Characteristics of California Emergency Departments in Centers for Disease Control and Prevention designated HIV Priority Counties

Abstract

Background:

Refocused national HIV testing initiatives include a geographic focus.

Objectives:

Using a geographic focus, we sought to identify which emergency departments (EDs) might be the most efficient targets for future HIV testing efforts using California as an example.

Methods:

Retrospective analysis of California EDs, emergency physicians, and patients served, along with county-level estimates of HIV prevalence and proportion of the population living in poverty. Emphasis is placed on characterizing EDs affiliated with teaching hospitals and those located in Centers for Disease Control and Prevention HIV priority counties.

Results:

Of the 320 EDs studied, 178 were in priority counties, 29 were affiliated with teaching hospitals, and 24 had both characteristics. Of the 12,869,889 ED visits included, 61.8% occurred in priority counties, 14.7% in EDs affiliated with teaching hospitals, and 12.0% in EDs with both characteristics. The subset of EDs in priority counties with teaching hospital affiliations (compared to priority and non priority county ED groups without a teaching hospital affiliation) had higher overall median visit volumes and higher proportions of visits by at-risk and CDC-targeted populations (e.g., individuals who were homeless, those who identified as Black or African American race, and those who identified as Hispanic or Latino ethnicity, all p<0.01).

Conclusions:

EDs in priority counties affiliated with teaching hospitals are major sources of healthcare in California. These EDs more often serve populations disproportionately impacted by HIV. These departments are efficient targets to direct testing efforts. Increasing testing in these EDs could reduce the burden of undiagnosed HIV in California.

Introduction

The Centers for Disease Control and Prevention (CDC) recommend universal opt-out HIV screening in all healthcare settings for all people, ages 13 to 64, at least once in their lifetime; for those at higher risk, the CDC recommends more frequent testing (1). However, many Americans have never been tested or are not tested at frequencies reflective of their risk profiles (2, 3). Approximately 1 in 7 people with HIV are unaware of their infection, and the estimated time from infection to diagnosis in the US is more than 3 years (4, 5). Individuals with undiagnosed HIV serve as a reservoir for ongoing transmission, accounting for 38% of new infections in the US (4). In light of the testing shortfall, the US Department of Health and Human Services and the CDC have refocused efforts and launched Ending the HIV Epidemic: A Plan for America (6). A focus of this initiative is diagnosing all people with HIV as early as possible. To obtain “maximum impact” the initiative targets a subset of “priority” jurisdictions - states and counties hardest hit by HIV and areas that account for more than half of new diagnoses (6, 7).

Despite CDC recommendations and more than 150 million visits to US emergency departments (EDs) each year, only a minority of ED patients are tested for HIV (e.g., 0.22% in 2009 to 1.05% in 2018 of visits included a HIV test) and less than one-quarter of EDs are estimated to offer routine HIV testing (8–12). Given this setting, it is unclear how US EDs should respond to these refocused efforts. EDs serve communities and patients with varying rates of HIV and the already-limited resources of many are tailored to the unique needs of the communities they serve (6, 11). A universal redoubling of effort may be neither feasible nor the most efficient use of already limited resources.

The primary goal of this work was to explore characteristics of different types of EDs in a state with a high burden of HIV and demonstrate that there is a subset that might represent the most efficient, high-yield, targets toward which to direct future HIV testing efforts. To do this, we focus on California and stratify EDs by CDC priority county status (in a CDC-designated priority county or not) and teaching hospital affiliation (affiliated with a teaching hospital or not). We then describe the differing characteristics of these EDs, the patients they serve, and the surrounding communities (6, 7). Emphasis is placed on characterizing a subset of at-risk populations served in these EDs along with populations targeted by the CDC in HIV-related efforts - persons who are homeless and individuals who identify as Black or African American race and individuals who identify as Hispanic or Latino ethnicity (1, 4, 6, 7, 13). The seroprevalence of HIV among persons who are homeless is estimated to range between 2–20%; individuals who identify as Black or African race and individuals who identify as Hispanic or Latino ethnicity represent approximately 40% and 25%, respectively, of persons living with HIV (13, 14).

