Sage Journals: Discover world-class research

Abstract

Objective: Federally qualified health centers (FQHCs) in Florida see large numbers of vulnerable patients with diabetes. Patient-centered medical home (PCMH) models can lead to improvements in health for patients with chronic conditions and cost savings for providers. Therefore, FQHCs are increasingly moving to PCMH models of care. The study objective was to examine the effects of initial transformation to a level 3 National Committee for Quality Assurance (NCQA) certified PCMH in 2011, on clinical diabetes outcomes among 27 clinic sites from a network of FQHCs in Florida. Methods: We used de-identified, longitudinal electronic health record (EHR) data from 2010-2012 and multivariate logistic regression to analyze the effects of initial transformation on the odds of having well-controlled HbA1c, body mass index (BMI), and blood pressure (BP) among vulnerable patients with diabetes. Models controlled for clustering by year, patient, and organizational characteristics. Results: Overall, transformation to a PCMH was associated with 19% greater odds of having well-controlled HbA1c values with no statistically significant impact on BMI or BP. Subanalyses showed transformation had less of an effect on BP for African American patients and HbA1c control for Medicare enrollees but a greater effect on weight control for patients older than 35 years. Conclusion: Transformation to a PCMH in FQHCs appears to improve the health of vulnerable patients with diabetes, with less improvement for subsets of patients. Future research should seek to understand the heterogeneous effects of patient-centered transformation on various subgroups.

Keywords

community health centers patient-centeredness access to care health outcomes disease management impact evaluation primary care

Introduction

In 2015, 13% of adult patients seen in federally qualified health centers (FQHCs) nationally had a diagnosis of diabetes, representing 11% of people with diabetes in the United States.¹ Individuals with diabetes have higher medical costs and are at greater risk for co-morbidities such as cardiovascular disease, stroke, and kidney disease.^2-4 Significant disparities in accessibility and receipt of quality diabetes care, are prevalent among those with low incomes.² FQHCs are often the main source of primary care in underserved areas where a majority of the patients may be uninsured and have multiple chronic conditions.⁵

FQHCs are increasingly moving toward patient-centered medical home (PCMH) models of care delivery. PCMHs focus on the whole patient and aim to provide accessible, comprehensive, and quality care with high levels of coordination among providers to improve the health of patients and achieve cost savings.^6,7 Providers may implement a PCMH independently or achieve recognition from a national organization such as the National Center for Quality Assurance (NCQA), which awards PCMH recognition based on a particular set of standards and performance metrics.⁴ Studies examining PCMH implementation on quality of care and effectiveness, including diabetes measures, reported mixed findings.Moreover, most of the studies were conducted in large integrated health systems, community-based clinics, or focused on specific payers.^2,3,7-26 While the literature in community-based practices and integrated health systems is substantial, little is known about the effects of FQHCs transitioning to a NCQA certified PCMH on clinical outcomes for patients with diabetes.¹

Gunter et al²⁴ examined PCMH characteristics and diabetes care quality using the Safety-Net Medical Home Scale and reported inconsistent and mixed findings for 800 patients with diabetes. Shi et al²⁵ used the Uniform Data System and surveys of Health Resources and Services Administration community health centers (CHCs) to compare clinical performance between PCMH and non-PCMH recognized CHCs. The study concluded that PCMH accredited centers performed better than nonaccredited CHCs for diabetes hemoglobin A1c (HbA1c) control and adult weight screening.²⁵ Finally, an examination of 800 FQHCs found that postimplementation, 1 of 6 PCMH domains (access/communication), was associated with improved outcomes for patients with diabetes.²⁶

Understanding how PCMH transformation influences care for vulnerable populations in FQHCs will provide vital information to improve care delivery for those with chronic illnesses and inform CHCs about successful models of implementation. Therefore, the objective of this study was to investigate the impact of the initial transformation to a level 3 (L3) NCQA certified PCMH on clinical outcomes for patients with diabetes receiving care in a network of FQHCs in Florida. We expect diabetes management to be improved after PCMH transformation given the principles of medical homes, which include providing comprehensive and coordinated care, a commitment to quality, and greater access to services.^6-8 This study adds to the literature by examining how implementation of the PCMH can affect diabetes clinical outcomes for subpopulations of vulnerable patients who visit FQHCs.

