PrEP Access in Federally Qualified Health Centers Across 11 United States Metropolitan Statistical Areas

Abstract

PrEP is a promising tool for HIV prevention, but uptake has been slow in key demographics and geographic areas including racial and ethnic minorities. Federally Qualified Health Centers (FQHCs), serving those with heightened risk of contracting HIV, including low income and minority patients regardless of ability to pay, are potential sites for PrEP delivery. This study aims to determine availability of PrEP at FQHCs in the US. FQHCs in the 11 largest U.S. metropolitan areas were included. The south included Atlanta, Dallas, District of Columbia, Houston, and Miami. Non-south included Boston, Chicago, Los Angeles, New York, Philadelphia, and San Francisco. We randomly selected 360 FQHCs for phone calls in which investigators queried the availability of PrEP for HIV prevention at each center. The study was powered to determine a 10% difference in proportion of clinics with PrEP services by region. We used a bivariate logistic regression to compare cities and regions. The percentage of FQHCs providing PrEP generally was low within this sample, with 0–28.0% offering PrEP services in the southern and 14.0–33.3% in non-southern metropolitan areas. Overall, 19.3% of clinics sampled indicated that they offered PrEP. Logistic regression did not show any difference between southern and non-southern regions (p=0.779). However, the total number of FQHCs was significantly lower in southern metropolitan areas (p=0.014). FQHCs in all metropolitan areas provided limited access to PrEP. Interventions, including technical assistance to increase PrEP availability in these settings catering to underserved populations, could be beneficial.

Introduction

The iPrEX trial of emtricitabine/tenofovir disoproxil for PrEP in men who have sex with men (MSM) and transgender women showed a 44% reduction of HIV infections in the modified intention to treat analyses, (1) and several studies have produced comparable and improved outcomes, including up to 86% reduction of HIV infection rates among groups at high risk for HIV (2–5). There are disproportionately high rates of new HIV infections in the southern cities in the United States (6). In addition to geographic disparities, HIV infection continues to disproportionately affect the MSM community. Approximately 1 in 14 MSM within the south live with HIV (7). Although existing data increasingly supports the efficacy and acceptability of PrEP among high-risk populations, only 5% of gay men have actually been prescribed and taken PrEP (8–11). A follow-up study from the Miami and San Francisco sites of the U.S. PrEP Demonstration Project demonstrated wide discrepancies in availability of PrEP. Participants at the Miami site reported frequent barriers to PrEP after study completion compared with respondents at the San Francisco site; access to a PrEP provider was a major limiting step in obtaining PrEP for this cohort (12).

A major barrier to acquisition of PrEP outside of study settings is primary care provider reluctance to prescribe PrEP (13, 14). For an individual seeking PrEP, a first step can be a call to a health center to determine whether a PrEP provider is available. As the initial point of contact in the process of acquiring PrEP, the health center can play a critical role in enhancing or discouraging PrEP uptake. Federally Qualified Health Centers (FQHCs) are potentially important venues for PrEP access because they see individuals regardless of ability to pay and largely serve low income and minority populations, those at highest risk for HIV (15). These institutions are uniquely capable of providing culturally appropriate and locally relevant interventions to support HIV prevention, making them ideal venues for PrEP dissemination(16). The quantity, quality, and consistency of information given to PrEP seekers during initial contact with a health center is unclear.

The aim of this study was to determine the availability of PrEP at FQHCs in the 11 largest Metropolitan Statistical Areas (MSAs) in the United States. We hypothesized that PrEP will be more readily available in non-southern MSAs (New York, Boston, Philadelphia, Chicago, Los Angeles, San Francisco) versus southern MSAs (Atlanta, Miami, Houston, Dallas, Washington D.C.).

Methods

Site Selection

The 11 largest MSAs according to Census Bureau data were included in the study. The non-southern MSAs were New York, Boston, Philadelphia, Chicago, Los Angeles, and San Francisco. The southern MSAs were Atlanta, Washington D.C., Miami, Houston, and Dallas. The list of all FQHCs in each MSA was obtained from the Health Resources and Services Administration (HRSA) website. We chose FQHCs for the survey because of the comprehensive primary and preventative care services regardless of the ability of patients to pay. The 11 cities were comprised of a total of 1303 FQHCs. A sample of 360 clinics was randomly selected. Clinics in primary or secondary schools were excluded from the random selection process. Duplicate clinics and clinics that did not have a working phone or had closed were removed, decreasing the original sample of 360 clinics to 332. Grantee-level characteristics data from 2016 was obtained using the HRSA website, and aggregated patient gender data was acquired from the National Association of Community Health Centers.

