The Science of Diabetes Self-Management and Care2023, Vol. 49(3) 239–246© The Author(s) 2023Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/26350106231169245journals.sagepub.com/home/tde
Abstract
Purpose: The purpose of the study was to examine differences among adult patients with diabetes who receive care through a telementoring model versus care at an academic specialty clinic on guideline-recommended diabetes care and self-management behaviors.
Methods: Endocrinology-focused Extension for Community Healthcare Outcomes (ECHO Endo) patients completed surveys assessing demographics, access to care, health care quality, and self-management behaviors at enrollment and 1 year after program enrollment. Diabetes Comprehensive Care Center (DCCC) patients completed surveys at comparable time points.
Results: At baseline, ECHO patients were less likely than DCCC patients to identify English as their primary language, have postsecondary education, and private insurance. One year postenrollment, ECHO patients visited their usual source of diabetic care more frequently. There were no differences in A1C testing or feet checking by health care professionals, but ECHO patients were less likely to report eye exams and smoking status assessment. ECHO and DCCC patients did not differ in consumption of high-fat foods and soda, physical activity, or home feet checks. ECHO patients were less likely to space carbohydrates evenly and test glucose levels and more likely to have smoked cigarettes.
Conclusions: Endo ECHO is a suitable alternative to specialty care for patients in underserved communities with restricted access to specialty care. Results support the value of the Project ECHO telementoring model in addressing barriers to high-quality care for underserved communities.
More than 37 million people in the United States live with diabetes, and prevalence continues to rise.1 Diabetes is a complex chronic disease and is often accompanied by multiple chronic comorbidities. Most patients with diabetes receive care for diabetes in primary care settings.1 Primary care providers (PCPs) are trained in the general treatment of diabetes; however, best management practices can be complex, recommendations change often, and PCPs are typically limited to short patient visits with competing comorbidities, and they can lack self-efficacy when faced with complicated medication management.2–4 Some research indicates that care provided by diabetes specialists is higher quality5,6 and leads to better outcomes than the care provided by PCPs alone7–11; however, the shortage of endocrinologists is expected to increase to about 2700 by 2025,12 and low-income individuals and those living in medically underserved areas experience additional barriers to accessing specialty care. New methods to support PCPs in the provision of optimal diabetes care are needed, particularly in underserved areas.
From Department of Population Health, New York University Grossman School of Medicine, New York, New York (Dr Berry, Dr Blecker, Dr Paul); School of Public and International Affairs, Virginia Polytechnic Institute, Blacksburg, Virginia (Ms Dávila Saad); Robert F. Wagner Graduate School of Public Service, New York University, New York, New York (Mr Billings); Division of Endocrinology, Diabetes, and Metabolism, University of New Mexico, Albuquerque, New Mexico (Dr Bouchonville); and Department of Internal Medicine, Division of Gastroenterology, University of New Mexico, Albuquerque, New Mexico (Dr Arora).
Corresponding Author:
Carolyn A. Berry, Department of Population Health, New York University Grossman School of Medicine, 180 Madison Avenue, Office 353, New York, NY 10016, USA. Email: Carolyn.berry@nyulangone.org
The Extension for Community Healthcare Outcomes (ECHO) model of care has proliferated globally as a means of enhancing primary care for complex conditions, especially for underserved and vulnerable populations, and several recent iterations of ECHO have focused on diabetes. Established in 2003 at the University of New Mexico (UNM) Health Sciences Center, Project ECHO extends the specialized knowledge of multidisciplinary care teams located in academic medical centers (“hubs”) to primary care clinics (“spokes”) through ongoing telementoring sessions. Participating PCPs and other community health care personnel gain confidence in complex disease management through these practice-based learning ECHO videoconference sessions13 and eventually become recognized throughout their clinics and communities as local resources for expert care.14 In the Project ECHO model, the specialty team does not have direct contact with patients, as in a telemedicine model; the specialty team provides mentoring and support from community providers to provide better care to their patients. In contrast to the telemedicine model in which a specialist cares for only a single patient at a time, Project ECHO actually expands capacity to manage many patients in a given clinical area through the enhancement of expertise among community providers.
The ECHO model has proven extremely popular in the United States and internationally, with more than 800 ECHO training centers in 93% of the world’s countries,15 including a growing number of diabetes-related ECHOs (diabetes ECHOs). ECHO was originally developed to enable PCPs to treat Hepatitis C cases in New Mexico but, due to widespread demand, has been adapted and expanded to enhance the care provided by PCPs across a wide range of complex, chronic conditions.16 UNM’s endocrinology-focused ECHO (Endo ECHO) connects a specialty care team at UNM with community PCPs throughout New Mexico through virtual clinics.