California accounts for 12% of the US population and is disproportionately impacted by HIV; eight California counties are CDC-designated as priority counties (Alameda, Los Angeles, Orange, Riverside, Sacramento, San Bernardino, San Diego, and San Francisco) (6, 7). Given this burden, in addition to CDC efforts, the California Department of Public Health (CDPH) has presented its first integrated strategic plan geared towards ending the HIV epidemic; this strategic plan includes a call for more routine HIV testing and linkage to care in California EDs (6, 15). Teaching hospitals, compared to non-teaching hospitals, have more resources and are thus more likely to have the institutional infrastructure (e.g., presence of social workers and infectious disease specialists) required to sustain ED-based HIV testing programs (16–18). Prior work also demonstrates that EDs with teaching hospital affiliations more frequently offer routine HIV testing and feel such testing is needed (11, 18). Identifying the subset of California EDs that might be the best, most efficient targets for HIV testing efforts could help guide policies on a state and, potentially, national level.

Methods

Study Design and Setting

To characterize California EDs, we used annually updated 2021 calendar year ED profile data from the California Department of Health Care Access and Information released in September 2022 on 322 EDs (all CDPH licensed EDs) in the state (19). Information included ED location and summary information on the demographics of patients served in these EDs; this data is sourced from two databases - an ED Treat-and-Release Database and an Inpatient Database (19). Individual ED location was categorized by medical service study area designators (i.e., urban, rural, frontier); these are an established classification approach based on US census information, incorporate socioeconomic and demographic data, and reflect ED-level information (e.g., frontier EDs being those in regions of sparsely populated areas geographically isolated from population centers) (19). Given our interest in EDs with general acute care hospital licenses, we excluded 2 facilities with psychiatric-specific licenses. The remaining 320 EDs were then classified by corresponding county location as priority or non-priority counties according to the CDC designation and then as affiliated or not affiliated with a teaching hospital according to ED profile data (19).

To describe the county-level density of emergency physicians (EPs) in California, we used publicly available Area Health Resource Files from the US Health Resources and Service Administration (20). These files provide county-level, population-adjusted (per 100,000 individuals) numbers of clinically active EPs. Physician information stems from the 2019 American Medical Association Physician Masterfile that is population adjusted using 2020 US Census population estimates (20). Inactive physicians (i.e., retired, semi-retired, working part-time, temporarily not in practice, or not active for other reasons) were not included. To describe county-level HIV prevalence rates, we used publicly available 2019 data from the CDC reflecting cases per 100,000 population among individuals aged 13 years and older from all races/ethnicities (21); HIV prevalence for Alpine and Modoc counties are suppressed in this data. We also used publicly available 2020 US Census Bureau Small Area Income and Poverty Estimates to identify median estimates of the proportion of persons in each California county, of all ages, living in poverty (22).

Analysis

All data were accessed on September 12, 2022 and were the most recent data available. Values were stratified and summarized with descriptive statistics (e.g., medians with interquartile ranges [IQRs] and proportions). Medians were used to measure central tendency instead of means given the skewness of the data. Analyses were conducted in R version 4.0.3 (https://www.r-project.org). Figures were created in R and ArcGIS Desktop (version 10.8.0; ESRI, Redlands, CA). Comparisons were conducted using statistical tests (e.g., χ², Wilcoxon rank sum, and Kruskal-Wallis tests) as appropriate. To account for multiple testing, a p value of <0.01 was considered significant. The (blinded) University Institutional Review Board determined that this study met criteria for exemption. Our study followed Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines for observational studies (23).

Results

The Table 1 shows the characteristics of California EDs, the EPs in these EDs, the patients served, and the surrounding communities stratified by priority county status and teaching hospital affiliation. A California county-level map demonstrating the location of clinically active EDs, stratified by priority county status and teaching hospital affiliation, is presented in Figure 1. A total of 320 clinically active EDs in California are presented: 178 in priority counties, 29 affiliated with teaching hospitals, and 24 with both characteristics. Of the 12,869,889 ED visits represented, 61.8% occurred in priority counties, 14.7% in EDs affiliated with teaching hospitals, and 12.0% in EDs with both characteristics. For context, Figure 2 demonstrates the location of other non-California CDC designated priority jurisdictions.

Table 1: Characteristics of California Emergency Departments, Patients Served, and Surrounding Communities.