Procedures

The investigators conducted scripted telephone calls to randomly selected FQHCs. The caller (male or female) stated that s/he was interested in obtaining PrEP and inquired about the availability of a provider who was willing to discuss this intervention. If the receptionist did not know what PrEP was, the caller proceeded as outlined in the script, requesting a pill to prevent HIV. We used the results to estimate availability of PrEP at the FQHCs in the various MSAs. Phone calls were made between August 1, 2015 to January 31, 2017.

Measures

In addition to transcripts of telephone calls, a checklist was developed to measure clinic staff knowledge of PrEP and availability of a clinician able and willing to prescribe PrEP at each clinic.

Data Management and Statistics

Data was collected on a password-protected computer and stored in a password-protected Excel spreadsheet. Other than MSA name, no identifying information on the health center was recorded. The study was powered to determine a 10% difference in proportion of clinics with PrEP availability in southern vs non-southern MSAs. Responses that were recorded were analyzed using chi-square analysis by MSA (10 degrees of freedom) and by southern versus non-southern MSA groups. The original sample size was 360 unique clinics. Clinics were coded as prescribing PrEP onsite, Not Prescribing PrEP onsite, Unknown availability (wherein clinic staff either did not know about PrEP or did not know whether PrEP would be prescribed by FQHC physicians) or as no contact after 2 attempts. We examined differences in response patterns across the southern and non-southern cities and across cities using chi-square analysis. Bivariate logistic regression comparing sites who confirmed availability of PrEP to all other sites was used to calculate confidence intervals around the probability of finding an FQHC in each city that would confirm availability of PrEP. Another bivariate logistic regression was also used to compare sites with confirmed availability of PrEP between southern and non-southern sites.

Provisions to Protect the Privacy Interests of Subjects

The institutional review board at the University of Miami determined that this study was exempt from review as human subjects were not involved (IRB# 20150580). FQHCs were assigned a study number after random selection (1–360). At the time of phone call to each FQHC, only the study number was recorded on the data collection sheet and MSA was identified. No information about clinic name, staff name, or phone number was recorded.

Results and Discussion

The FQHCs surveyed largely served racial and ethnic minorities (Table 1). Except for Boston (82%), all MSAs had greater than 90% of the patients they served living at or below 200% of the federal poverty line. Dallas, Houston, and Miami, municipalities located in non-Medicaid expansion states, had over 50% uninsured clients. The non-southern MSAs, all within states with Medicaid expansion, had fewer than 25% uninsured patients. The percentage of patients receiving HIV care services was universally low, ranging from 0–8%. Characteristics varied widely across grantees within the same MSAs.

Table 1.

Grantee-Level Characteristics

		Number FQHCs	Adults (18–64) (%)	Female (%)	Male (%)	Black (%)	Hispanic (%)	Non-English (%)	Below 200% poverty line (%)	Uninsured (%)	Medicaid (%)	Medicare (%)	Mental Health (%)	Drug txa (%)	HIV txb (%)
Southern
	Atlanta	20	66	39	61	80	11	15	95	41	34	6	6	0	2
	Dallas	25	68	38	62	27	51	24	96	52	29	5	4	1	0
	D.C.	43	73	45	55	54	32	32	92	27	47	7	14	4	8
	Houston	62	69	44	56	36	48	35	96	50	28	4	10	1	6
	Miami	77	63	47	53	43	43	28	90	52	28	3	18	0	5
Non-Southern
	Boston	81	70	46	54	34	36	42	82	21	44	8	8	0	1
	Chicago	229	60	41	59	45	40	25	97	25	60	5	10	3	1
	Los Angeles	270	70	43	57	17	61	42	96	27	62	4	6	0	1
	New York	201	66	43	57	49	42	27	92	18	60	8	13	1	4
	Philadelphia	71	70	42	58	70	40	15	94	17	63	8	13	1	5
	San Francisco	224	63	47	53	21	40	36	95	24	61	9	10	4	1

^a)Drug treatment services

^b)HIV primary care

1. Drug treatment services

2. HIV primary care

There was a significant difference in the average number of grantee sites in the southern MSAs and the non-southern MSAs (10.8 vs 28.2, p=0.014). A test was done to determine if the average of the number served was statistically different between southern and non-southern MSAs. The difference was 18,379 patients for southern MSAs versus 26,162 for non-southern MSAs (p=0.052).