Methods

Data

We used de-identified, longitudinal electronic health record (EHR) data and a pre-post design to examine the impact of initial transformation to NCQA L3 PCMH status in 2011, on clinical outcomes for patients with diabetes receiving care in a network of FQHCs. Health Choice Network (HCN) is one of the largest health center controlled networks in the United States supported by HRSA with member centers that include 30 FQHCs and 2 non-FQHCs. Of those, 5 centers representing 27 clinics in Florida, sought and were awarded L3 PCMH recognition from NCQA in 2011.

To assess whether transformation to a PCMH improved outcomes for vulnerable patients with diabetes served in FQHCs, we examined 2010, 2011, and 2012, EHR data obtained from the 27 clinic sites. Patients were excluded from the sample if they were younger than 18 years, were pregnant, or had given birth in the previous 12 months. Additionally, observations that had implausible values for the outcome measures were removed from the analytic data set. The final analytic data set had 14 136 observations. The unit of analysis was a person-year, meaning each person could contribute up to 3 observations to the analysis.

Measures

The main predictor was clinic transformation to a L3 NCQA certified PCMH in 2011, constructed as a bivariate variable (1 = transformation in 2011). Consistent with the American Diabetes Association guidelines, 3 outcome measures were examined; the odds of having well-controlled glucose defined as HbA1c <7.0, the odds of having well-controlled blood pressure (BP) at 2 thresholds, <140/90 mm Hg or 130/80 mm Hg, and the odds of being at a normal weight, defined as body mass index (BMI) between 18.5 and 24.9 kg/m².⁴ For patients with multiple lab values or measurements during the year, the last recorded value in the patient’s record of the calendar year was used for each outcome variable.

Control variables included indicators for health center, age, gender, race, ethnicity, clinic size, baseline BMI, payer source, and primary language spoken. Age was categorized into 3 groups; 18 to 34 (reference), 35-64, and >65 years. Race was categorized into Caucasian (reference), African American, Asian, and other or unknown race. Ethnicity and primary language spoken were bivariate variables coded as Hispanic or non-Hispanic and English or non-English, respectively. Payer source was categorized into 4 groups, uninsured (reference), Medicaid, Medicare, and private insurance.

Analyses

Generalized estimating equation (GEE) models, also known as population average logistic regression models were used to estimate the association between an FQHC transforming to a L3 NCQA certified PCMH and diabetes clinical outcomes. The GEE models controlled for clustering by year, by patient, and by health center. All analyses were conducted using STATA 13.0.

Based on initial results from the primary analyses, we decided to test whether African American race, age group, or payer source moderated the effects of PCMH transformation on the outcome measures. We conducted 3 sub-analyses using an interaction term with PCMH by African American race, age group, and payer source. For the payer source interaction, uninsured, Medicaid, Medicare, and private insurance were used to examine moderating effects. The uninsured category was the most common source of payment.

Results

Population Characteristics

Table 1 displays the characteristics of the final analytic data set. The mean age was 59 years and approximately 60% were female. Half of the patients were Caucasian, 42% were African American, 6% were other or unknown race, and 2% were Asian. Forty-six percent of the patients were of Hispanic ethnicity and 58% spoke English as their primary language. In the pre-period, before controlling for other factors, 41% of the patients had HbA1c levels less than 7.0, 62% had BP values <140/90 mm Hg, and 14% had a BMI between 18.5 and 24.9 kg/m².

Table 1.

Sample Characteristics (N = 14 136).

