Nationally, rates of sexually transmitted diseases (STDs) are on the rise. From 2013 to 2017, rates of chlamydial infection and gonorrhea rose 22% and 67%, respectively.1 Chlamydia is the most commonly reported notifiable disease in the U.S. with a rate of 528.8 cases of chlamydia infections per 100,000 persons reported in 2017.2,3 Gonorrhea, although not as prevalent as chlamydia, is the second most commonly reported infection in the U.S. and had a rate of 171.9 cases per 100,000 persons the same year (2017).3
Presenting for STD care is a complex decision that often includes multiple factors.4 From a clinical treatment perspective, STD care is often best addressed in a clinic setting because these locations have procedures in place to follow patients longitudinally and provide retesting when needed, and provide specialty services specific to STD testing not always available elsewhere (eg, rectal and pharyngeal sampling).5 Yet, emergency departments (EDs) which are less equipped to provide these services are seen as acceptable venues for STD-related care by high-risk populations.6 Factors which can prompt individuals to seek care in the ED for nonurgent matters include negative perceptions of other local care options and the perceived convenience of the ED as compared to other points of care,7 particularly among those seeking STD treatment.8 Many individuals may also be unaware of local clinic-based STD services.6
The literature on the subject of clinic-based STD care location use versus ED use from a spatial perspective is sparse. Studies illustrating the utility of spatial analysis for public health problems such as the opioid epidemic and optimal placement of naloxone supplies at pharmacies located near known overdosing locations have been effective9 so the potential for the use of spatial analysis in terms of STD care utilization is promising. Individuals who live within close proximity to a hospital are likely to visit it for nonurgent complaints which could be treated by a general practitioner.10 Strengthening the argument that individuals will use proximity as a factor in deciding where to go to seek care, geography has been shown to be a better predictor for hospital readmission rates than hospital system affiliation.11
The City of St. Louis, Missouri carries a high burden of chlamydia (1,279 cases per 100,000 persons) and gonorrhea infection (750 cases per 100,000 persons).12 Despite widespread local and national increases in cases of chlamydia and gonorrhea and the continuing need for care,13 the availability of clinical services for STD testing and treatment has been declining. This reduction in the availability of services has an established, direct, measurable impact on those needing STD care, wherein a lower percentage of infected patients get tested.14 A recent study of approximately one fourth of the 1224 local health departments in the United States found that 61.5% reported budget cuts in the past year, and over 40% of those health departments reported reductions in overall clinic hours, routine screening, and partner services.15 Clinic closures may be contributing to the increased incidence of gonorrhea and chlamydia, because individuals lack the packaged services including follow-up that were available only a few years prior.16
The St. Louis region, like other regions of the country, has experienced closures of traditional STD clinics, making access to quality STD care for infected individuals more difficult to obtain. Each time a STD clinic closes the safety net becomes more porous,17 and individuals seeking care must turn to other options, such as EDs. Therefore, the purpose of this article is to investigate whether clinic-based STD care locations exist in close proximity to the home address of individuals who utilize the ED for chlamydia and gonorrhea testing and treatment.
MATERIALS AND METHODS
The clinical data used for this study were collected from 4 hospital-based EDs, 2 urban and 2 suburban, in the St. Louis area which are all part of the same health system. One of the urban EDs is for children. All visits where an STD test was administered or an STD diagnosis code was recorded between January 1, 2010, and May 31, 2016, were included, although 1 suburban hospital did not begin STD testing until 2012. Data associated with each visit included patient age, sex, race, International Classification of Diseases (ICD)-9/ICD-10 diagnosis codes, STD test type and result, patient address, and date and time of presentation to the ED. The home address of each patient was geocoded.
The sample was narrowed to include visits specifically where STD-related care was provided, ones where we could reasonably assume the patient could have gone to a clinic-based location offering STD care services instead of the ED. We defined a visit in which STD-related care was provided as when both a nucleic acid amplification test (NAAT) for gonorrhea and chlamydia was administered and an STD diagnosis code was recorded; only visits that met both criteria were included in the sample. Additionally, only visits made by individuals 13 years and older who listed a home address in the City of St. Louis or St. Louis County18 at the time of their visit were included. Visits with a diagnosis code indicating sexual assault were excluded because their primary intent in visiting the ED was not necessarily STD-related. For patients who made more than 1 visit and received STD-related care during the study period, only their first visit was included.