For Endo ECHO in particular, the multidisciplinary care team at the UNM hub consisted of adult and pediatric endocrinologists, a psychiatrist, nephrologist, pharmacist, registered nurse manager, diabetes educator, community health worker (CHW) trainer, and social worker. The spokes, 10 federally qualified health centers in New Mexico, comprised PCPs and CHWs. The community PCPs and CHWs participated in a 2-day face-to-face training at Project ECHO and received intensive hands-on training in physical examination, diabetes technology, motivational interviewing, and team building. ECHO-specific training included use of technology, delivery of case presentations, and the process of outcomes tracking. The weekly 2-hour telementoring clinics included brief didactic presentations by members of the multidisciplinary care team or invited speakers that provided interprofessional perspectives on commonly encountered endocrine disorders. Practice-based learning via case presentations followed the didactic presentations. Community PCPs presented challenging cases and received guidance from the specialty team and from community peers.
A comprehensive evaluation of the demonstration phase of Endo ECHO has demonstrated dramatic increases in diabetes-related patient reported measures,17 decreases in A1C,18 and modest changes in health care utilization.19 The work presented in this article addresses two research questions: (1) Are patients receiving care from Endo ECHO-trained providers a more underserved population than those directly receiving specialty care at the UNM Diabetes Comprehensive Care Center (DCCC)? and (2) Is the care received and outcomes experienced by patients receiving care from Endo ECHO-trained providers for 1 year comparable to those experienced by patients receiving specialty care at the UNM DCCC?
All research was approved by institutional review boards at UNM and New York University.
The study employed a quasi-experimental design with patients enrolled in Endo ECHO as the treatment group and patients treated at the UNM DCCC as the contemporaneous comparison group, serving as a benchmark of patients receiving direct specialized diabetes care. Surveys were administered to the Endo ECHO group upon enrollment in Endo ECHO and 1 year after enrollment. Surveys were administered to UNM DCCC patients at comparable time points.
Endo ECHO sample. Individuals were eligible for the study if they were 18 years or older with “complex” diabetes: type 1 diabetes or type 2 diabetes and insulin dependent and/or with an A1C of 9% or higher. Providers and staff at the participating Endo ECHO clinics recruited patients into the study. Patients were also recruited from the broader community through local ads and referral networks. Individuals were eligible to receive care from the participating provider even if they declined to enroll in the associated research. All enrolled patients were treated by a PCP participating in Endo ECHO and had at least 1 visit with a participating CHW based at the same clinic. The enrollment period began in January 2015 and ran through April 2017. Of 861 patients who participated Endo ECHO, 533 completed both baseline surveys and surveys 1 year later, for a 62% follow-up rate. The sample of patients completing both surveys did not differ in demographic characteristics from the baseline-only sample.17
DCCC sample. Providers and staff working at the DCCC clinic recruited patients into the study, and individuals had to meet the same eligibility criteria as the Endo ECHO sample. The comparison group survey period began in March 2016 and ran through October 2018. Patients were enrolled in the study at the clinic by a research coordinator. All enrolled patients continued to receive care as usual from their providers at DCCC. Among the DCCC sample, 245 patients completed baseline surveys, and 139 completed both baseline surveys and surveys 1 year later (57% follow-up rate).
CHWs at the 10 Endo ECHO spoke clinics consented patients for participation in the evaluation and administered baseline surveys upon enrollment into Endo ECHO, and a research assistant at UNM received consent from and administered the survey to DCCC patients. A research assistant surveyed patients in both groups telephonically 1 year later. All patients provided written consent to participate in the study and received a $10 incentive for each completed survey.
The baseline survey included measures of demographics, insurance status, health literacy, and history with diabetes, and the follow-up survey included outcomes reflecting access to diabetes care, quality of diabetes care, and self-care and health behaviors.
Baseline measures. The baseline survey included measures of demographic characteristics (sex, age, preferred language, race/ethnicity, educational level), insurance coverage, diabetes diagnosis, whether the patient had ever been seen by an endocrinologist, and general health status; these survey questions were modeled on the Behavioral Risk Factor Surveillance System Survey (BRFSS) Questionnaire.20 The survey also included a brief validated measure of health literacy.21
Access to diabetes care. The measures of access to care included usual source of diabetes care (developed for this survey), number of annual visits to usual source of diabetes care (BRFSS), and travel time to usual source of diabetes care (developed for this survey).
Quality of diabetes care. Quality-of-care indicators included number of A1C tests and feet checks by professionals in the previous 12 months and date of last eye exam (BRFSS) and assessment/referral for smoking (Summary of Diabetes Self-Care Activities).22
Self-care and health behaviors. Self-care items included consumption of high-fat foods, carbohydrate spacing, cigarette smoking, physical activity, feet checks in the last 7 days (Summary of Diabetes Self-Care Activities), and soda consumption in the last 30 days (BRFSS).
Bivariate tests (χ2 tests and independent groups t tests) compared Endo ECHO and DCCC scores at baseline and 1 year later. Logistic and ordinary least squares regressions assessed the contribution of baseline differences in patient characteristics to clinic group differences in quality of care and self-care behaviors 12 months postbaseline.