The table includes 2021 calendar year information from California Area Health Resource Files released in 2022; patient demographic information was limited to a subset of available information and represented populations at higher risk for HIV.^1,4,13,14 Emergency Physician data reflects California county-level physician density data from the 2019 American Medical Association Physician Masterfile adjusted with 2020 county-level US Census population estimates. Estimates of percent living in poverty (for all ages) are from the 2020 US Census Bureau county-level Small Area Income and Poverty Estimate report released in 2021; HIV data reflect 2019 county-level CDC prevalence rates adjusted per 100,000 population; they are inclusive individuals aged 13 years or older, from individuals of all races/ethnicities. HIV prevalence for Alpine and Modoc counties are suppressed in this data. All data were publicly accessed and were the most recent data available.

	All California Emergency Departments (n=320)
	EDs in Priority Counties (n=178)				EDs in Non-Priority Counties (n=142)
	Teaching (n=24)		Not Teaching (n=154)		Teaching (n=5)		Not Teaching (n=137)
	n	(%)	n	(%)	n	(%)	n	(%)	P value
ED Characteristics
Medical Service Study Area									<0.01
Frontier	0	0.0	2	1.3	0	0	11	8.0
Rural	0	0.0	5	3.2	0	0	52	38.0
Urban	24	100.0	147	95.5	5	100	74	54.0
Licensed Bed Size									<0.01
1 – 99	0	0.0	16	10.4	0	0	54	39.4
100 – 199	1	4.2	52	33.8	1	20	41	29.9
200 – 299	4	16.7	35	22.7	2	40	23	16.8
300 – 499	14	58.3	44	28.6	0	0	18	13.1
500 or more	5	20.8	7	4.5	2	40	1	0.7
Median Visit Volume (IQR)	63,932 (49,864–81,315)		36,428 (20,720–56,436)		63,653 (56,891–78,208)		27,880 (14,107–47,658)		<0.01
Patient Demographics
Male	781,034	50.7	2,992,732	46.7	177,184	50.4	2,128,015	46.6	<0.01
Homeless	99,294	6.4	174,603	2.7	13,029	3.7	100,245	2.2	<0.01
Age Categories									<0.01
10 – 19	123,060	8.0	480,564	7.5	27,411	7.8	387,194	8.5
20 – 29	230,826	15.0	951,810	14.9	58,389	16.6	643,367	14.1
30 – 39	236,214	15.3	952,751	14.9	59,297	16.9	655,884	14.4
40 – 49	199,104	12.9	780,644	12.2	47,143	13.4	535,813	11.7
50 – 59	209,479	13.6	828,685	12.9	48,214	13.7	555,137	12.2
60 – 69	177,624	11.5	767,974	12.0	40,977	11.6	529,674	11.6
Hispanic or Latino Ethnicity	742,332	48.1	2,573,197	40.1	157,228	44.7	1,694,910	37.1	<0.01
Black or African American Race	230,034	14.9	827,957	12.9	29,797	8.5	289,977	6.3	<0.01
Patient Insurance Status									<0.01
Medicare	284,500	18.5	1,513,150	23.6	60,657	17.2	1,147,209	25.1
Medicaid (Medi-Cal Program)	816,724	53.0	2,524,446	39.4	171,670	48.8	1,756,395	38.5
Self-Pay or Uninsured	84,200	5.5	351,215	5.5	23,633	6.7	224,094	4.9
	EDs in Priority Counties (n=178)				EDs in Non-Priority Counties (n=142)
	Median (IQR)				Median (IQR)
Department Characteristics
EPs per 100,000 residents, per county	18 (14–21)				11 (4–20)				0.18
Surrounding Population Characteristics
HIV prevalence rate per 100,000	446 (321–499)				153 (128–220)				<0.01
Estimates of Poverty for All Ages	11 (9–13)				13 (10–15)				0.24

Abbreviations: CDC (Centers for Disease Control and Prevention), ED (emergency department), EP (emergency physician), HIV (human immunodeficiency virus), IQR (interquartile range), US (United States)

Figure 1: County-level Map of California Emergency Departments by County Priority Status and Teaching Hospital Affiliation.

California has 8 CDC-designated priority counties (Alameda, Los Angeles, Orange, Riverside, Sacramento, San Bernardino, San Diego, and San Francisco); these counties are shaded in grey. Individual EDs in these counties are superimposed and colored according to their teaching affiliation.

