Accuracy of Primary Care Medical Home Designation in a Specialty Mental Health Clinic

Abstract

Objective:

To assess whether primary care medical homes (PCMHs) are accurately identified for patients receiving care in a specialty mental health clinic within an integrated public delivery system.

Methods:

This study reviewed the electronic records of patients in a large urban mental health clinic. The study defined ‘matching PCMH’ if the same primary care clinic was listed in both the mental health and medical electronic records. This study designated all others as ‘PCMH unknown.’ This study assessed whether demographic factors predicted PCMH status using chi-square tests.

Results:

Among 229 patients (66% male; mean age 49; 36% White, 30% Black, and 17% Asian), 72% had a matching PCMH. Sex, age, race, psychiatric diagnosis, and psychotropic medication use were not associated with matching PCMH.

Conclusions:

To improve care coordination and health outcomes for people with severe mental illness, greater efforts are needed to ensure the accurate designation of PCMHs in all mental health patient electronic records.

Introduction

People with severe mental illness (SMI) such as psychosis, bipolar disorder, or severe major depressive disorder have shorter life expectancies, dying on average 10–25 years earlier than the general US population [1]. High incidences of diabetes, obesity, coronary artery disease, hyperlipidemia, and cancer in the setting of limited access to preventive services and fragmentation of care contribute to this stark disparity [1,2]. Many patients with SMI are prescribed psychotropic medications, further increasing risk of cardiometabolic disease [3]. Yet patients with SMI are less likely to be screened for cardiovascular risk factors [4]. National efforts to improve preventive care and the overall health of people with SMI have promoted a shared care model, with coordination of care between specialty mental health and medical care settings, to promote cardiometabolic risk screening [3, 5–7]. While this approach has been used in the United Kingdom, fragmentation of specialty mental health and medical care settings makes this approach more difficult to implement in the United States.

A critical first step to improve coordination of care between specialty mental health and medical care is having accurate provider contact information readily available to healthcare team members [8,9]. The adoption of electronic health records to create a digital health care infrastructure is intended to improve coordination of care and facilitate communication [10]. Not surprisingly, prior research has shown that patients with SMI who receive primary care services are more likely to get cardio-metabolic screening [4,11]. Yet keeping accurate provider information available and updated is challenging because individuals with SMI receive fragmented care, with services for physical health and mental health financed and delivered under separate systems, and by multiple public and private agencies [7, 10]. Medicaid carve out and billing issues have contributed to separate electronic health records for specialty mental health and physical health. It is unknown how accurate primary care clinic identification is in mental health systems’ electronic medical records.

In this cross-sectional study, we sought to determine the accuracy of primary care medical home designation within the specialty mental health electronic health record. We hypothesized that many people with SMI would not have accurate primary care information documented in the mental health clinic’s electronic health record.

Methods

Study Design:

This cross-sectional study examined primary care medical home identification among people with SMI receiving care in a large, community specialty mental health clinic in San Francisco.

Setting and Subjects:

The clinic selected for this study is the largest specialty mental health clinic within the San Francisco City and County integrated public health delivery system, the San Francisco Health Network (SFHN). It is one of 14 specialty mental health clinics designated as the Medicaid carve-out for specialty mental health clinics in the county of San Francisco. These mental health clinics are part of an integrated healthcare system that includes a large public hospital with both psychiatric and medical emergency services, inpatient psychiatric, medical and surgical departments, as well as 20 free-standing community health centers, outpatient specialty clinics, jail health services, and a nursing home. Within this same public health care system, the specialty mental health and primary care clinics use distinct electronic health records that do not share data.

This study was part of a larger study examining the effectiveness of a novel system-level intervention to improve metabolic screening [12]. As such, this ancillary study included only subjects if they were actively enrolled in one of the five intensive case management programs in the study clinic. Intensive case management at this clinic includes direct services, multidisciplinary team-based treatment, a small caseload per case manager (10–15 patients), 24/7 coverage and crisis intervention. Other inclusion criteria for subjects were: 1) age ≥18 and; 2) received care at the clinic between October 2014 and February 2015. In order to capture all primary care and laboratory data in the electronic medical record, we further limited participants to those who receive primary care in one of the 14 primary care clinics in the San Francisco Health Network (SFHN). There were no additional exclusion criteria.

Data Collection:

The primary outcome measure was “matching primary care medical home,” defined as having the same primary care medical home listed in both the specialty mental health and the medical electronic records. The team collected designated primary care medical home from both the mental health and the medical electronic health records. Records without a primary care medical home listed in the specialty mental health record (data missing), or with a different primary care medical home than in the medical electronic health record (data discordant), were defined as “unknown primary care medical home.’ For the purposes of this study, the primary care clinic listed in the medical electronic record was considered the gold standard.

