Purpose: Nearly half of all emergency department (ED) visits in the US are for nonurgent conditions. This study was the first to examine to what extent access to and use of patient-centered medical home (PCMH) practices can achieve cost savings in nonurgent ED care among the uninsured population nationwide.

Methodology: This retrospective cohort study used the 2010-2011 Medical Expenditure Panel Survey data. Difference-in-difference methods, multivariate logit model, marginal effect, and survey procedures were employed.

Findings: The cost savings amounted up to $586 million for the full PCMH group than the “no regular provider” group (95% CI: $225 million, $947 million) among the uninsured population at the national level. The mean nonurgent ED costs for the full PCMH group and the “no regular provider” group were $455.3 (95% CI: $55.9, $854.8) and $760.1 (95% CI: $527.6, $992.7) respectively among the uninsured population nationwide, with an average of $304.8 lower costs for each patient each year in the full PCMH group.

Conclusions: PCMH models demonstrate higher odds of reduced nonurgent ED care costs among the uninsured individuals, compared to insured enrollees. The room for improvement in nonurgent ED care cost reduction had large magnitudes.

Originality: For the PCMH models to effectively reduce nonurgent ED care costs, priority should be given to target the uninsured population.

Keywords: PCMH, nonurgent emergency department care, cost savings, uninsured, nationwide

Research indicates that near half of all emergency department (ED) visits in the US were for nonurgent medical care or potentially preventable.^1–3 It is estimated that nonurgent ED care can be $600 to $1050 more expensive than care received in physicians’ office.² This nonurgent ED use and costs problem may be more pronounced among the uninsured subpopulation. Some primary care doctors and specialists refuse to see uninsured patients due to concerns about underpayment, whereas EDs are open always, and cannot turn away patients regardless of their ability to pay, thus, uninsured individuals may use primary care less often and EDs more frequently.

Recently, the patient-centered medical home (PCMH) has risen as an innovative healthcare delivery model. It has features such as coordination and integration of care, a whole person orientation, and enhanced access to primary care.^4,5 Thus, the PCMH model holds promise in reducing nonurgent ED care costs for uninsured people by improving their access to and use of primary care.

To date, no studies have examined to what extent PCMH models can achieve cost reductions or cost savings as opposed to non-PCMH models among the uninsured populations. This current study is the first to examine the empirical evidence on this topic. The cost savings estimation depends on whether access to and use of PCMH practices was associated with reduced nonurgent ED care costs. Thus, this study examined whether there was such an association at all before the cost saving magnitude estimation, specifically, whether PCMHs’ general effect on potentially reduced nonurgent ED care costs was more pronounced for the uninsured population than the insured populations. This study built on existing work that examined PCMHs’ effects on reduced nonurgent ED utilization⁶ mainly as this current study had the similar algorithm in classifying ED care into urgent and nonurgent costs. However, this current study is distinct in that examines and analyzes nonurgent ED care costs as the outcome, rather than care utilization. This study will inform health policies on how PCMH models can contribute to cost reductions for nonurgent ED care, especially among the uninsured populations.

Data and study population

This study analyzed the Medical Expenditure Panel Survey (MEPS) 2010-2011 panel with a retrospective cohort study design. Individuals were included in this analysis if they were 18 years or older, had any ED visit in 2011, and had data from all five survey rounds. Detailed MEPS data description and features relating to this current study can be found in the previous work.⁶

Outcome variables

The outcome measure was the cost group status that was urgent versus nonurgent ED care costs in 2011. The cost group status, instead of the cost amount itself, was of interest. Detailed ED care use and cost group status and classification can be found in the previous work.⁶ The cost group of urgent ED care was the reference group.

Independent variables and control variables

The study by Beal et.al.⁷ used the MEPS data to measure the PCMH,⁷ which was further refined and categorized into four groups by the previous work.⁶ This current study adopted the refined measure and PCMH group classification. The PCMH group status included the full PCMH group, partial PCMH group, unknown PCMH group, and no regular source of care group, which was measured in the baseline year 2010.⁶ The control variables were the same as those in the previous work to ensure consistent analysis.⁶

Statistical analysis

There is a potential selection issue between nonurgent ED care costs and access to and use of PCMHs, this study attempted to attenuate this problem by including controls for the education level and the health status, as well as using the lagged time effect in our regression model. Detailed analysis description can be found in the previous work.⁶ Logit models were used to analyze urgent and nonurgent ED care costs. The analysis employs a difference-in-difference (DID) approach, comparing the difference in odds of nonurgent ED care cost between the full PCMH group and the “no regular provider” group for the uninsured individuals versus the insured individuals. The survey procedures in STATA version 13 were used to account for clustering in the sampling design and obtain correct standard errors. Detailed analysis description can be found in the previous work.⁶ The unit of analysis was each individual. Patients who had at least one nonurgent ED visit among multiple ED visits in 2011 were classified as having used nonurgent ED care. For the cost reduction analysis, this study focused on examining only the independent variable of interest: PCMH intervention groups and their interactions with insurance status. If any of them was significant, this study would calculate the nonurgent ED care cost magnitude that could be reduced between PCMH groups among the uninsured populations nationwide.