Abbreviations: ED (emergency departments), CDC (Centers for Disease Control and Prevention)

Figure 2: United States Map of Centers for Disease Control and Prevention HIV Priority States and Counties.

Figure adopted from Centers for Disease Control and Prevention US priority jurisdictions listed in the Ending the HIV Epidemic.^6,7

EDs in priority counties (compared to those in non-priority counties) were more frequently affiliated with teaching hospitals (13.5% versus 3.5%, p<0.01). They had higher median visit volumes (40,677 [21,422–62,862] versus 28,330 [14,682–49,688], p<0.01) and larger hospital licensed bed sizes (e.g., 39.3% versus 14.8% were x≥300 beds, p<0.01). The patient populations served by EDs in priority counties were more often composed of those at higher risk and CDC-targeted populations (e.g., persons who are homeless [3.4% versus 2.3%, p<0.01], individuals who identified as Black or African American race [13.3% versus 6.5%, p<0.01], and individuals who identified as Hispanic or Latino ethnicity [41.7% versus 37.7%, p<0.01]).

EDs affiliated with teaching hospitals (compared to those not affiliated with teaching hospitals) had higher median visit volumes (63,908 [52,023–81,272] versus 32,211 [16,063–51,010], p<0.01) and larger hospital licensed bed sizes (e.g., 72.4% versus 24.1% were x≥300 beds, p<0.01). They also more often served those at higher risk and CDC-targeted populations (e.g., persons who are homeless [5.9% versus 2.5%, p<0.01], individuals who identified as Black or African American race [13.7% versus 10.2%, p<0.01], and individuals who identified as Hispanic or Latino ethnicity [47.5% versus 38.9%, p<0.01]).

The subset of EDs in priority counties with teaching hospital affiliations (compared to both ED groups without a teaching hospital affiliation) had higher overall median visit volumes and higher proportions of visits for those at higher risk and CDC-targeted populations (e.g., individuals who were homeless, individuals who identified as Black or African American race, and individuals who identified as Hispanic or Latino ethnicity, all p<0.01) (Table 1 and Supplemental Figure). Further, although all California EDs groups cared for sizable proportions of patients who were uninsured, this same subset more often served Medi-Cal (i.e., California’s Medicaid program serving low-income individuals) or Self-Pay or Uninsured individuals (Supplemental Figure). As seen in our map (Figure 1), there is a higher density of EDs, both all EDs and those with teaching hospital affiliations, in California priority counties compared to those in non-priority counties.

Discussion

We demonstrate that California EDs in CDC-designated priority counties, notably the subset also with teaching hospital affiliations, are major sources of healthcare in the state. Further, compared to other EDs in priority counties, those with teaching hospital affiliations more often serve populations known to be disproportionately impacted by HIV. Given their teaching hospital affiliation, these EDs are also linked to healthcare systems with resources and capabilities not present in either smaller EDs or those not affiliated with teaching hospitals (e.g., presence of social work services, infectious disease specialists, and institutional infrastructure that would be conducive to routine ED-based HIV testing services). Although these EDs account for a minority of all ED care provided in California, from an epidemiologic perspective, they serve populations with higher pre-test probabilities of undiagnosed HIV infections and are linked to systems with the highest odds of sustaining routine ED-based HIV testing. Compared to the other EDs, they are likely the most efficient and pragmatic venues for targeted HIV efforts that align with recent state and national initiatives.

Despite previous CDC recommendations encouraging universal opt-out HIV testing in all healthcare settings, rates of HIV testing remain low (8); this includes low rates of testing in EDs that have persisted in recent years (8, 10, 12). Some of this may stem from EPs’ concerns that HIV testing will increase costs; other EPs report worry that it might increase length of stay or divert resources needed elsewhere (11). Furthermore, some EPs express opposition and feel that preventative care services should not be offered in the ED (11).

Regardless, patients with undiagnosed HIV are seen in EDs, not tested, and contribute to the ongoing epidemic – on both a state and national level. As more acute care is shifted away from primary care settings and into EDs, there is an increasingly clear need to better understand how EDs can contribute to ending the HIV epidemic (6, 7, 15, 24). Recently refocused national initiatives incorporate a targeted, geographic, focus on areas hardest hit by HIV. Why should the ED response be any different?