Variable	Mean
Age, years	59.0
Gender, %
Female	59.8
Male	40.2
Race, %
Caucasian	49.7
African American	42.4
Asian	1.5
Other/Unknown race	6.4
Ethnicity, %
Hispanic	46.3
Non-Hispanic	53.7
Primary language spoken, %
English	58.1
Non-English	41.9
Clinic characteristics
Clinic size, n	485.7
Center 1, %	12.9
Center 2, %	8.0
Center 3, %	40.8
Center 4, %	22.0
Center 5, %	16.3
Payer type, %
Medicaid	16.9
Medicare	12.5
Private	3.3
Uninsured	67.4
Clinical measures
Body mass index, kg/m²	31.9
HbA1c, %	8.0
Systolic blood pressure, mm Hg	134
Diastolic blood pressure, mm Hg	80
HbA1c <7.0, %	40.7
Blood pressure less than 140/90 mm Hg, %	61.9
Normal weight: BMI between 18.5 and 24.9 kg/m², %	13.6

Effect of PCMH Transformation

Table 2 shows that transformation to a L3 NCQA PCMH in 2011, was associated with 19% greater odds of having well-controlled HbA1c but had no statistically significant impact on BP control or weight control. While the overall effect of transformation on blood pressure levels was not statistically significant in the primary multivariate analysis, African American patients had significantly worse HbA1c (odds ratio [OR] = 0.78, P = .001) and BP values (OR = 0.56, P < .001) compared with Caucasian patients after PCMH transformation. Moreover, initial transformation appeared to effect patients within certain age groups and with certain types of payers differently. Sensitivity analyses were conducted to test whether African American race, age, and payer source moderated the impact of PCMH transformation.

Table 2.

Multivariate Results on the Effect of FQHC PCMH Transformation on Diabetes Outcomes.

Measure	Odds Ratio	P (95% CI)
HbA1c < 7.0	1.19	.004 (1.05-1.34)
Blood pressure <140/90 mm Hg	1.05	.422 (0.93-1.18)
Blood pressure <130/80 mm Hg	1.06	.338 (0.94-1.20)
Normal weight: BMI between 18.5 and 24.9 kg/m²	1.06	.308 (0.94-1.20)

Abbreviations: FQHC, federally qualified health center; PCMH, patient-centered medical home; HbA1c, hemoglobin A1c; BMI, body mass index.

Interaction Effects of Race, Age, and Payer Type

Table 3 shows that medical home transformation had less of an effect for African American patients on BP control at 140/90 mm hg (OR = 0.77, P = .015) compared with all other patients. Race did not have a moderating effect on HbA1c values, BP control at 130/80 mm Hg, or BMI within the normal range. Also displayed, patients older than 35 years were more likely to achieve normal weight after PCMH transformation (OR = 2.67, P = .010) compared with younger patients. Finally, PCMH transformation had significantly less of an effect on controlling blood glucose for Medicare enrollees (OR = 0.63, P = .005) compared to uninsured patients.

Table 3.

Interaction Effects of Race, Age, and Payer Source on the Impact of PCMH Transformation on Diabetes Outcomes.

Measure	Odds Ratio	P (95% CI)
HbA1c <7.0
AA * PCMH	0.84	.114 (0.68-1.04)
Age 35-64 * PCMH	1.75	.420 (0.45-6.86)
Age 65 * PCMH	1.72	.437 (0.44-6.75)
Medicaid * PCMH	0.94	.647 (0.72-1.23)
Medicare * PCMH	0.63	.005 (0.46-0.87)
Private * PCMH	0.81	.576 (0.39-1.68)
Blood pressure <140/90 mm Hg
AA * PCMH	0.77	.015 (0.62-0.95)
Age 35-64 * PCMH	0.75	.605 (0.25-2.21)
Age 65 * PCMH	0.77	.638 (0.26-2.28)
Medicaid * PCMH	0.91	.502 (0.69-1.20)
Medicare * PCMH	1.19	.337 (0.84-1.68)
Private * PCMH	1.08	.846 (0.50-2.31)
Blood pressure <130/80 mm Hg
AA * PCMH	0.81	.074 (0.64-1.02)
Age 35-64 * PCMH	1.35	.528 (0.53-3.45)
Age 65 * PCMH	1.41	.481 (0.55-3.62)
Medicaid * PCMH	0.88	.400 (0.66-1.18)
Medicare * PCMH	1.24	.274 (0.84-1.83)
Private * PCMH	1.51	.357 (0.63-3.66)
Normal weight: BMI between 18.5 and 24.9 kg/m²
AA * PCMH	0.97	.774 (0.78-1.21)
Age 35-64 * PCMH	2.67	.010 (1.27-5.62)
Age 65 * PCMH	2.35	.025 (1.11-4.97)
Medicaid * PCMH	0.98	.913 (0.74-1.31)
Medicare * PCMH	1.02	.897 (0.71-1.48)
Private * PCMH	0.56	.132 (0.26-1.19)

Abbreviations: PCMH, patient-centered medical home; AA, African American; BMI, body mass index.