The study team extracted additional covariates, including age, sex, race/ethnicity, psychiatric diagnosis, and current prescription of psychotropic medications. Psychiatric diagnosis was a hierarchical definition, with the primary diagnosis categorized as follows: schizophrenia, bipolar disorder, major depressive disorder, personality disorder, PTSD, or other mental health diagnosis.

Analysis:

We performed descriptive analysis of the study population using chi-square tests, analyzing whether patient characteristics predicted primary care medical home status.

Results

Our study population consisted of 229 patients with SMI. Half were 50 years or older (50%), 66% were male, and the population was diverse (36% non-Hispanic White, 30% non-Hispanic Black, 17% Asian, and 17% Latinx/Other/Mixed race). Sixty-nine percent had a primary diagnosis of schizophrenia, and 87% were prescribed psychotropic medications.

Seventy-two percent (166/229) of study subjects met criteria for “matching primary care medical home.” No patient-level characteristics (race, age, psychiatric diagnosis, or taking any psychotropic medication) were associated with having a matching primary care medical home recorded in the specialty mental health electronic record (p >0.05).

Discussion

In a large, urban, public specialty mental health clinic, 28% of patients with SMI did not have a primary care medical home correctly identified in the specialty mental health electronic record. This is particularly concerning given that all of the patients in this setting received a high-level of care with intensive case management services. To improve outcomes for patient with SMI—both morbidity and mortality—we need a focus on the prevention and treatment of cardiometabolic conditions. In order to achieve this, improved communication, enhanced coordination, and clearer role definition between mental health specialists and primary care providers will be critical [7].

Efforts have been made to reduce fragmentation of care and improve linkage to medical care for patients with SMI [12–15], but a critical first step must be a clear and accurate designation of primary care medical homes in the specialty mental health records. Having inaccurate information about primary care medical homes in the specialty mental health electronic health record impedes coordination of care between primary care and specialty mental health providers for patients with SMI. While patients in our study were enrolled in an intensive case management program that seeks to coordinate and improve care for patients with SMI, more than a quarter of the patients in our study did not have a clearly identified primary care medical home. For effective communication to occur between mental health and primary care providers, everyone needs to be aware with whom they need to communicate. Case managers may have developed ‘work-arounds’ to compensate for the lack of information, or misinformation, in the health record; however, this remains a clear barrier to communication and coordination of care.

Our study has several limitations. First, it was based in a single mental health clinic, and may not be generalizable to other settings. Second, since only about 5% of the people served in specialty mental health clinics in San Francisco receive intensive case management., it is possible that we are underestimating the number of people with inaccurately designated primary care medical homes [16]. Likewise, we did not ask patients to confirm their primary care medical home, which would have represented a true gold standard, nor review charts to establish where primary care was received. We also did not have access to utilization data and therefore we are not able to assess the strength of a patient’s relationship with either a primary care medical home or the mental health clinic.

Conclusions

Our study underscores one of the potential drivers of the poor communication between specialty mental health and primary care providers in the safety net. While many current specialty mental health electronic record systems rely on individuals to update and maintain information manually, devising automated systems to ensure accurate identification of the primary care medical home would undoubtedly improve coordination of care. Policy changes to facilitate information sharing will further improve the care and ultimate health of individuals with SMI. This study should be replicated to determine the extent of misidentification of primary care medical homes in other settings without intensive case management, and the degree to which this impacts communication across specialties and clinical outcomes.

Author Biographies

Maria Esteli Garcia

As a clinician investigator, Dr. Garcia focuses on co-morbid mental health and chronic diseases and their disproportionate impact on vulnerable and marginalized populations. As an Assistant Professor in the Division of General Internal Medicine, she conducts research on mental health integration in primary care, with a focus on racially, ethnically and linguistically diverse populations. Dr. Garcia completed residency training in Internal Medicine at UCSF, in the San Francisco General Hospital Primary Care Program, and subsequently the UCSF Primary Care Research Fellowship. During her fellowship, she developed an interest in improving mental health service delivery for individuals with limited English proficiency. Dr. Garcia has researched the unique challenges that patients with co-morbid mental health and chronic diseases face and focused on implementation work in mental health integration and improvement of service delivery for populations with language barriers.

