Changes in Barriers to Primary Care and Emergency Department Utilization

The 2010 Patient Protection and Affordable Care Act¹ (ACA) aimed to expand health insurance coverage, improve access to medical care, and control health care costs.² An implicit goal of health insurance expansion is to improve access and utilization of primary care services and divert patients from what is believed to be higher-cost emergency care. Indeed, annual emergency department (ED) visit rates have increased from 90 million in 1996 to 124 million in 2008.³

Using data from the 2005 National Health Interview Survey (NHIS), Rust et al⁴ reported an association between barriers to timely primary care and ED utilization. In Massachusetts, where health care reform legislation similar to the ACA was enacted in 2006, ED visits have remained high despite achieving near universal coverage (>98% of working-age adults).⁵ Thus, expanded national health insurance coverage in the setting of a primary care provider shortage may add to the problem of limited access to primary care services. To further evaluate this hypothesis, we quantified changes in national barriers to timely primary care access from 1999 to 2009, and their association with ED utilization.

We analyzed data from the NHIS, a cross-sectional household interview survey that approximates noninstitutionalized US civilian population.⁶ From 1999 to 2009, NHIS collected household interview data for a total of 317 497 adults (age, ≥18 years). Barriers to timely primary care captured in the survey included the following: (1) “Couldn't get through on the telephone”; (2) “Couldn't get an appointment soon enough”; (3) “Once you got there, you have to wait too long to see the doctor”; (4) “The (clinic/doctor’s) office wasn't open when you could get there”; and (5) “Didn't have transportation.” These barriers were used to predict self-reported ED visits during the past 12 months.

We performed statistical analyses using Stata 10.1 (StataCorp, College Station, Texas). Survey commands were used to create nationally representative estimates. Multivariable analyses adjusted for demographic, socioeconomic status, health conditions, and access to care variables.

Overall, 9.7% of adults per year had at least1 barrier to timely primary care and 20.1% had at least 1 ED visit. Adults with a higher number of barriers were more likely to have at least 1 ED visit (18.8% for 0 barrier, 29.5% for 1 barrier, and 36.5% for ≥2 barriers). After adjusting for potential confounders, barriers to timely primary care were associated with increased ED utilization (compared with 0 barriers: adjusted odds ratio [AOR], 1.37 [95% confidence interval {CI}, 1.31-1.43] for 1 barrier and AOR, 1.68 [95% CI, 1.60-1.78] for ≥2 barriers).

Over the past decade, the prevalence of barriers to timely primary care among all adults, and specifically among those with at least 1 ED visit, has increased (Figure). From 1999 to 2009, the prevalence of having at least 1 barrier increased from 6.3% (95% CI, 6.0%-6.6%) to 12.5% (95% CI, 11.9%-13.1%). During that time, the strength of the association between number of barriers and ED utilization remained constant; for individual years, the AOR for at least 1 ED visit ranged from 1.29 to 1.44 (mean, 1.37) for 1 barrier and 1.54 to 1.91 (mean, 1.68) for 2 or more barriers. However, among adults with at least 1 ED visit, the prevalence of having at least 1 barrier increased from 12.0% (95% CI, 11.0%-13.0%) to 18.9% (95% CI, 17.6%-20.3%).

As previously reported using 2005 NHIS data,³ barriers to timely primary care were associated with increased ED utilization during 1999 to 2009. We extended these findings to demonstrate that these barriers have increased over the past decade, and were increasingly prevalent among those with ED visits. These results suggest that limited access to primary care services is an increasingly important contributor to rising ED volumes.

While policy makers attempt to reduce ED utilization through expanded health insurance coverage, the well-documented primary care provider shortage will likely accelerate the observed trend of increased barriers to timely primary care.⁷ On the basis of our study results, we beleive that the increasing prevalence of these barriers may result in even higher patient ED utilization.

This study has some potential limitations. By using data from an existing survey, we were limited to questions that were already in the survey and could not alter or add other questions. The reported associations and temporal trends may be confounded by unmeasured factors not included in the NHIS. In addition, the NHIS was based on self-reported data, so barriers and ED utilization could not be confirmed and was subject to recall bias.

In conclusion, the association between barriers to timely primary care and increased ED utilization has been consistent over the past decade, and the prevalence of these barriers has been rising. In the setting of limited primary care workforce resources, health insurance expansion and increased demand for services may contribute to even more barriers to timely primary care.⁸ Optimal health care delivery and attempts to limit ED utilization will likely require solutions beyond expanded health insurance coverage including improved access to primary care services through increasing the supply and availability of primary care providers.

Correspondence: Dr Ginde, Department of Emergency Medicine, University of Colorado School of Medicine, 12401 E 17th Ave, Mail Stop B-215, Aurora, CO 80045 (adit.ginde@ucdenver.edu).

Author Contributions:Study concept and design: Cheung and Ginde. Acquisition of data: Cheung and Ginde. Analysis and interpretation of data: Cheung, Wiler, and Ginde. Drafting of the manuscript: Cheung. Critical revision of the manuscript for important intellectual content: Wiler and Ginde. Obtained funding: Cheung and Ginde. Administrative, technical, and material support: Ginde. Study supervision: Ginde.

Funding/Support: This study was supported by grants from the Emergency Medicine Foundation, Society for Academic Emergency Medicine, and Adler Scholarship.

Previous Presentation: An abstract of this study was presented at the Society for Academic Emergency Medicine Conference; June 5, 2011; Boston, Massachusetts.