Five hundred thirty-three ECHO patients and 139 DCCC patients completed baseline surveys and surveys 1 year later (Table 1). At baseline, ECHO patients were much less likely to identify English as their primary language (69% ECHO vs 91% DCCC, P < .001), to have postsecondary education (40% ECHO vs 60% DCCC, P < .001), or to have private insurance (16% ECHO vs 23% DCCC, P = .04). DCCC patients were much more likely to have type I diabetes (43% vs 7%, P < .001), to have ever been seen by an endocrinologist (91% vs 24%, P < .001), and to report overall health status as good, very good, or excellent (54% vs 30%, P < .001).
Endo ECHO and DCCC patients were overwhelmingly and equally likely to report having a usual source of diabetes care (Table 2). DCCC patients were more likely to report no visits to their source of care in the past year (<1% ECHO vs 5% DCCC, P < .001), while Endo ECHO patients were much more likely to have visited that source of care 4 or more times (72% ECHO vs 58% DCCC, P < .001). DCCC patients reported longer travel times on average to their usual source of care than Endo ECHO patients (26 minutes for DCCC vs 20 minutes for ECHO, P = .006).
In terms of guideline-recommended care, there were no statistically significant differences in reported A1C testing or feet checking by health care professionals. Endo ECHO patients were less likely to report eye exams in the past year (69% ECHO vs 87% DCCC, P < .001) and smoking status assessment (70% ECHO vs 82% DCCC, P = .004).
ECHO and DCCC patients did not differ significantly in frequency of consumption of high-fat foods and soda, 30-minute sessions of physical activity, or feet checks (see Table 3). ECHO patients were less likely than DCCC patients to space carbohydrates evenly 7 days in the previous week (34% ECHO vs 51% DCCC, P < .001) and test glucose levels (70% ECHO vs 83% DCCC, P < .001 = 0.002) and more likely to have smoked a cigarette (17% ECHO vs 6% DCCC, P < .001).
A series of regression analyses controlling for preferred language, educational status, insurance status, type of diabetes, and health literacy (results not shown) assessed whether clinic differences in care (Table 2) and selfmanagement (Table 3) were attributable to baseline differences in the patient populations served by the clinics (Table 1). None of the statistically significant clinic differences were rendered nonsignificant when controlling for these patient characteristics, suggesting that the clinic differences were independent of the differences in patient populations.
A telementoring model like Project ECHO holds promise for delivering high-quality care to underserved communities. The first question in this study was whether UNM’s Endo ECHO reached underserved, vulnerable patients with complex diabetes relative to its specialty clinic located at the UNM campus in Albuquerque. The results indicate that Endo ECHO served a less educated, more Hispanic and Spanish-speaking population with lower rates of private insurance than the DCCC clinic. Furthermore, ECHO patients were less health literate, reported poorer health status, and were dramatically less likely to have ever visited an endocrinologist. In terms of serving a population more disadvantaged than that able to access the only specialty clinic in the state, Endo ECHO was very successful.
Project ECHO does not seek only to serve a broader array of patients than specialty clinics; it aims to provide equivalent care to those patients. Accordingly, the second research question concerned the quality of diabetesrelated care provided to patients and patient outcomes related to self-management of diabetes. Reducing barriers to accessing care is an explicit goal of Project ECHO, and reported access to care was in fact better for ECHO patients than DCCC patients, with ECHO patients reporting more visits to their usual source of diabetes care and less travel time to get to care. The DCCC clinic is the referral center for the state, and it is not surprising that average travel times for its patients are longer than for the community-based ECHO clinics and that patients face barriers in accessing DCCC more frequently; many patients travel for hours to receive care at DCCC.
Quality-of-care and self-management outcomes indicated important points of comparability but also notable gaps in care at the ECHO clinics. In considering differences in self-management behaviors in particular, it is notable that the ECHO population is more disadvantaged than the DCCC population, including lower educational levels and health literacy, which might have a negative effect on self-management behaviors. However, controlling for these differences in patient characteristics did not diminish the clinic differences in the analyses, suggesting room for improvement exists in the Endo ECHO clinics in the areas of assessing smoking status, referring patients for eye exams, and promoting optimal self-management in smoking, spacing carbohydrates, and daily blood glucose testing.
This study relied on self-report surveys, which may be seen as a limitation. The survey used valid, accepted questions where available, and there is no reason to believe that any self-report biases differed by clinic type. Generalizability of the findings may be limited to populations similar to those included in this study.
Given the increasing shortage of endocrinologists and limited access to specialty care,12 alternative models of care for patients with complex diabetes are needed. Endo ECHO appears to be a suitable alternative to specialty care for patients in medically underserved communities where access to specialty clinic is restricted. More broadly, this study provides support for the value of the Project ECHO telementoring model in addressing barriers to highquality care for underserved communities.
This work was funded by a grant from The Leona M. and Harry B. Helmsley Charitable Trust.
Carolyn A. Berry https://orcid.org/0000-0003-3671-3080
Supplemental material for this article is available online.
