Telementoring Primary Care Clinicians to Improve Geriatric Mental Health Care

Abstract

Health care delivery and payment systems are moving rapidly toward value-based care. To be successful in this new environment, providers must consistently deliver high-quality, evidence-based, and coordinated care to patients. This study assesses whether Project ECHO^® (Extension for Community Healthcare Outcomes) GEMH (geriatric mental health)—a remote learning and mentoring program—is an effective strategy to address geriatric mental health challenges in rural and underserved communities. Thirty-three teleECHO clinic sessions connecting a team of specialists to 54 primary care and case management spoke sites (approximately 154 participants) were conducted in 10 New York counties from late 2014 to early 2016. The curriculum consisted of case presentations and didactic lessons on best practices related to geriatric mental health care. Twenty-six interviews with program participants were conducted to explore changes in geriatric mental health care knowledge and treatment practices. Health insurance claims data were analyzed to assess changes in health care utilization and costs before and after program implementation. Findings from interviews suggest that the program led to improvements in clinician geriatric mental health care knowledge and treatment practices. Claims data analysis suggests that emergency room costs decreased for patients with mental health diagnoses. Patients without a mental health diagnosis had more outpatient visits and higher prescription and outpatient costs. Telementoring programs such as Project ECHO GEMH may effectively build the capacity of frontline clinicians to deliver high-quality, evidence-based care to older adults with mental health conditions and may contribute to the transformation of health care delivery systems from volume to value.

Introduction

Health care delivery and payment systems in the United States are moving rapidly toward value-based care as a result of state and federal health care reform efforts.¹ New health care management models (eg, accountable care organizations, patient-centered medical homes) require clinicians to be more aware of the full spectrum of care and costs for patients. To be successful, providers operating in a value-based environment need to be able to consistently deliver high-quality, evidence-based and coordinated care to patients. Given that an estimated 26% of the US adult population have a behavioral health condition (and the fact that individuals with such conditions are significantly more likely to have decreased physical health and poor treatment adherence), it is widely recognized that providing effective behavioral health care in these settings is critical.² Yet, few of the new health care management models have fully integrated behavioral health into their comprehensive approach to health care delivery.³

The shortage of mental health clinicians in the United States makes the delivery of consistent, evidence-based mental health care a major challenge. Approximately 89.3 million Americans currently live in the 4000 communities designated as Health Professional Shortage Areas that lack a sufficient number of mental health care experts to address the needs of these communities.⁴ These challenges are more salient in rural areas, where the shortage of mental health counselors is particularly severe and the population is aging rapidly.^5,6 In these settings, primary care clinicians (PCCs) are often the frontline mental health care workforce despite limited training in effectively diagnosing and treating mental health conditions.⁷

Remote learning and mentoring programs offer a promising approach to building the capacity of PCCs to better address geriatric mental health challenges. In particular, Project ECHO (Extension for Community Healthcare Outcomes)—a collaborative care management, mentoring, and education model designed by specialists at the University of New Mexico—has a proven track record of improving treatment for chronic and complex health conditions.^7,8 The Project ECHO model uses web-based videoconferencing to host teleECHO clinic sessions during which clinicians at health care worksites (spokes) are trained by an interdisciplinary team of specialists at academic medical centers (hubs) using didactic presentations and case-based learning. This study evaluates whether use of the Project ECHO model is an effective strategy to address geriatric mental health challenges in rural and underserved communities.

Methods

From September 2014 to February 2016, the University of Rochester Medical Center (URMC) implemented a Project ECHO geriatric mental health (GEMH) hub connecting a team of URMC specialists (geriatric psychiatry and medicine, nursing, social work, psychology, and pharmacy) to spokes of primary care and social service sites. Thirty-three teleECHO clinic sessions were conducted during the 2-year project. Fifty-four primary care and case management spoke sites in 10 New York counties joined the program, and approximately 154 unique individuals participated over its course. The curriculum consisted of case presentations and didactic lessons that provided participants with information on screening, treatment, and diagnosis of geriatric mental health conditions with a focus on issues faced by older adults related to depression, anxiety, and dementia. Didactic portions of the clinics focused on education and best treatment practices for medication therapies, behavioral interventions, social services, caregiver support, and sleep hygiene. Case presentations reviewed patients with a range of mental and physical health conditions and comorbidities, including but not limited to depression, generalized anxiety, post-traumatic stress disorder, Alzheimer's disease, and Lewy Body Dementia. Subsequent discussions focused on treatment options for the case at hand.

