Safety net utilization among a cohort of the uninsured in Nashville, TN: An examination of a program designed to increase healthcare access

Purpose: To address the needs of the uninsured in Nashville, TN, the public health department developed and implemented a program, Bridges to Care (BTC) that linked low-income, uninsured individuals to primary care safety net clinics and provided access to low-cost prescriptions. Based on eligibility (i.e. <=200% of poverty, county resident status) individuals were assigned a medical home that matched their needs, e.g., language, location, health conditions. The goals of the program were to increase access to primary care and decrease emergency department (ED) utilization for non-emergent conditions. The purpose of this study was to assess ED and clinic utilization prior to and following enrollment into the BTC program.

Method: Health services utilization was assessed from 2002-2007 among a cohort of 5,246 individuals who enrolled into BTC program during the 2004 calendar year. Encounter information, including visit dates and primary diagnosis (ICD-9-cm codes), were obtained from twenty safety net clinics and nine hospitals (ED and inpatient) during the study period. Measures included number of persons, frequency of visits to the clinics, EDs and hospital stays, rates of non-emergent ED visits, ED rates of chronic disease visits, and relative rates of ED use.

Results: Seventy-five percent of the cohort had at least one visit to the safety net clinics prior to enrollment; 40% continued to receive care following enrollment; however 35% did not. Another 9% were new patients to the safety net. Clinic use increased following enrollment whereas the number of persons using the safety net declined; demonstrating an increase in utilization rate per person. During the study period there were no significant changes in ED use for non-emergent conditions (range=34%-35%) or for chronic diseases (range=3.3%-5.6%). The relative rate of ED use decreased; defined as percentage of total healthcare usage.

Conclusions: Although 40% of the cohort used the safety net prior to BTC enrollment, access to low-cost pharmaceuticals may have provided incentive for enrollment. The increase in utilization rates suggest that a segment of the cohort were in need of and subsequently received frequent health care. The relative decrease in ED use is explained by the increase in clinic visits; the absolute ED use did not change. Additional interventions are needed to influence ED use for non-emergent conditions and chronic disease management. It appears that a segment of the uninsured meet their healthcare needs by continuing to use the ED combined with prescription programs through BTC or a local pharmacy.