The effect of North Carolina free clinics on hospitalizations for ambulatory care sensitive conditions among the uninsured

Conditions that are treated in an acute care setting that could have been mitigated through access to appropriate primary care are known as Ambulatory Care Sensitive (ACS) conditions. Lack of access to primary care results in the absence of regular preventive care, monitoring of chronic illnesses, and early treatment of acute conditions [1, 2]. As untreated health conditions worsen, hospitalization may be required; examples include uncontrolled asthma triggering a lung infection or unmanaged diabetes resulting in a stroke. High rates of ACS conditions that result in admission to hospitals or emergency departments are therefore an indicator of poor access to primary care.

Despite modest improvements in the number of preventable hospitalizations for ACS conditions in recent years, total costs for potentially preventable hospitalizations are estimated to exceed $30 billion annually [3]. Uninsured adults, who are less likely to have regular source of care and more likely to have unmet medical needs [4], are hospitalized for ACS conditions more frequently than Medicaid recipients or individuals with commercial insurance [5]. Efforts to improve primary care access for populations with the highest rates of hospitalizations for ACS conditions, such as the uninsured, could aid in reducing aggregate health care costs in the U.S. [6].

Free clinics, which provide medical care for free or minimal cost, are one avenue to address the medical needs of the uninsured. While free clinics do not have the resources to meet all the medical needs of these populations, they often provide care for chronic illnesses through regular monitoring, dispensing medications, and providing lab tests. Their services may be more limited than those of other primary care organizations that accept insurance due to limited funding for free clinics and reliance on medical and administrative volunteers. Clients served by free clinics frequently come from demographic groups identified as having an increased likelihood of being admitted for ACS conditions, such as uninsured [5], individuals from low-income areas [6, 7], and minorities [8, 9].

This study examines whether free clinics in North Carolina (NC) reduce hospitalizations for ACS conditions for uninsured adults. More precisely, this work considers whether hospitalization of an uninsured adult in a community served by a free clinic has lower odds of being admitted for an ACS condition. We also consider whether this relationship is stronger for ACS hospitalizations resulting from chronic conditions as compared to ACS hospitalizations resulting from acute conditions.

Cecil P. Sheps Center for Health Services Research provided NC hospital discharge data for 2003 through 2007. From 2003 to 2007, eighteen free clinics opened in NC [25] (Additional file 1). The sample was restricted to NC adults who were designated self-pay (n = 270,325).

Table 4 also presents the demographic and community characteristics of the sample by whether the hospitalization was for an ACS condition or not. Hospitalizations for ACS conditions occurred more frequently in counties with a free clinic and without an FQHC. Uninsured men and women were hospitalized with an ACS condition at the same rate. However, uninsured middle-aged, older and Black Americans were more often hospitalized with an ACS condition as compared to their reference group (i.e., younger adults and White Americans, respectively). Although t-tests indicated the number of hospital beds, number of MDs, percent of the population minority, percent of the population living in poverty and the percent of the population without health insurance differed statistically for the two groups, actual differences were minimal in practical terms.

The lack of significance of the odds ratio for free clinics when examining hospitalization related to an acute ACS condition is consistent with the notion that free clinics are more focused on providing preventive care for chronic conditions than addressing time sensitive acute conditions. None of the community factors were statistically significant, likely due to the lack of variation in these measures within counties over time in the period examined.

Examining all ACS conditions, uninsured women had slightly higher odds of being hospitalized for an ACS condition as compared to uninsured men (OR women: 1.08). Increasing age was associated with steadily increasing odds of an ACS hospitalization for the uninsured (OR 30 to 39: 1.23; OR 40 to 49: 1.67; OR 50 to 64: 2.12; and OR 65 or older: 2.90). In addition, uninsured Black Americans had 1.84 times the odds of being admitted for an ACS condition as compared to uninsured White Americans, while Asian Americans had a 38% reduced odds of being admitted for an ACS condition versus White Americans.

For ACS conditions related to chronic illnesses, the magnitude of the odds ratios for uninsured middle-aged, older, and Black Americans increased relative to the odds ratios for all ACS conditions. Black Americans experience over twice the odds (OR: 2.15) of being hospitalized for ACS conditions related to chronic illnesses compared to White Americans. The odds ratios for the age categories for chronic illness related ACS conditions followed a similar pattern as those for overall ACS conditions, but the effect was greater for each age category. However, in limiting ACS conditions to only those related to chronic illnesses, women had 9.0% lower odds of being hospitalized for a chronic condition versus their male counterparts.

Uninsured individuals with limited access to primary care are at greater risk of being hospitalized for ACS conditions [5, 27], incurring potentially unnecessary costs for hospitals and health care systems. However, few large-scale, multi-year studies have focused on how free clinics affect hospitalizations for ACS conditions for uninsured adults. This study finds that proximity to a free clinic significantly reduces an uninsured individual’s odds of a hospitalization for an ACS condition.