To understand the short-term impact of the program, this study explored changes in the GEMH knowledge, confidence, and treatment practices of participants. It also examined satisfaction with the program and obtained health insurance claims data from a private payer to assess changes in health care utilization and costs before and after the implementation of Project ECHO GEMH. This study was approved by The New York Academy of Medicine Institutional Review Board.

Interviews

Semistructured interviews were conducted with individuals who had participated in at least 1 teleECHO clinic to explore their perceptions and experiences in the program. Interviews were conducted between May and December 2015 with 26 participants, and they ranged from 15 to 59 minutes in length. Once completed, interviews were professionally transcribed and researchers coded and analyzed data according to pre-existing and emerging themes using NVivo 8.0 software (QSR International Pty Ltd, Doncaster, Victoria, Australia). All interviewees provided verbal informed consent.

Interviewees included physicians, nurse practitioners, physician assistants, social workers, care managers, pharmacy and medical students, and program managers. Practice settings ranged in terms of size and geographic location and included small practices, large academic medical centers, and government social service programs (Table 1).

Table 1.

Interview Participant Characteristics

Profession	Number of participants (N = 26)	Percent of total
Physician	4	15
Nurse practitioner	4	15
Physician assistant	1	4
Care manager/case worker	10	38
Student	5	19
Program manager	2	8
Practice setting
Hospital	12	46
Private practice (ACO-affiliated)	7	27
Private practice or FQHC	2	8
Older adults-focused social service or long-term care setting	4	15
Other	1	4
Geographic location
Rural	7	27
Nonrural (small or medium-sized city or metro area)	19	73

ACO, accountable care organization; FQHC, Federally Qualified Health Center.

Health insurance claims data

Claims data were used to examine the impact of practice participation in Project ECHO GEMH on health care utilization and costs. De-identified data were obtained from 1 health plan for beneficiaries attributed to 35 practices that had signed Memorandums of Understanding with URMC's Project ECHO GEMH team. Beneficiaries aged 65 years or older who had visited participating practices during the study period were included. Using the practice as the unit of analysis, data were reviewed from claims for services rendered 6 months before and 6 months after the practice joined Project ECHO GEMH (dates ranged from August 2014 to August 2015) to assess changes in average health care utilization and costs per patient. Patients were attributed to a practice based on the first medical or drug claim incurred after the practice joined the program.

The study team explored health insurance claims data for patients ages 65 years or older who had been diagnosed with a mental health condition of interest or had filled a prescription for a relevant psychotropic medication (antipsychotics, antidepressants/stimulants, antianxiety agents, and medications used to treat attention and substance use disorders). For this study, patients with mental health conditions of interest were identified using International Classification of Diseases diagnoses in claims data that corresponded to the following categories of mental health diagnoses: neurocognitive disorders, depressive disorders and anxiety and related disorders (consisting of anxiety disorders, obsessive-compulsive and related disorders, and trauma and stressor-related disorders) defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.⁹ The team also obtained data for those beneficiaries older than age 65 who had not received a relevant mental health diagnosis or filled a relevant prescription. This allowed for exploration of (1) the targeted impact of the intervention and (2) how the program may have impacted other patients receiving health care services.

Results

Interviews with Project ECHO GEMH participants

Most interviewees felt the program was valuable, although many offered suggestions for improvement. Rural clinicians and those in small cities consistently found the program beneficial and viewed it as a valuable resource that helped them provide better care for patients. Participants with greater access to professional support—either because they were students or were clinicians closely affiliated with a larger hospital system—were more varied in their support of the program, although most still saw value in participating.

Impact on knowledge and confidence

Participants reported noticeable improvements in their knowledge and confidence related to treating patients with GEMH needs after participating in Project ECHO GEMH. They felt that attending clinics was an effective way to stay up-to-date and/or learn best practices in their own and other fields. Discussions around medications and nonpharmacological treatments for GEMH patients were most frequently seen as valuable. Other topics highlighted included diagnosis of GEMH conditions and cognitive exams, end-of-life care, care related to sleep disturbances, and resources for social support:

[The ECHO GEMH specialist] talked a lot more about behavioral interventions as opposed to medication… In the nursing home setting, especially, that was really valuable to hear because as I said, there's so much regulation about what medications you are and are not allowed to use… Everybody wants you to do behavioral interventions first, which I think are generally appropriate. But as physicians…we're not so trained in behavioral interventions. (Student/Fellow)

Many interviewees noted that they had begun putting new knowledge into practice. They reported implementing ECHO GEMH lessons related to sleep management, medication adjustments, nonpharmacological interventions, screening tools, and diagnostic assessments. Those who had not yet implemented ECHO GEMH recommendations explained they had not yet had an appropriate opportunity.