Lauren (Liz) Goldman

Dr. Goldman is a practicing internist at San Francisco General Hospital where she is the medical lead for the Emergency Department Case Management team, an innovative interdisciplinary intensive case management program assisting the most socially and medically complex clients in the San Francisco Health Network, and the medical lead for an academic clinic-based complex care management team. She also works as a Clinical Informaticist on data analytics, partnering on value based care efforts and data access for research. In addition, she supervises residents, has an outpatient primary care clinic, and attends on the inpatient service. Her research and policy work focuses on improving and transforming health systems for underserved diverse populations. She has developed and implemented systems innovations that improve care transitions, value of care, and care coordination efforts across medical and behavioral health settings.

As part of her work to improve the science of quality measurement, she led the California Office of Statewide Health Planning and Development Validation Project which assessed the accuracy of California patient discharge data, and was awarded AHRQ grant funding to evaluate the impact of administrative data quality on hospital performance metrics. Her expertise in using administrative data for quality assessments has been nationally recognized, through her participation in the AHRQ Quality Indicator Expert Panel.

Marilyn Thomas

Dr. Thomas is a postdoctoral scholar in the Department of Psychiatry at UCSF. Her passion is to reduce social and health disparities experienced by women and minorities and she completed doctoral training in epidemiology at UC Berkeley. As a social epidemiologist, she investigates the effects of structural racism on morbidity, mortality, and life expectancy. Over her career, she hopes to address racial disparities in health and in education.

Stephen Chan

Internal Medicine resident at the University of California, at Davis.

Fumi Mitsuishi

Dr. Mitsuishi is an Associate Professor in the Department of Psychiatry at the University of California in San Francisco.

Dean Schillinger

Dean Schillinger MD previously served as Chief of the UCSF Division of General Internal Medicine at San Francisco General Hospital (SFGH). In 2006, he founded the UCSF Center for Vulnerable Populations, and currently directs the UCSF Health Communications Research Program. Dr. Schillinger served as Chief Medical Officer for the Diabetes Prevention and Control Program for California from 2008–13. He previously directed the ambulatory care clinics at SFGH. He co-directs a national course on Medical Care of Vulnerable and Underserved Patients and edits a textbook of the same name. Author of over 275 peer-reviewed articles, he is an international expert in health communication and in diabetes-related healthcare and public health. He has focused his research on health communication for vulnerable populations, carrying out a number of studies exploring the impact of limited health literacy on prevention and control of diabetes and heart disease. He has been awarded grants from NIDDK, AHRQ, CDC, NLM and private foundations in the field of health communication science.

Christina Mangurian

Dr. Mangurian is a Professor of Clinical Psychiatry at UCSF School of Medicine and the UCSF Department of Psychiatry’s Vice Chair for Diversity and Health Equity. Dr. Mangurian founded and directs the UCSF Program of Research on Mental health Integration among Underserved and Minority populations (PReMIUM) which is based at CVP (http://integration.ucsf.edu/). She is a community psychiatrist whose NIH-funded research program focuses on improving diabetes screening and HIV care of people with severe mental illness (e.g., schizophrenia, bipolar disorder), particularly among underserved minority populations.

Dr. Mangurian has a successful track record in implementation in the public sector, most notably being her work as a Special Coordinator for the Medical Director of the New York State Office of Mental Health to implement health screening of 15,000 outpatients served within the New York State public mental health system (Mangurian et al., 2010). With her prior NIDDK-funded grant (Diabetes screening and care received by the mentally ill in an integrated system), Dr. Mangurian has partnered with Kaiser Permanente Northern California Division of Research to leverage Kaiser’s tremendous EHR to examine patient-, provider and clinic-level variables that might influence diabetes care in this vulnerable population. Her current NIMH-funded R01 focuses on improving the HIV care of people with schizophrenia nationally.

In addition to her research experience, Dr. Mangurian is the Director and co-Founder of the UCSF Public Psychiatry Fellowship at San Francisco General Hospital. This was the first Public Psychiatry Fellowship in California, and the only one nationally to have a formal mental health services research component. Dr. Mangurian also serves as the Chair of the American Psychiatric Association’s Council on Minority Mental Health and Health Disparities.

Dr. Mangurian received her BA in Biology from Reed College. She graduated AOA from the UCSF School of Medicine, and completed her Psychiatry Residency and Chief Residency at Columbia University. She also completed the Columbia University Public Psychiatry Fellowship. Dr. Mangurian joined the faculty at UCSF Department of Psychiatry at Zuckerberg San Francisco General in 2009. She joined the faculty of the UCSF Center for Vulnerable Populations in 2014, and the faculty of the Philip R. Lee Institute for Health Policy Studies in 2018. She received a UCSF Master’s Degree in Clinical Research, with Implementation Science Track coursework, in 2015.