The demographic and socioeconomic characteristics of the sample and multivariate analysis results are presented in the previous work.⁶ In summary, 1,287 adults had any ED visit in 2011, which represented weighted 29,463,684 people in the U.S. population. Reductions in odds of nonurgent ED care costs between the full PCMH group and the “no regular provider” group was significantly larger for the uninsured group than publicly and privately insured groups (β = -1.70, p=0.009, and β= -1.04, p=0.040 respectively). Since the full PCMH group was significantly associated with lower odds of nonurgent ED care costs than the “no regular provider” group among the uninsured population, their corresponding nonurgent ED care cost reduction magnitude was further examined. The cost magnitude results are presented in Table 1–3. These cost savings amounted up to $586 million for the full PCMH group with 1.0 million patients than the “no regular provider” group with 2.0 million patients (95% CI: $225 million, $947 million) among the uninsured population at the national level. The mean costs for the full PCMH group and the “no regular provider” group were $455.3 (95% CI: $55.9, $854.8) and $760.1 (95% CI: $527.6, $992.7) respectively among the uninsured population, with average cost savings of $304.8 for each patient each year nationwide.

This study found that among the uninsured population at the national level, $586 million spent on nonurgent ED care can be reduced each year between the full PCMH group and the “no regular provider” group, which is a substantial magnitude. The cost savings were $304.8 for each patient each year nationwide. These findings have highlighted the promise of PCMH models to perform better than ordinary primary care practices to reduce nonurgent ED care costs. Literature reports that ordinary ambulatory or primary care practices may not be effective in reducing nonurgent ED care use or costs. One study found that two-thirds of patients who visited the ED for nonurgent reasons had a regular source of care in an ambulatory care setting.⁸ In addition, higher levels of physician visits in other ambulatory care settings were reported for patients with nonurgent ED use over a one-year period. Another study found that one-third of ED visits occurred when patients have access to primary care physicians during their normal business hours.⁹ At the same time, ED physicians believed even if patient conditions can be treated by a primary care physician patients may still use ED services due to ED convenience.¹⁰ Compared with ordinary ambulatory or primary care practices, PCMH models may be more effective in reducing nonurgent ED care use or costs.

Findings from this study have important policy implications. Continuously rising health care costs have imposed great challenges to policy makers, health plans, providers, and communities. It is reported that American healthcare costs neared $3.3 trillion in 2016, an average of $10,348 per person (11). Based on the nationwide healthcare cost challenges, identifying strategies for cost containment have thus become imperative. This study provides empirical evidence to inform policies of PCMH models’ effect in reducing nonurgent ED care costs. Specifically, the uninsured population would be the target population for PCMH models to achieve maximum nonurgent ED care cost savings.

Having acknowledged the value of PCMH models, large health plans, such as the Centers for Medicare & Medicaid Services (CMS), the largest single payer, has offered a 5% annual bonus until 2024 with the alternate payment model of the Medicare Access & CHIP Reauthorization Act for physicians who have a medical home.¹²

This study has some limitations. First, certain PCMH criteria involve provider-side measures at the system level that go beyond the scope and capacity of survey information from individual participants.¹³ However, the MEPS data is still a good source to capture or summarize common components within various PCMH models to estimate PCMH’s effect on nonurgent ED care cost savings nationwide, especially from patients’ perspectives. Another main limitation is the length of time of PCMH implementation, which cannot be reflected by the PCMH measure in the one-year baseline period. Detailed discussion can be found in the previous work.⁶

In conclusion, the empirical evidence suggests that at a national level, the PCMH model could save $586 million in nonurgent ED care costs within the uninsured population, amongst which are proxies for incorporating patient perspectives. The room for improvement in nonurgent ED care cost reduction had large magnitudes. This study informs policies on promising and pertinent strategies of PCMH models in reducing nonurgent ED care costs, especially, priority should be given to the target group of the uninsured population.

None.

The author declares there is no conflict of interest.

1. Niska R, Bhuiya F, Xu J. National Hospital Ambulatory Medical Care Survey: 2007 emergency department summary. Natl Health Stat Report. 2010;(26):1–31.
2. Excellus Blue Cross Blue Shield. Potentially avoidable emergency room visits in New York State.
3. Division of Health Care Finance and Policy. Preventable/Avoidable Emergency Department Use in Massachusetts. 2010.
4. Stange KC, Nutting PA, Miller WL, et al. Defining and measuring the patient-centered medical home. J Gen Intern Med. 2010;25(6):601–612.
5. Moreno L, Peikes D, Krilla A. Necessary but Not Sufficient: The HITECH Act and Health Information Technology’s Potential to Build Medical Homes. (Prepared by Mathematica Policy Research under Contract No. HHSA290200900019I TO2.) AHRQ Publication No.10-0080-EF. Rockville, MD: Agency for Healthcare Research and Quality. 2010.
6. Xin H, Kilgore ML, Sen B. “Is Access to and Use of Patient Perceived Patient-Centered Medical Homes Associated with Reduced Nonurgent Emergency Department Use?” Am J Med Qualy. 2017;32(3):246–253.
7. Beal A, Hernandez S, Doty M. Latino access to the patient-centered medical home. J Gen Intern Med. 2009;3(Supple 3):514–520.
8. Cunningham P. The Use of Hospital Emergency Departments for Nonurgent Health Problems: A National Perspective. Med Care Res Rev. 1995;52(4):453–474 .
9. Compact Action Brief: A Roadmap for Increasing Value in Health Care. Reducing Emergency Department Overuse: A $38 Billion Opportunity. 2010.
10. California HealthCare Foundation, Overuse of Emergency Departments among Insured Californians, Oakland, Calif. 2006. p. 1–10.
11. Hartman M. National Health Care Spending In 2016: Spending And Enrollment Growth Slow After Initial Coverage Expansions. Health Affairs. 2017.
12. The Medicare access & CHIP reauthorization act of 2015: path to value. 2015.
13. Agency for Healthcare Research and Quality. 2011. Patient Centered Medical Home Resource Center. 2011.