The provider data for this study is a list of clinic-based STD care locations in the St. Louis area, compiled by the St. Louis Sexually Transmitted Infection Regional Response Coalition in 2016 and under the direction of the senior author.19 Specific information on each clinic-based STD care location includes organization name, address, chlamydia, and gonorrhea testing availability (yes/no), walk-in availability, no charge or low cost, location's days and hours, and type of clinic-based STD care location. Low cost was defined as clinic-based STD care locations which have a sliding scale for billing. All locations were open to the general public, and all provided testing for gonorrhea and chlamydia. The addresses of each clinic-based STD care location and the 4 hospital-based EDs were geocoded.
The software R was used for all spatial analyses and mapping.20 To approximate clinic-based STD care coverage for the patient population visiting the ED for STD-related care, the number of patients within multiple radii of each clinic-based STD care location was calculated: 0.25, 0.5, 1, 2, and 5 miles. These radii created circular zones around each clinic-based STD care location in a point to point distance measurement (often referred to as an “as the crow flies” distance); we then counted the number of patients that reported a home address within each circular zone or radius. A quarter mile was chosen because it is the standard distance which is assumed by Americans to be preferable for a walk over a drive,21 a half-mile was chosen because this distance is approximately the mean distance people cover when they do walk,22,23 and a mile was chosen as a distance that indicates a relatively close place that can be accessed via car or a walk.24 The 2-mile designation was chosen based on a previous spatial analysis where the average distance from residence to a human immunodeficiency virus testing site was slightly above 2 miles.25 The 5-mile designation was chosen as a distance which has been cited as a close market area of hospital catchment, specifically in reference to alternate specialized clinical services.26 Patients were stratified based on their county of residence. St. Louis City is an urban core city and as such is classified here as “urban.” St. Louis County is the county adjacent to St. Louis City and does not contain St. Louis City within its boundaries and is classified for this article as “suburban.” Both of these classifications are based on population density (>3,000 people per sq/mi for urban and 1000–3000 people per sq/mi for suburban).27 Work with spatial data in R was enabled by the simple features package.28
To determine the percentage of patients who lived closer to a clinic-based STD care location than the ED they visited, we computed odometrical route distances for (1) the distance from each patient's address to the ED they visited, and (2) the distance from each patient's address to every clinic-based STD care location, and used the shp2graph package29 which converts the spatial road data into a network. That network was then processed using the igraph package.30 We assumed that patients drove the shortest distance from their home address to the hospital they visited and would also do so for each clinic-based STD care location. We defined a missed clinic opportunity as occurring when a patient visited the ED while a clinic-based STD care location (1) was within the same distance or closer to the patient's home than was the emergency room they visited, (2) was open at the time of the visit, (3) offered no-charge services, and (4) had walk-in availability. For each visit, we matched the ED visit day-of-the-week and time of day with the open hours of each clinic-based STD care location—and in some cases day of the month when clinic-based STD care locations were open (eg, third Thursday of the month). Clinic-based STD care locations that were tagged as having no-charge services and walk-in availability were combined with distance results and day/time of the visit for each patient to produce the number of patients who missed a clinic opportunity, per our definition.
Maps were created using the tmap package.31s The heat maps were produced using a kernel density estimation based on the location of all individuals in the sample and for individuals with a missed clinic opportunity. These analyses were approved by the institutional review board at Washington University in St. Louis.
There were 59,879 visits in the original data set. Visits were excluded in the following order: first, visits in which there was not an STD diagnosis in combination with NAAT testing were removed (n = 48,447). Second, visits where the patient was under the age of 13 (n = 46) were removed. Third, 146 individuals were excluded on the basis of having a sexual assault ICD-9/ICD-10 codes. Fourth, visits from patients who lived outside of the study area were excluded (n = 654). Fifth, visits where a geocoder could not match to a point or street address were removed (n = 1590). For individuals who presented to the ED multiple times within the study period, 2676 repeat visits were excluded. Finally, observations where we could not identify which hospital the patient presented to were removed (n = 220). The final sample size was 6100 first visits by unique individuals.