Discussion

The study objective was to examine the association between initial transformation to a L3 NCQA PCMH and diabetes outcomes for patients in FQHCs. We expected PCMH transformation to improve clinical outcomes based on the defining elements of medical homes, including patient-centeredness, providing coordinated care, and a focus on quality improvement, among others.^5-8 PCMH implementation in FQHCs appears to have a mixed or varying effect on the health of vulnerable patients with diabetes. Overall, there were significantly greater odds of a patient with diabetes having well-controlled HbA1c after PCMH transformation, likely from increased care utilization and chronic disease management.^14,26 There was also evidence that PCMH transformation in FQHCs has varying effects among different patient subgroups. For instance, after transformation patients with diabetes who were older than 35 years were more likely to achieve normal weight. However, there was less of an impact on blood pressure control for African American patients and less of an impact on glucose levels for Medicare patients. It is possible the results for achieving normal weight and controlling blood pressure in this study were because of the fact that producing changes in these outcomes can take more time compared to changes in glucose levels. The mixed findings reported in this study are consistent with the previous literature.^3,7-26

Given the differential effects of PCMH transformation among certain groups of patients, it is essential to further examine the impact of care delivery models on patient outcomes, particularly for patients with chronic illnesses. If differences in the impact of PCMH transformation among subgroups are affirmed in future studies, then FQHCs may want to consider how to tailor PCMH components to better meet the needs of particular subgroups of patients.

FQHCs are an essential part of our health care delivery system and will continue to have an increasing role in the provision of primary care services for those with limited resources who are often sicker, uninsured, and who may have otherwise sought care in hospital emergency departments or not at all.^5,27 For instance, a recent national study observed that FQHCs had 5% more visits for diabetes than other private community primary care providers.⁵

Thus, FQHCs can play an important role in improving the health of vulnerable populations with diabetes and help to reduce disparities in health outcomes.^5,27,28 As delivery systems continue to move toward care that is patient-centered, future research is needed to investigate whether specific aspects of PCMH transformations can have differential effects on health outcomes.

Our study has several strengths, including analyzing clinical outcomes extracted from the EHR data stored in HCN’s data warehouse. Moreover, all adults older than 18 years who met the inclusion criteria, as well as all payer sources utilized by this sample of patients were included in the sample, as opposed to limiting the analyses to a single payer source or relying solely on process measures.

This study has a few limitations. With any observational study, there is the possibility of unmeasured confounding. Unmeasured time variant factors could potentially affect the results given the pre-post design without a control group. However, all analyses controlled for patient and clinic characteristics and included center fixed effects to minimize potential confounding. Because, the participating FQHCs were all from HCN of Florida, generalizability may be limited. However, Florida’s diverse population strengthens the generalizability of the study.

Despite these limitations, it appears that PCMH transformation might be associated with improved outcomes for vulnerable patients living with diabetes. Calman et al²⁹ assessed utilization among patients with diabetes in FQHCs that transitioned to an NCQA L3 PCMH and observed increased use of support services but lower utilization of primary care services among all patients with diabetes. The Centers for Medicare and Medicaid Services FQHC Advanced Primary Care Practice Demonstration revealed that for Medicare beneficiaries with diabetes, transition to a PCMH resulted in more HbA1c tests, eye examinations, and nephropathy tests.³⁰ In the current analysis we looked at PCMH transformation as a singular event, when in actuality it is the composite of multiple practice changes. Thus, future research should examine the components of certification that may drive change. Our findings track with similar studies, help contribute to the evidence supporting transformation to a PCMH, and fills an important gap by examining patient-level clinical outcomes in a diverse safety-net population with multiple payer sources served in FQHCs that achieved NCQA L3 PCMH accreditation.