Refocusing ED-based HIV testing initiatives and incorporating a targeted response might be the most feasible path forward – it may be the best chance of diagnosing new infections and may represent the most sustainable option. Refocused efforts could utilize either physician- or nursing-initiated screening approaches and be modeled after existing ED-based programs (18, 25). These efforts could reflect universal screening for all ED patients without prior HIV testing or instead use a targeted approach based on either clinical judegment or incorporating a validated quantitiave risk score; the latter of which would likely require fewer tests and less person hours (25). Further, beyond simply uncovering new infections, new diagnoses in these EDs could more quickly link persons with HIV to care, decrease time from diagnosis to viral suppression, and improve long-term HIV/AIDS-related outcomes (7, 26 ,27). Given the unique nature of these teaching hospitals, targeted efforts aimed at initiating or expanding HIV-related efforts could also help spur ED-based pre-exposure prophylaxis (PrEP) programs for at-risk individuals; this could include PrEP prescribed in the ED or alternatively ED-based referrals to outpatient programs (27, 28). Recent state and national action reducing both access- and cost-barriers to PrEP initiation suggest such programs could be feasible and sustainable (29, 30). Although diagnosing HIV is only the first of many steps in patient engagement, uncovering these infections still requires a test.

Limitations

Our findings should be interpreted in the context of several limitations. First, the data was annotated with Black or African American race and Hispanic or Latino ethnicity classifications (19); we acknowledge the limitations inherent to these classifications (31). In line with updated guidance on the reporting of race and ethnicity, we have concurrently reported additional sociodemographic factors on both ED patients and their surrounding communities (31). Second, we are unable to comment on which EDs in this study currently have ED-based HIV testing services and the details of any such services (e.g., opt-in, opt-out, or targeted screening). It is possible that some EDs may already offer some degree of routine HIV testing. However, we and others have shown that a minority of EDs offer HIV testing and only a minority of ED patients are tested for HIV (8, 10, 14). We have also shown that this low rate of testing persists even for persons – from both populations and regions known to be hardest hit by HIV - who present to EDs with targeted chief complaints (e.g., signs or symptoms of sexually transmitted infections or those presenting after sexual assault) (12). Even if some (or all) of these EDs currently offer some degree of HIV testing, these efforts appear insufficient. New and refocused efforts that can efficiently reach at-risk populations and areas are required (6, 7, 15). Third, this work was limited to California. These findings may not be generalizable to EDs in other states or different US regions; future work by our group is geared towards characterizing a national sample of EDs. Fourth, given limitations of the data, we were unable to account for other important populations at elevated risk for HIV infection (e.g., men who have sex with men and those who engage in intravenous drug use). Incorporating these populations into future work will be critical.

Conclusions

Given the ongoing epidemic, recent state and national initiatives have been refocused to target areas hardest hit by HIV. Despite an increasing amount of care being provided in US EDs, rates of ED-based HIV testing are persistently low. There is an unmet need for HIV testing in US EDs. An important, unanswered, question to consider is how EDs can most efficiently respond to refocused efforts and best contribute to helping end the HIV epidemic. One approach could be focusing efforts on EDs with teaching hospital affiliations in priority counties. As we demonstrate here, these EDs are major sources of healthcare in the state. They serve patient populations known to be disproportionately impacted by HIV. They are also linked to systems capable of supporting routine HIV testing services. Ending the HIV epidemic requires acknowledging that we continue to miss undiagnosed infections in EDs. Without more routine testing in US EDs, many of these infections will remain undiagnosed, and the epidemic will continue.

Article Summary.

Why is this topic important?

There is a need to increase HIV testing in US EDs. Recent state and national initiatives aimed at ending the HIV epidemic in California include a geographic component and target “priority” counties hardest hit by HIV.

What does this study attempt to show?

There are a subset of EDs in California that might be the most efficient, high-yield, targets toward which to direct future HIV testing efforts.

What are the key findings?

A subset of EDs in California priority counties with teaching hospital affiliations are major sources of healthcare in the state. Compared to other EDs groups, these may be more capable of supporting HIV testing services.

How is patient care impacted?

Increasing rates of HIV testing in EDs could be the most efficient way to reduce the burden of undiagnosed infections in California. Increasing the rate at which persons living with HIV are diagnosed in California could more quickly link patients to care, improve HIV/AIDS-related outcomes, and help end the ongoing epidemic.