The sample was predominantly female (78.4%) (Table 1). The mean age and standard deviation of males (28.92; 10.81) were similar to that of females (27.45; 9.59). The majority of both males (60.8%) and females (65.7%) reported addresses in St. Louis County. St. Louis City was home to 2158 (35.4%) patients and 3942 (64.6%) patients lived in St. Louis County. Hospital A in the city and hospital C in the county saw the majority of patients (35.0% and 46.8%, respectively). Overall, slightly under half of all visits were made between the weekday hours of 8:00 AM and 5:30 PM for both males (41.6%) and females (43.2%).
Of the 33 clinic-based STD care locations in the analysis, 60% had walk-in availability,20 33.3% were no charge,11 and another 60.6% were considered low cost20 (Table 2). Almost all of the clinic-based STD care locations (31, 93.9%) had at least 1 weekday in which they were open a minimum of 8 hours. Twenty-five (75.8%) had hours before 8:00 AM or after 5:30 PM at least 1 day between Monday and Friday. Thirteen (39.4%) had weekend hours. The majority of clinic-based STD care locations in the study area were Federally Qualified Health Centers (FQHC) (18, 54.5%).
At a radius of 2 miles, 98.9% of city residents in the study were covered by at least 1 clinic-based STD care location (Table 3). At this same radius, only 49.5% of county residents were covered by at least 1 clinic-based STD care location. Over half of the city residents lived within a mile of a no-charge clinic-based STD care location (54.4%), whereas only 5.2% of the county residents did. At the 0.5-mile radius, designation of 35.3% of city residents was covered by a clinic-based STD care location but only 5.6% of county residents were. At a quarter of a mile, 11.6% of city residents and 1.3% of county residents lived within that distance from a clinic-based STD care location. We investigated the proximity of individuals' home addresses to the only traditional health department-based specialty STD clinic in the region and found that 50.6% (3086/6100) lived closer to that clinic-based STD care location than the hospital where they presented. In the city, almost all (99.1%) individuals were closer to at least 1 clinic-based STD care location than the ED they visited (Table 4). In the county, this number fell to 82.2%. Fifty-seven percent (1,242) of individuals residing in the city were closer to an open clinic-based STD care location than the ED they visited, compared with 45.8% (1804) individuals in the county. Ninety-eight percent (2114) of individuals residing in the city were closer to a no-charge clinic-based STD care location compared with 57.2% (2255) in the county. The most explicit indicator of a missed clinic opportunity encompasses clinic-based STD care locations that are within the same distance as the hospital, open at the time of ED admission, no charge, and available for walk-in appointments. Forty-six percent (993) of the individuals in the city were closer to a clinic-based STD care location with all of these criteria than the ED where they presented, and in the county, 26.1% (1027) of individuals were closer to a clinic-based STD care location with these criteria. For the entire sample investigated (city and county), there were 2020 (33.1%) visits in which a clinic-based STD care location was closer, open, no charge, and available for walk-in appointments.
Figure 1 displays the locations of hospitals and clinic-based STD care locations in relation to the density of the entire study population (6100). Figure 2 displays only those individuals who missed the opportunity to use an open, no-charge, walk-in clinic-based STD care location within the same distance as the hospital (2020).
Availability and cost are commonly cited barriers to accessing STD care 32s and are strong predictors of service utilization across all types of health services. Our study demonstrates that both of these barriers have the potential to be overcome. Because the most frequent barrier to STD care for our population was time-related availability, with half of our sample presenting to the ED when no other clinic-based STD care location was open and closer to their home residence than the ED, a potential implication is that clinic-based STD care locations may not have the hours that their clients require. Expanding clinical care hours may provide the flexibility that patients need to find STD services in a clinic location as opposed to the ED.
If cost was a perceived barrier for the patients in our study, only 29.4% did not have a no-charge clinic-based STD care location closer to their residence than the ED where they presented. Given the recent defense of traditional STD clinics,33s it is notable that over half of the sample population lived closer to the only no-charge clinic-based STD care location of this type in the region. These no-charge clinic-based STD care locations may not have been open at the time that the patients presented to the ED, but with knowledge of hours and availability, these patients may have been able to receive no-charge services instead of ED services where the cost of an STD-related visit can be US $1736 or more (in 2006 dollars).34s Potentially, insurance status may have played a role in where patients sought care, possibly due to assumptions that the ED would treat them if they were uninsured, whereas the clinic-based STD care location might not. If uninsured patients are encouraged to make use of these no-charge clinic treatment opportunities, it could result in substantial cost savings to the hospital system, which often covers the cost of care for these vulnerable populations.