We did the mini-cog exam, how to do a mini-cog. Oh, there was something I learned and I've actually used it on two patients. And I didn't know about it before that. So that was very helpful. (Care Manager)

Impact on patient health outcomes

When they were able to implement recommendations, interviewees generally noticed improvements in patient health outcomes. Positive outcomes attributed to lessons from Project ECHO GEMH included improved quality of life, improved social and behavioral patterns, fewer psychotic symptoms, and improved relationships with caregivers:

The patient actually did very, very well. She was paranoid, and isolating herself in the room, and when we started to implement the various recommendations [from the ECHO team], she was able to start coming out of her room. She was now participating with some [nursing home activities]. She was going down to the cafeteria with her husband. This is a huge breakthrough for her. (Nurse Practitioner)

Impact on professional support community

Participants frequently reported that the program increased their sense of professional support, both because of their engagement in the teleECHO clinic sessions and their newly formed professional relationships with specialists in the ECHO community. This increased access to specialists was particularly beneficial for clinicians in rural areas:

I think it has [increased my sense of professional support], just because the capacity in which they function is not something that we've had access to before. It's nice to know that should anything come up, I have a group of people I can contact and say, “What do you think about this?” I can go outside of our department to folks who have more specialized knowledge and experience. (Social Worker)

Some participants noted the value of gaining access to specific types of specialist clinicians, such as a pharmacist with in-depth knowledge of medication management for psychiatric patients or a social worker who could provide insight on housing options. However, for many, it was the interdisciplinary nature of the team that made the ECHO GEMH program unique and effective:

We get the pharmacologist; there's a social worker, a psychiatrist, and all those people together and it's just a nice way to hear different people's slants on whatever case you're presenting or whatever case is being presented. [Without the program,] I would have to make three or four or five different phone calls. (Physician)

Knowledge dissemination and broader practice impacts

Participants varied in the extent to which they felt knowledge from teleECHO sessions spread beyond themselves to their colleagues. Some reported that they shared the information from the program with others in their care setting and a few noticed changes in the practices of those clinicians. Perceived barriers to knowledge spread included a lack of clinician participation (often because of time limitations), a delay between receiving recommendations and seeing patients, and a lack of openness to new information or advice among particular colleagues:

What I'll do is, I'll get the recommendations. I'll go back to the other NP, PAs or our docs at the facility, especially in our community, and discuss the case and discuss the recommendations, and help them understand… I try to go back and talk to the providers I work with. (Nurse Practitioner)

Health care utilization and costs

The study team analyzed data on health care utilization and costs before and after the implementation of Project ECHO GEMH at the 35 spoke sites for which health plan data were available. The number of health plan beneficiaries (aged 65 years and older) attributed to each of the sites during the 6 months before joining Project ECHO GEMH ranged from 11 to 20,643 (median 561); the number attributed to each practice in the 6 months after they joined Project ECHO GEMH ranged from 11 to 19,410 (median 577). During both periods, an average of 17% of patients at each site had a mental health diagnosis. Of the 3 conditions of interest, depressive disorders were the most common class of mental health diagnoses at 34 out of 35 practices. Table 2 presents additional information on the mental health conditions of patients included in the quantitative analyses.

Table 2.

Mental Health Diagnoses for Patients Served by Participating Practices

Diagnosis	Before ECHO GEMH	After ECHO GEMH
Median number of health plan beneficiaries (aged 65+) attributed to each participating practice during the study period	561	577
Average percentage of health plan beneficiaries (aged 65+) attributed to each practice diagnosed with a mental health condition^a (%)	17	17
…with a depressive disorder	10	11
…with an anxiety-related disorder^a	4	4
…with a neurocognitive disorder	3	3

For the purposes of this analysis, patients with anxiety-related disorders include those with diagnoses of anxiety, obsessive-compulsive, or trauma- and stressor-related disorders, or those who have filled prescriptions for a medication that treats one of these mental health conditions.

ECHO, Extension for Community Healthcare Outcomes; GEMH, geriatric mental health.

Table 3 reports the health care utilization and costs for patients with a mental health condition of interest. There was a 24% reduction in costs associated with emergency room use (P = .049). For all other metrics (except antipsychotic prescriptions) utilization and costs trended downward, but these results were not statistically significant at P < .05.

Table 3.