Footnotes

Authors’ Note

Preliminary results were presented at the Health Choice Network’s 22nd Annual Educational Conference, June 24-26, 2016 in Miami, Florida and at the AcademyHealth Annual Research Meeting, June 25-27, 2017 in New Orleans, Louisiana.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Heidi S. Kinsell, PhD, is a Research Faculty member in the Department of Behavioral Sciences and Social Medicine in the College of Medicine at Florida State University. Dr. Kinsell is a health services researcher whose research centers around access to care for vulnerable populations, health economics, and examining the effects of state health care policies on utilization of services and patient outcomes.

Allyson G. Hall, PhD, is a Professor in the Department of Health Services Administration at the UAB. Dr. Hall’s broad research interests emphasize improving healthcare access and systems for vulnerable populations. Current scholarly activity focuses on reducing fragmentation within the delivery system including assessments of patient-centered medical homes and transition to primary care programs for emergency department patients.

Jeffrey S. Harman, PhD, a health economist and health services researcher, is a Professor in the Department of Behavioral Sciences and Social Medicine at the Florida State University College of Medicine. His research explores utilization and expenditures of health services, with an emphasis on uncovering health disparities in the delivery of care and on the impact of health policies on services for vulnerable populations, such as Medicaid beneficiaries and individuals suffering from mental illness.

Sweta Tewary, PhD, is an Assistant Professor at Nova Southeastern University. Her background is in health informatics, business intelligence with professional experience in the areas of disease management, care burden and health promotion.

Andrew Brickman, PhD, is Director of Research at Health Choice Network. His training is in Health Psychology with a career in academic clinical research and a decade working in community health systems.

References

Health Resources and Services Administration. Health Center Data. US Department of Health and Human Services. Rockville, MD: Health Resources and Service Administration; 2015. http://bphc.hrsa.gov/uds/datacenter.aspx. Accessed October 30, 2017.

Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2017. Estimates of Diabetes and Its Burden in the United States. Atlanta, GA: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services; 2017. https://www.cdc.gov/diabetes/pdfs/data/statistics/national-diabetes-statistics-report.pdf. Accessed October 30, 2017.

Hoff

Weller

DePuccio

The patient-centered medical home: a review of recent research. Med Care Res Rev. 2012;69:619-644.

American Diabetes Association. Glycemic targets. Diabetes Care. 2017;40(suppl 1):S48-S63. doi:10.2337/dc17-S009.

Kaiser Commission on Medicaid and the Uninsured. Community health centers in an era of health reform: an overview and key challenges to health center growth. http://kaiserfamilyfoundation.files.wordpress.com/2013/03/8098-03.pdf. Published March 2013. Accessed October 30, 2017.

Patient-Centered Primary Care Collaborative. Defining the medical home. A patient-centered philosophy that drives primary care excellence. https://www.pcpcc.org/about/medical-home. Accessed October 30, 2017.

National Committee for Quality Assurance. NCQA Patient-centered medical home. Improving experiences for patients, providers and practice staff. https://www.ncqa.org/Portals/0/PCMH%20brochure-web.pdf. Accessed October 30, 2017.

Nielsen

Buelt

Patel

Nichols

. The patient-centered medical home’s impact on cost and quality. Annual review of evidence, 2014-2015. https://www.pcpcc.org/resource/patient-centered-medical-homes-impact-cost-and-quality-2014-2015. Published February 2016. Accessed October 30, 2017.

Neal

Chawla

Colombo

Snyder

Nigam

Medical homes: cost effects of utilization by chronically ill patients. Am J Manag Care. 2015;21:e51-e61.

10.

Jackson

Powers

Chatterjee

et al . Improving patient care. The patient-centered medical home: a systematic review. Ann Intern Med. 2013;158:169-178.

11.

Reid

Johnson

Hsu

et al . Spreading a medical home redesign: effects on emergency department use and hospital admissions. Ann Fam Med. 2013;11(suppl 1):S19-S26.

12.

Bojadzievski

Gabbay

RA.

Patient-centered medical home and diabetes. Diabetes Care. 2011;34:1047-1053.

13.

Pines

Keyes

van Hasselt

McCall

Emergency department and inpatient hospital use by Medicare beneficiaries in patient-centered medical homes. Ann Emerg Med. 2015;65:652-660. doi:10.1016/j.annemergmed.2015.01.002.

14.