Women made up a large percentage of our sample. This is consistent with STD testing and care rates in all settings where women are disproportionately affected by chlamydia and gonorrhea.3 A potential implication for this is that women need better STD care opportunities at clinic-based STD care locations rather than the ED. Additionally, women are more often asymptomatic than men, and this may play a role in their increased presentation at the ED rather than a clinic-based STD care location. Improving messaging from clinic-based STD care locations specifically targeting women, informing them of services provided and options for care may improve utilization.
Time of day and day of the week in which STD care clinical services were sought and proximity of residence to clinic-based STD care location are the biggest factors impacting a missed clinic opportunity. A third of our sample missed an opportunity to use a clinic-based STD care location that was closer, open, no charge, and available for a walk-in appointment. These missed opportunities were greater in the urban environment of the city where there are more clinic-based STD care locations; 46.0% of city residents missed an opportunity to use a clinic-based STD care location that was closer, no charge, and available. There are fewer community-based organizations in the suburban area than in the urban area and the opportunity to go to a closer no-charge clinic-based STD care location in the suburban area is 57.2% compared with the urban area of 98.1%. This gap in services in the suburban area is particularly important because the Northern sector of St. Louis County has high rates of both chlamydia (>1161 per 100,000) and gonorrhea (791 per 100,000).35s Sexually transmitted disease clinic-based care locations should be strategically located in geographic areas where chlamydia and gonorrhea rates are high and rising.
Our analysis was limited to nonpublic transport routing. Travel time is a known barrier to accessing care with disparities embedded in transportation by bus versus car.36s However, for this analysis, only direct transport was investigated (car, bicycle, walk). Public transportation routes which would likely be more circuitous were not investigated. Further investigation regarding specifically what type of transportation and the subsequent routes individuals use to go to the ED would be of interest. This would provide insight on whether more transportation options such as additional bus stops or bus line extensions may be warranted. However, because 3 of our 5 radii were a mile or less, this may not be a substantial limitation, given that these distances are walkable for many individuals. However, another potential limitation of this study is that the patients were assumed to have been coming from their residences but could have gone to the ED from an alternate location. Additionally, the distance assumed to be walkable may have contained obstacles in the built environment (street with no crosswalk) or may have been an unsafe journey in some other way.
Individuals may have had access to the clinic-based STD care locations that were not part of the analysis as a member of a special group (ie, students at the nearby school where a clinic was available), in which case, they would have had more access to clinic-based STD care and potentially leading us to underestimate the number of people who might have missed a clinic opportunity. Additionally, the list of clinic-based STD care locations might include locations that were available at the beginning of the study period and subsequently closed or might not contain locations that opened later.
Clinics providing STD services can increase marketing and advertising efforts to increase public awareness of the services which they provide and potentially expand their operating hours, if feasible to accommodate a greater percentage of patients needing their services. However, this may be too simplistic. Patients may not have been aware that they had signs or symptoms consistent with an STD until diagnosed in the ED, favoring the ED as a healthcare setting capable of diagnosing and treating a wide range of conditions. Of concern is the role that stigma may play in patients' choice of where to seek care for sexual health. People with an STD concern may make the choice to seek care in the ED intentionally, either because they believe the ED would do a superior job of diagnosing and treating them, or potentially to preserve their anonymity in going to a place further from their residence where they are less likely to encounter someone they know. Future research on intent, particularly qualitative interviews with patients could shed additional light on this issue.
2. Centers for Disease Control and Prevention. Selected nationally notifiable disease rates and number of new cases: United States, selected years 1950–2016. U.S. Department of Health and Human Services, 2017.
3. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance. 2017. U.S. Department of Health and Human Services; 2018.
4. Brackbill RM, Sternberg MR, Fishbein M. Where do people go for treatment of sexually transmitted diseases? Fam Plann Perspect 1999; 31:10–15.
5. Golden MR, Kerndt PR. Improving clinical operations: Can we and should we save our STD clinics? [editorial]. Sex Transm Dis 2010; 37:264–265.