Utilization and Costs for Patients with a Mental Health Condition of Interest

Utilization variables (average use per patient)	Mean: before ECHO GEMH	Mean: after ECHO GEMH	% Change	P
Inpatient hospitalizations	0.276	0.23	−16	0.21
Outpatient visits	4.175	3.93	−6	0.29
Emergency room visits	0.829	0.67	−20	0.08
Antipsychotics	0.372	0.41	10	0.26
Benzodiazepines	0.590	0.56	−4	0.60
Cost variables (average cost per patient, $)
Inpatient hospitalization costs	$2560.72	$2198.09	−14	0.22
Outpatient costs	$1405.57	$1402.93	0	0.98
Emergency room costs	$406.37	$310.71	−24	0.049
Prescription costs	$1938.35	$1712.85	−12	0.60
Total costs	$7725.34	$7142.97	−8	0.16

Patients who have a mental health condition of interest are those who have been diagnosed with a depressive disorder, neurocognitive disorder, or an anxiety-related disorder (including anxiety, obsessive-compulsive, and trauma- and stressor-related disorders), or who have filled a relevant psychotropic medication.

ECHO, Extension for Community Healthcare Outcomes; GEMH, geriatric mental health.

Table 4 reports the health care utilization and costs for patients who did not have a diagnosis of the GEMH conditions selected. Patients at participating practices without a GEMH diagnosis experienced a 12% increase (P = .02) in outpatient visits and a 25% increase in outpatient costs (P = .000). Patients without a mental health condition of interest also experienced a 24% increase in prescription costs (P = .004). All other metrics (with the exception of inpatient hospitalizations) trended in an upward direction but were not statistically significant at P < .05.

Table 4.

Utilization and Costs for Patients with No Diagnosis of a Mental Health Condition of Interest

Utilization variables (average use per patient)	Mean: before ECHO GEMH	Mean: after ECHO GEMH	% Change	P
Inpatient hospitalizations	0.162	0.146	−10	0.75
Outpatient visits	3.198	3.591	12	0.019
Emergency room visits	0.397	0.408	3	0.84
Cost variables (average cost per patient, $)
Inpatient hospitalization costs	$1804.10	$1556.49	−14	0.68
Outpatient costs	$892.84	$1119.42	25	0.000
Emergency room costs	$192.74	$198.78	3	0.83
Prescription costs	$1097.10	$1355.29	24	0.004
Total costs	$5126.86	$5584.46	9	0.53

ECHO, Extension for Community Healthcare Outcomes; GEMH, geriatric mental health.

Discussion

Findings from this study suggest that remote learning programs such as Project ECHO GEMH may offer an effective approach to building the capacity of frontline clinicians to deliver high-quality, evidence-based care to older adults with mental health care needs. Among the participants interviewed for this study, the program led to improvements in clinicians' self-reported GEMH knowledge and confidence, encouraged changes in treatment practices, increased access to interdisciplinary professional support, and improved health outcomes for patients facing a myriad of health challenges. Benefits accrued across the range of professionals participating in the program, including physicians, nurse practitioners, care managers, and social workers.

The health care system is in a constant state of transformation, making it difficult to ascribe changes and outcomes to a specific program without a randomized controlled trial. Still, results from this health insurance claims data analysis support the qualitative findings and suggest that—among the 35 practices included in the analyses—costs associated with use of the emergency room decreased by 24% for patients with a mental health diagnosis when comparing the 6 months before the practice began ECHO GEMH to the 6 months after they joined. For those patients with GEMH conditions, most other cost and use indicators trended downward but were not statistically significant at P < .05. These findings suggest that GEMH patients were better managed in the community practice setting, which enabled them to avoid emergency room presentations related to mental health. Although additional research is needed, these findings are consistent those reported by others who have found that care models in which mental health care is provided in integrated care settings can lead to improved patient health outcomes and a consistent reduction in costs (usually 12%–15% cost improvements).^10,11

Conversely, patients at participating practices without a mental health diagnosis experienced a 12% increase in outpatient visits, a 25% increase in outpatient costs, and a 24% increase in prescription costs. These results may be reflective of a co-occurring shift toward models of care that increase reliance on PCCs in managing complex patients; that is, Project ECHO GEMH may have mitigated the impact of external factors (eg, state and federal programs and initiatives) shifting health care utilization toward a higher reliance on primary care while also better equipping these practices to achieve the goals of these new care models (eg, reduced reliance on high cost services, such as emergency department visits).

The results presented here suggest that telementoring models such as Project ECHO GEMH have the potential to play an important role in transforming health care delivery systems from volume to value. As the health care landscape becomes more integrated and value oriented, programs that spread best practices beyond academic medical centers to clinicians serving patients in rural and underserved communities will be important tools for organizations looking to improve their population health management practices. This is especially true for older adults with mental health conditions who often have complex medical needs that result in high medical costs.