Paustian

Alexander

El Reda

Wise

Green

Fetters

MD.

Partial and incremental PCMH practice transformation: implications for quality and costs. Health Serv Res. 2014;49:52-74.

15.

Werner

Duggan

Duey

Zhu

Stuart

EA.

The patient-centered medical home: an evaluation of a single private payer demonstration in New Jersey. Med Care. 2013;51:487-493.

16.

Flottemesch

Fontaine

Asche

Solberg

LI.

Relationship of clinic medical home scores to health care costs. J Amb Care Manag. 2011;34:78-89.

17.

Rosenthal

Alidina

Friedberg

et al . A difference-in-difference analysis of changes in quality, utilization and cost following the Colorado multi-payer patient-centered medical home pilot. J Gen Intern Med. 2016;31:289-296. doi:10.1007/s11606-015-3521-1.

18.

Rosenthal

Alidina

Friedberg

et al . Impact of the Cincinnati aligning forces for quality multi-payer patient centered medical home pilot on health care quality, utilization, and costs. Med Care Res Rev. 2016;73:532-545. doi:10.1177/1077558715618566.

19.

Rosenthal

Sinaiko

Eastman

Chapman

Partridge

Impact of the Rochester Medical Home initiative on primary care practices, quality, utilization, and costs. Med Care. 2015;53:967-973.

20.

Friedberg

Rosenthal

Werner

Volpp

Schneider

EC.

Effects of a medical home and shared savings intervention on quality and utilization of care. JAMA Intern Med. 2014;175;1362-1368. doi:10.1001/jamainternmed.2015.2047.

21.

Fifield

Forrest

Burleson

Martin-Peele

Gillespie

Quality and efficiency in small practices transitioning to patient centered medical homes: a randomized trial. J Gen Intern Med. 2013;28:778-786. doi:10.1007/s11606-013-2386-4.

22.

DeVries

Sridhar

Hummel

Breidbart

Barron

JJ.

Impact of medical homes on quality, healthcare utilization, and costs. Am J Manag Care. 2012;18:534-544.

23.

Solberg

Asche

Fontaine

Flottemesch

Anderson

LH.

Trends in quality during medical home transformation. Ann Fam Med. 2011;9:515-521.

24.

Gunter

Nocon

Gao

Casalino

Chin

MH.

Medical home characteristics and quality of diabetes care in safety net clinics. J Community Health. 2017;42:303-311. doi:10.1007/s10900-016-0256-9.

25.

Shi

Lee

Chung

Liang

Lock

Sripipatana

Patient-centered medical home recognition and clinical performance in US community health centers. Health Serv Res. 2017;52:984-1004.

26.

Shi

Lock

Lee

et al . Patient-centered medical home capability and clinical performance in HRSA-supported health centers. Med Care. 2015;53:389-395.

27.

Adashi

Geiger

Fine

MD.

Health care reform and primary care—the growing importance of the community health center. N Engl J Med. 2010;362:2047-2050.

28.

National Association of Community Health Centers. A sketch of community health centers. Chartbook December 2014. http://nachc.org/wp-content/uploads/2015/11/Chartbook_December_2014.pdf. Accessed October 30, 2017.

29.

Calman

Hauser

Weiss

et al . Becoming a patient-centered medical home: a 9-year transition for a network of federally qualified health centers. Ann Fam Med. 2013;11(suppl 1):S68-S73.

30.

Kahn

Timbie

Friedberg

et al . Evaluation of CMS’s Federally Qualified Health Center (FQHC) Advanced Primary Care Practice (APCP) Demonstration: Final Second Annual Report. Santa Monica, CA: RAND Corporation; 2015. http://www.rand.org/pubs/research_reports/RR886z1.html. Accessed October 30, 2017.

Impacts of Initial Transformation to a Patient-Centered Medical Home on Diabetes Outcomes in Federally Qualified Health Centers in Florida

Abstract

Keywords

Introduction

Methods

Data

Measures

Analyses

Results

Population Characteristics

Effect of PCMH Transformation

Interaction Effects of Race, Age, and Payer Type

Discussion

Footnotes

Authors’ Note

Declaration of Conflicting Interests

Funding

Author Biographies

References