6. Mehta SD, Shahan J, Zenilman JM. Ambulatory STD management in an inner-city emergency department: Descriptive epidemiology, care utilization patterns, and patient perceptions of local public STD clinics. Sex Transm Dis 2000; 27:154–158.
7. Uscher-Pines L, Pines J, Kellermann A, et al. Deciding to visit the emergency department for non-urgent conditions: A systematic review of the literature. Am J Manag Care 2013; 19–47.
8. Pearson WS, Peterman TA, Gift TL. An increase in sexually transmitted infections seen in US emergency departments. Prev Med 2017; 100:143–144.
9. Dodson ZM, Enki Yoo E-H, Martin-Gill C, et al. Spatial methods to enhance public health surveillance and resource deployment in the opioid epidemic. Am J Public Health 2018; 108:1191–1196.
10. Ingram D, Clarke D, Murdie R. Distance and the decision to visit an emergency department. Soc Sci Med D 1978; 12:55–62.
11. Rising KL, Karp DN, Powell RE, et al. Geography, not health system affiliations, determines patients' revisits to the emergency department. Health Serv Res 2018; 53:1092–1109.
12. Missouri Department of Health and Senior Services. Epidemiologic profiles of HIV, STD, and hepatitis in Missouri 2016.
13. Hoover KW, Parsell BW, Leichliter JS, et al. Continuing need for sexually transmitted disease clinics after the affordable care act. Am J Public Health 2015; 105(Suppl 5):S690–S695.
14. Centers for Disease C, Prevention. Impact of closure of a sexually transmitted disease clinic on public health surveillance of sexually transmitted diseases—Washington, DC, 1995. MMWR Morb Mortal Wkly Rep 1998; 47:1067–1069.
15. Leichliter JS, Heyer K, Peterman TA, et al. US public sexually transmitted disease clinical services in an era of declining public health funding: 2013–14. Sex Transm Dis 2017; 44:505–509.
16. Ducharme J. Why STD rates are higher than They've ever been. Time 2018; 8/29/2018.
17. Pathela P, Klingler EJ, Guerry SL, et al. Sexually transmitted infection clinics as safety net providers: Exploring the role of categorical sexually transmitted infection clinics in an era of health care reform. Sex Transm Dis 2015; 42:286–293.
18. Office of Management, Budget. 2010 standards for delineating metropolitan and micropolitan statistical areas; notice. Fed Regist 2010; 75:37246–37252.
20. Team RC. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org/
21. Atash F. Redesigning suburbia for walking and transit: Emerging concepts. J Urban Plan Dev 1994; 120:48–57.
22. Buehler R, Pucher J, Merom D, et al. Active travel in Germany and the US: Contributions of daily walking and cycling to physical activity. Am J Prev Med 2011; 41:241–250.
23. Pucher J, Buehler R, Merom D, et al. Walking and cycling in the United States, 2001–2009: Evidence from the National Household Travel Surveys. Am J Public Health 2011; 101(Suppl 1):S310–S317.
24. Yang Y, Diez-Roux AV. Walking distance by trip purpose and population subgroups. Am J Prev Med 2012; 43:11–19.
25. Eberhart MG, Share AM, Shpaner M, et al. Comparison of geographic methods to assess travel patterns of persons diagnosed with HIV in Philadelphia: How close is close enough? J Biomed Inform 2015; 53:93–99.
26. Luft HS, Robinson JC, Garnick DW, et al. The role of specialized clinical services in competition among hospitals. Inquiry 1986; 83–94.
27. Centers for Medicare, Medicaid Services H. Medicare program; Medicare prescription drug discount card. Interim final rule with comment period. Fed Regist 2003; 68:69839.
28. Pebesma E. Simple features for R: Standardized support for spatial vector data. The R Journal 2018; 10:1439–446.
29. Lu B, Sun H, Harris P, et al. Shp2graph: Tools to convert a spatial network into an Igraph graph in R. ISPRS Int J Geoinf 2018; 7:293.
30. Csardi G, Nepusz T. The igraph software package for complex network research. InterJournal, Complex Systems 2006; 1695:1–9.
For further references, please see “Supplemental References,” http://links.lww.com/OLQ/A377.
Supplemental Digital Content
© Copyright 2019 American Sexually Transmitted Diseases Association