Strengths and limitations

The mixed-method approach is a major strength of this study, but both the qualitative and quantitative components of the work had important limitations. First, the study team was only able to interview a small portion of program participants and interviews were voluntary. Second, the team only had access to utilization and cost data from a single health plan. Although this health plan provides coverage to a substantial portion of upstate New Yorkers, utilizing data from only 1 health plan means that data were missing on many of the patients who may have been impacted by this intervention. Moreover, health plan data were available at the practice level only (eg, average utilization and cost of patients in a given practice), limiting the ability to understand health care utilization and cost patterns for individual clinicians or patients. Still, the study team projected the estimated sample sizes required to demonstrate statistically significant changes in health care utilization and cost outcomes associated with telementoring. For example, the estimated patient-level sample size required to see the 14% observed cost reduction in average inpatient/hospitalization costs for those with no mental health diagnosis (assuming a power of .80, an alpha of .05 and using the actual standard deviations and the pre-post correlation coefficient) would have been 1627. Thus, with individual-level data this would have shown more statistically significant results (in total, more than 70,000 patients were included in the study aggregate health insurance claims data), but the study team used the more conservative approach of using the number of practices as the sample size as the data were available only at that level. Additionally, Project ECHO GEMH focused on select mental health disorders; conditions targeted in other teleECHO clinics, such as alcohol and prescription drug misuse, may further influence outcomes of interest.

Conclusion

Additional research is needed to better understand the potential impact of programs such as Project ECHO on patient health outcomes, as well as long-term impacts on health care utilization and costs. Moreover, building the capacity of PCCs and other frontline staff to treat mental health conditions in older adults may be particularly valuable given the persistent stigma related to seeking mental health care among older generations.^12,13 Future research should explore opportunities to optimize and sustain clinician engagement, effects of program dosage, and long-term impacts.

Footnotes

Author Disclosure Statement

The authors declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors received the following financial support: Support for this work was provided by The New York State Health Foundation (NYSHealth) and the Health Foundation for Western and Central New York (Grant Nos. 5-23331 [NYSHF] and 1027 [HFWCNY]). The views presented here are those of the authors and do not necessarily reflect those of the foundations or their directors, officers, and staff.

References

Abrams

, Nuzum

, Zezza

, Ryan

, Kiszla

, Guterman

. The Affordable Care Act's payment and delivery system reforms: a progress report at five years. Issue Brief, 2015; 12:1–16.

Kessler

, Chiu

, Demler

, Merikangas

, Walters

. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry, 2005; 62:617–627.

Lewis

, Colla

, Tierney

, Van Citters

, Fisher

, Meara

. Few ACOs pursue innovative models that integrate care for mental illness and substance abuse with primary care. Health Aff, 2014; 33:1808–1816.

US Department of Health and Human Services, Health Resources and Services Administration. Shortage Designation: Health Professional Shortage Areas & Medically Underserved Areas/Populations. www.hrsa.gov/shortage/ Accessed September 12, 2016 .

Smalley

, Yancey

, Warren

, Naufel

, Ryan

, Pugh

. Rural mental health and psychological treatment: a review for practitioners. J Clin Psychol, 2010; 66:479–489.

Glasgow

, Brown

. Rural ageing in the United States: trends and contexts. J Rural Stud, 2012; 28:422–431.

Scott

, Unruh

, Catlin

, et al. Project ECHO: a model for complex, chronic care in the Pacific Northwest region of the United States. J Telemed Telecare, 2012; 18:481–484.

Arora

, Thornton

, Murata

, et al. Outcomes of treatment for hepatitis C virus infection by primary care providers. N Engl J Med, 2011; 364:2199–2207.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 5th edition. Arlington, VA: American Psychiatric Publishing, 2013.

10.

Chiles

, Lampert

, Hatch

. The impact of psychological interventions on cost offset: a meta-analytic review. Clin Psychol Sci Pract, 1999; 6:204–220.

11.

Coventry

, Hudson

, Kontopantelis

, et al. Characteristics of effective collaborative care for treatment of depression: a systematic review and meta-regression of 74 randomised controlled trials. PLoS One, 2014; 9:e108114.

12.

Bartels

. Improving the United States' system of care for older adults with mental illness. Findings and recommendations for the President's New Freedom Commission on Mental Health. Am J Geriatr Psychiatry, 2003; 11:486–497.

13.

Conner

, Copeland

, Grote

, et al. Mental health treatment seeking among older adults with depression: the impact of stigma and race. Am J Geriatr Psychiatry, 2010; 18:531–543.