This large-scale study is the first to examine the impact of free clinics serving a community on hospitalizations for uninsured individuals with ACS conditions during a period when multiple new clinics opened. The incorporation of the county-level and year fixed effects creates a pre/post study design, with the results driven by counties gaining free clinic services during the study period. The study design allows us to argue that our finding is causal in nature. During the time period examined (2003 to 2007), 18 new free clinics were opened in NC serving an additional 23 counties. Although free clinics only serve approximately 87,000 uninsured [28], equating to approximately 5.5% of the uninsured adults in NC (pre-ACA) [29], the model indicates they contribute to a statistically and practically significant reduction in the odds of a hospitalization for an ACS condition by an uninsured individual, an effect which would be increasingly magnified as the proportion of uninsured served increases.

We found that FQHCs had no significant effect on the odds of hospitalization for an ACS condition for the uninsured. This finding is consistent with other research on the uninsured [30, 31], implying that FQHCs may be more successful at providing primary care to Medicaid and/or Medicare recipients than to uninsured individuals. Administrative and/or economic requirements for care at FQHCs could impede access for uninsured individuals [32, 33]. However, the result in our study may be due to limited entry of FQHCs over the period examined. To the extent that FQHCs are not providing accessible or adequate primary care for the uninsured, other providers, such as free clinics, are necessary to minimize avoidable and costly use of hospitals for ACS conditions.

As expected, free clinics’ contribution to providing medical care for the uninsured appears to be most pronounced in aiding management of chronic diseases. Patients at free clinics may benefit from ongoing regular contact with a provider enabling them to diagnose chronic conditions, maintain prescriptions, adjust treatment as needed, and recognize symptoms of declining health. For acute ACS conditions, limited appointment availability and a lack of specialists and equipment may restrict free clinics’ ability to address time sensitive care needs.

While care for acute conditions is a necessary part of primary care treatment, the provision of adequate ongoing regular medical care for chronic illness is an important need for low income populations, who are at higher risk of having one or more chronic illnesses. Previous studies of individual free clinics have shown improved self-care management among the uninsured in the form of increased exercise time, improved blood pressure control, and reducing HgbA1c levels [19–21], supporting free clinics’ focus on addressing chronic illnesses amongst the uninsured. Further investigation of free clinics’ programming may uncover practices that could be adopted at other health care safety net organizations in treating chronic illnesses for the uninsured.

The current study highlights the interconnectedness of chronic illness among the uninsured and hospitalizations for ACS conditions. Our study found that uninsured Black Americans had higher odds of being hospitalized for ACS conditions, with the effect larger for chronic ACS conditions. This pattern is consistent with prior research indicating increased rates of hospitalization for chronic illness related ACS conditions for Black Americans versus rates of hospitalizations for all or only acute ACS conditions [9]. We also found that middle-aged and older uninsured individuals had higher odds of being hospitalized for ACS conditions, with the effect larger for chronic ACS conditions. Hospitalizations for ACS conditions for middle-aged adults create added costs from lost days of work, and higher out-of-pocket costs that can have long-term ramifications for the individual and their families [8]. These demographic groups (Black, middle-aged and older Americans) are at greater risk for having one or more chronic illnesses [34–36], and with limited access to health care as a result of lack of insurance, they are likely to have unattended conditions that require costly hospitalizations. Future research is warranted to understand whether disease management programs at free clinics are effective in improving outcomes for uninsured Black Americans, middle-aged, and older adults with one or more chronic illnesses.

Finally, the current study again confirms the ongoing inequities in health and health care for Black Americans. Uninsured Black Americans had an 84.0% increased odds of being hospitalized for an ACS condition as compared to White Americans. This finding is consistent with earlier studies finding of higher rates of ACS hospitalizations for Black Americans for all types of payers [8, 9, 37]. Organizations within the health care safety net need to develop partnerships with social organizations directed toward, operated by, and frequented by Black Americans to improve access as well as improve our understanding of barriers to care for Black Americans.

Although uninsured hospital stays for ambulatory care sensitive conditions are twice as common as ACS hospital stays for Medicaid or private insurance [5], few studies have examined whether free clinics aid in reducing the odds that an uninsured individual is hospitalized for an ACS condition by providing effective primary care. Despite the success in reducing the number of uninsured in the US with the implementation of the ACA, states that rejected the Medicaid expansion will continue to have higher rates of uninsured, and therefore potentially higher rates of ACS admissions. This study indicates that free clinics dedicated to providing care for uninsured adults in NC, despite serving less than 6% of the uninsured, contribute to statistically and practically significant lower proportion of hospitalizations for ACS conditions by the uninsured.

Free clinics in NC have been successful in addressing the needs of uninsured adults with chronic conditions. However, given the increased odds of being hospitalized for an ACS condition related to a chronic illness for middle-aged, older and Black Americans, further research is warranted on the effectiveness of free clinics in meeting the needs of these groups.

As NC and other southern states continue to opt out of the Medicaid expansion, and in the absence of universal health care, these states need to investigate how to make primary care accessible to the uninsured in order to improve their health and minimize costly hospital use for ACS conditions. To increase free clinics’ reach, state and local policy makers should encourage development of free clinics in high need areas, such as low income and minority communities. Meeting the health care needs of the uninsured could improve health outcomes for the uninsured while reducing healthcare costs for the community.