Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Health Outcomes and Policy Research

Research Advisor

Lawrence M. Brown, PharmD, Ph.D.


Cyril F. Chang, Ph.D. Joy A. Clay, Ph.D. Doug Imig, Ph.D. Junling Wang, Ph.D., M.S.


nonurgent ED use, privately insured, regular source of care, nationally representative


Purpose. The purpose of this study was to examine the influence of patient‑perceived provider availability on nonurgent emergency department use (NUEDU). The study population consisted of a nationally‑representative population of non‑elderly adults who were continuously privately insured (CPI) for at least a year before an emergency department visit that was determined to be nonurgent, and who reported having a regular source of care (RSC).

Methods. Data were obtained from a nationally‑representative longitudinal survey, the Medical Expenditure Panel Survey (MEPS). The classification of ED visits as nonurgent was based on patient report using a method developed for this study. Patient­‑perceived provider availability was based on patient report of how difficult it was to contact their RSC by phone for medical advice during office hours (Regular Hours Contact), how difficult it was to reach their RSC after hours in case of an urgent problem (After Hours Contact), and whether their RSC had office hours at night or on weekends (Night or Weekend Hours). Multiple logistic regression analysis was used to determine whether the patient-reported provider availability factors predicted NUEDU.

Findings. Nonurgent ED visits made by continuously insured non-elderly adults with a RSC accounted for an estimated 2,309,399 ED visits in 2006. This number represents 9.31% of ED visits made by people of all ages, and 12.08% of all ED visits made by non‑elderly adults. The percentage ED visits categorized as nonurgent in the study population was 45.15%. Reported lack of night or weekend hours predicted an increased likelihood of a nonurgent ED visit; OR 1.371(1.368, 1.374). Increasing reported difficulty in contacting a RSC by telephone outside of regular office hours predicted decreasing likelihood of NUEDU. In contrast, increasing difficulty in contacting a RSC during office hours by telephone predicted increasing likelihood of NUEDU. Those who reported that it was "very difficult" to contact a RSC by phone during regular office hours were more than 4 times as likely to have a nonurgent ED visit as those who reported that it was "not at all difficult"; OR 4.136(4.122, 4.151). Difficulty in reaching a RSC during regular office hours was a consistent predictor of increased likelihood of NUEDU regardless of how the population of nonurgent ED users was segmented, almost without exception.

Conclusions. The factors representing availability of night or weekend hours, ability to make after‑hours contact by telephone with a RSC, and regular‑hours contact by telephone were found to be statistically significant predictors of NUEDU among non‑elderly adults with CPI and a RSC. These findings suggest that privately‑insured persons with a RSC rely on EDs for a significant portion of their acute but nonurgent health care needs.