly: “there’s definitely a problem with access of referring a person to a [primary care] practitioner in a timely manner.”

Lack of Prioritization of Metabolic Screening

Psychiatrists consistently noted that prioritization of metabolic screening and care quickly diminished in light of more pressing issues. One psychiatrist noted:

“A lot of patients… have multiple semi-crisis issues in one visit, and lipids seem to be drifting way down the priority list. Without realizing it, multiple visits occur; each of them oriented toward crises, without even just basic health care maintenance much less lipids floating to the top.”

Another psychiatrist acknowledged a difference based on functioning: “I find that the lower-functioning the patients are the less that I talk to them about [diet and exercise].” In addition, the administrative group also acknowledged that lack of reimbursement for metabolic screening impacted prioritization of screening, saying, “[Medicaid] will deny any services that are documented in such a way that looks like it’s providing a medical service…. It puts a lot of stress on the staff.”

Expansion of Scope of Practice

When discussing expansion of psychiatric scope of practice to include administering medications for some metabolic disorders, patients believed that psychiatrists could succeed with training, and many patients would rather attend one comprehensive doctor’s visit that could incorporate various services. One participant captured the overall sentiment: “A doctor is a doctor to me.” Another patient thought this expanded scope of practice would actuality reduce his anxiety: “It will actually ease my mind just being able to do things through one doctor and not have to go here and there… I’m like a chicken with my head cut off going to three, four different doctors.”

Psychiatrists had mixed opinions regarding expanding their scope of practice, saying “I think many people do not want to go down that path and will be quite resistant to being asked to do so.” Others had reservations around ongoing management of these cardiometabolic risk factors: “If I ignore it, that could be a problem. If I prescribe something and it’s out of my scope—that could be a problem. And my question would be: Where does it end? Like, how far down the algorithm are we comfortable going?” Some psychiatrists agreed that prescribing medications for some cardiometabolic disorders should be within their scope of practice given the side effects of antipsychotic medications: “I feel like [prescribing medications for metabolic disorders] would be a good idea for us to do, especially since our meds are responsible for the [metabolic disorders] that we are seeing. So I feel a responsibility to do it.” However, they all recognized the need for additional training to do this safely.

The administrative group also found value in providing these services in behavioral health, with one stating “It’s kind of like striking while the iron is hot, when the person has the motivation to seek services.” However, one administrator captured the double-edged sword of encouraging this expanded scope of practice: “So, imagine that we actually all start to do this and we build a competency around prescribing statins and somehow that may inhibit the patient from going to a [primary care] medical home because everything is covered. And if we inhibit them from going to a [primary care] medical home, are they getting the best care? Should we be the medical home? Can we be the medical home?”

DISCUSSION

These focus groups highlight the potential opportunity offered by behavioral health medical homes in the effort to overcome patient, provider, and systemic challenges to cardiometabolic care for people with severe mental illness.

The US health care system is challenging even for high functioning patients (Rein, 2007). Though outliers exist, most patients report difficulty navigating the healthcare systems, and both primary care providers and psychiatrists agree that the current healthcare system is confusing and burdensome. When compounded by the problems in thinking that accompany severe psychiatric disorders, the system can become prohibitively complex. Our findings indicate that some people with SMI are very knowledgeable of health care coordination and have strong, valuable relationships with their primary care providers and psychiatrists. Even among those who can navigate the system, some struggle with the need to see different doctors for various medical problems. Other people with SMI are far too psychiatrically ill to coordinate their own care, and providers noted that metabolic screening rarely was the primary purpose of healthcare visits. As such, accessing primary care remained a low priority. A behavioral health home model appears to be an acceptable and practical solution for patients who cannot navigate complex health care systems. Unfortunately, many providers and administrators appear resistant to changing the system to accommodate the needs and preferences of this vulnerable population.

While some psychiatrists recognized the desirability of expanding the scope of practice for community psychiatrists to treat metabolic abnormalities, primary care providers were hesitant. Though patients overwhelmingly agreed that such a concept was both palatable and preferred, providers were less enthusiastic, and many felt that they either lacked necessary training to feel comfortable prescribing such medication, or that such culture change within the field would be met with resistance. Although this shift in culture would be no small change, it would create a more patient-centered approach to treat patients in one setting and take advantage of internal resources (staff psychiatrists with medical training).

This study is limited in that all participants received care within the same safety net system. Though large and diverse, it would be ideal to replicate this study in another setting. In addition, those who are too psychiatrically ill to consent for participation are critical target populations. Our data indicate that these people are less likely to have a primary care provider than people with consent capacity (57% vs. 91%). Future studies should determine barriers and preferences of these particularly high-risk patients who likely are most needing of help. Similarly, psychiatrists and primary care physicians who participated likely do not comprehensively represent the views of their discipline, thus hindering generalizability of our results.

US health care systems are exceedingly difficult for people with SMI to navigate. Novel approaches such as expansion of scope of practice among community psychiatrists and reorientation of community mental health clinics to serve as medical homes could help address early morbidity and mortality in this vulnerable population. To accomplish this, we have several recommendations. First, we strongly believe that policymakers should consider adding a consulting internist to clinics, just as collaborative care adds a consulting psychiatrist (Alakeson, et al., 2010; Amiel & Pincus, 2011; Unutzer, Katon, Callahan, Williams, Hunkeler, Harpole, Hoffing, Della Penna, Noel, Lin, Arean, Hegel, Tang, Belin, Oishi, & Langston, 2002), so that providers could advise on preliminary treatment of common primary care issues among patients who have challenges accessing traditional primary care. Second, we suggest expanding psychiatric residency programs to include comprehensive training on screening and preliminary treatment of cardiometabolic risk factors, as this would undoubtedly benefit this vulnerable population. Since these recommendations would require health care delivery changes, we also recommend altering Medicaid reimbursement policies so that providers can bill for these important activities. Finally, despite several models being tested and developed (Benjamin G Druss, et al., 2016; Scharf DM, 2014), there is still no strong evidence base supporting systemic solutions to improve primary care for people with severe mental illness served in specialty mental health settings (Reilly, et al., 2013). As such, we recommend further research in this important area.