Only 19.3% of clinics confirmed on-site availability of PrEP, and 26.8% confirmed that PrEP was not available at their location. Of the remaining sites, 27.7% either did not know about PrEP or did not know whether physicians on site might prescribe PrEP and suggested making an appointment to find out, and 26.8% never answered the phone or returned phone calls. There were no differences in these proportions between southern and non-southern sites, however, there was significant variability across cities (p=.003).

Availability of PrEP services in FQHCs generally remained low within this sample, with 0–28.0% indicating provision of these services across southern MSAs and 14.0–33.3% offering PrEP services across non-southern MSAs, with New York City having the greatest percentage of FQHCs offering PrEP services in our survey (Figure 1). The logistic regressions presented in Figure 1 did not show any significant difference between the MSAs in terms of PrEP availability (p=0.214) relative to non-availability. When comparing southern to non-southern, the p-value was 0.779.

Figure 1. Proportion of Clinic Sites Offering PrEP by Metropolitan Statistical Areas (MSAs) with 95% Confidence Interval.

* No FQHCs in Dallas reported PrEP availability, and thus the confidence interval cannot be calculated.

Although scale-up of PrEP has occurred in certain MSAs in recent years, PrEP knowledge among FQHC staff remained universally low across MSAs surveyed and less than 20% of clinics could confirm availability of PrEP over the phone. This study highlights an urgent need for an educational campaign within healthcare systems to teach staff about the availability of PrEP for HIV prevention. According to CDC Surveillance data from 2017, of the 38,281 HIV diagnoses in the US, 44% are among African Americans and 25% are among Hispanics and Latinos. Further, HIV incidence is much higher in the south (17). The FQHCs surveyed overwhelmingly served racial and ethnic minorities who have the highest risk for HIV, especially among the southern MSAs. Therefore, interventions such as technical assistance to increase PrEP availability among resource-limited populations, could be beneficial. While we did not detect a difference between southern and non-southern MSAs in terms of proportion of FQHCs offering PrEP, this was likely due to lack of power to detect a difference in the overall lower-than-expected number of PrEP-providing FQHCs. Additionally, there were significantly more FQHCs in the non-southern MSAs. The sites in the non-south also tended to serve more participants on average than those in the south.

Limitations to this study include the 18-month study period, over which there were large PrEP campaigns in some of the included MSAs. However, the study revealed a general lack of knowledge and availability nationwide that remains consistent with low rates of PrEP coverage. Whereas the administrative staff’s knowledge may not reflect the actual provision of PrEP in these locations, these frontline staff are a major source of information for patient inquiries so this information is relevant to a patient’s perceived availability. Another limitation was quality of data available from the HRSA website. Many FQHCs had incorrect phone numbers or did not offer medical care and had to be excluded from the analysis. The data presented in Table 1 is a representation of grantee-level data. In some cases, a grantee had sites located outside of the MSA that were also incorporated into their characteristics data. This should not have impacted the aggregated statistics significantly since outlying FQHCs were generally close geographically. Administrative FQHC sites are also included in the HRSA data, but were excluded from our analysis. Additionally, this study does not address the disproportionate impact that a single center with strong PrEP services can have in a region. A single center with a well-developed PrEP program, whether in an FQHC or other venue, can result in more PrEP availability than is suggested by the percentages for each MSA. We consider, however, that our survey may represent the experience of individuals unaware of specialized centers and attempting to seek PrEP through a locally convenient and economically accessible primary care center

Conclusions

Despite these limitations, our findings parallel those of a recent report that MSM in all geographic regions have limited PrEP provider access and that availability is greatest in the urban Northeast and lacking most profoundly within non-urban areas and the south (18). This report provides important evidence that PrEP availability remains scarce nationwide for individuals seeking care through FQHCs. Educational campaigns for healthcare workers, including clinic staff might be effective in increasing availability of this important HIV prevention tool in cities with large vulnerable populations.