Skip to main content
NCBI home page
AIDS Patient Care and STDs logoLink to AIDS Patient Care and STDs
. 2014 Apr 1;28(4):188–197. doi: 10.1089/apc.2013.0362

Late HIV Diagnosis: Differences by Rural/Urban Residence, Florida, 2007–2011

Mary Jo Trepka 1,,2,✉, Kristopher P Fennie 1, Diana M Sheehan 1,,2, Khaleeq Lutfi 1, Lorene Maddox 3, Spencer Lieb 4
PMCID: PMC3985529  PMID: 24660767

Abstract

The purpose of this retrospective cohort study was to identify individual-level demographic and community-level socioeconomic and health care resource factors associated with late diagnosis of HIV in rural and urban areas of Florida. Multilevel modeling was conducted with linked 2007–2011 Florida HIV surveillance, American Community Survey, Area Health Resource File, and state counseling and testing data. Late diagnosis (defined as AIDS diagnosis within 3 months of HIV diagnosis) was more common in rural than urban areas (35.8% vs. 27.4%) (p<0.0001). This difference persisted after controlling for age, sex, race/ethnicity, HIV transmission mode, country of birth, and diagnosis year (adjusted OR 1.39; 95% CI 1.17–1.66). In rural areas, older age and male sex were associated with late HIV diagnosis; zip code-level socioeconomic and county level health care resource variables were not associated with late diagnosis in rural areas. In urban areas only, Hispanic and non-Hispanic black race/ethnicity, foreign birth, and heterosexual mode of transmission were additionally associated with late HIV diagnosis. These findings suggest that, in rural areas, enhanced efforts are needed to target older individuals and men in screening programs and that studies of psychosocial and structural barriers to HIV testing in rural and urban areas be pursued.

Introduction

Early diagnosis and treatment of human immunodeficiency virus (HIV) infection confers multiple benefits on the health of individuals as well as population health. People living with HIV infection who receive care and treatment have decreased morbidity and improved survival.1,2 The general population benefits because antiretroviral treatment can decrease a person's viral load and thus decrease the probability of transmission.3–5 Increasing the percentage of people living with HIV infection who are treated with antiretroviral therapy was associated with a decrease in community viral load and a decrease in the incidence of HIV in British Columbia and San Francisco.6,7 Additionally, it was estimated that for 2008, half of the HIV transmissions in the United States were from people with undiagnosed HIV infection.8 Studies indicate that awareness of HIV infection and counseling about transmission prevention lead to safer sexual practices among people living with HIV infection.9 There may also be long-term cost savings associated with early diagnosis and treatment because people who present late for care have higher long-term direct medical costs.10

Despite the benefits of early diagnosis and treatment and substantial nationwide efforts to promote HIV counseling and testing, an estimated 18% of Americans infected with HIV were unaware of their infection as of 2009.11 Data from the 2005–2009 Behavioral Risk Factor Surveillance System indicate that people living in rural areas, particularly remote rural areas, were less likely to be tested for HIV within the past year than those living in central urban areas, and that these differences persisted after controlling for demographic and self-reported risk factors.12 A study of people diagnosed with HIV in South Carolina during 2001–2005 found that rural residence was associated with late HIV diagnosis,13 and a study of Veterans Administration patients found that rural residence was associated with late HIV care.14 We identified no studies in either rural or urban settings that examined the role of the availability of public HIV counseling and testing sites or other health care resources in late HIV diagnosis. The objective of this retrospective cohort study was to examine the role of individual demographic and community-level socioeconomic and health care resource factors in late diagnosis of HIV [defined as diagnosis of acquired immunodeficiency syndrome (AIDS) within 3 months of HIV diagnosis] in rural and urban areas of Florida.

Methods

Study sample

De-identified records of Florida residents aged 13 years or older who were reported with HIV infection that met the Centers for Disease Control and Prevention (CDC) case definition15 for the first time during 2007–2011 were obtained from the Florida Department of Health (DOH) Enhanced HIV/AIDS Reporting System (eHARS). The majority of data in eHARS come from health care provider reports, laboratory reports, and data abstracted from medical records by county health department staff. Records of diagnosed HIV cases with missing or invalid residential zip codes and records with HIV diagnoses made in a correctional facility were excluded from the analysis.

Individual-level characteristics

Individual-level variables available in the eHARS dataset for cases included residential zip code and county at time of HIV diagnosis and AIDS diagnosis (if applicable); month and year of HIV diagnosis and AIDS diagnosis (if applicable); country of birth; age at HIV diagnosis; sex; race/ethnicity; HIV transmission mode; and whether the diagnosis was at a correctional facility. During 2007–2011, a case met the HIV case definition if the person's medical record or laboratory report indicated evidence of a confirmed positive HIV test or a detectable viral load; the AIDS case definition was met if the person's medical record indicated the development of an AIDS-defining illness, a CD4 lymphocyte count <200 cells/μL, or CD4% of total lymphocytes <14.16,17 People were classified as being born in the United States (US) if they were born in any of the 50 states, District of Columbia, or any US dependency. People with a reported mode of transmission of men who have sex with men (MSM) combined with injection drug use (IDU) were grouped with people whose reported mode of transmission was IDU only. Follow-up AIDS diagnosis data were available through December 2012, allowing for at least 1 year of follow up for all records.

Community-level characteristics

Five-year estimates of zip code-level socioeconomic data from the 2007–2011 American Community Survey were obtained from the US Census Bureau18 and linked using the zip code tabulation area (ZCTA).19 The US Census Bureau reports data by ZCTA. A ZCTA approximates a zip code and is built by aggregating US Census Bureau blocks based on the zip code of addresses in these blocks. Zip code-level poverty was measured using the 5-year estimate of the percentage of population living below the poverty line in that ZCTA. Unemployment was measured as the 5-year estimate of the percentage of people aged 16 and older in a ZCTA who were unemployed. Educational attainment was measured as the 5-year estimate of the percentage of the population aged 18 and older who were high school graduates in that ZCTA. Florida ZCTAs were divided into rural and urban areas using the zip code approximation from Version 2.0 of Rural-Urban Commuting Area (RUCA) data codes developed at the University of Washington WWAMI Rural Research Center.20,21 There are several ways to categorize the codes into rural/urban areas. Categorization C was used because it groups zip codes for isolated small rural towns, other small rural towns, and large rural city/towns into one rural category.21 This was chosen because of the relatively small number of rural HIV cases in the isolated small rural town group (n=30, 0.1% of all cases) and the other small rural town group (n=213, 0.8% of all cases).

County level health care resources were obtained from the 2012–2013 Area Health Resource File.22 The Area Health Resource File is a dataset of county-level health care resource information that is assembled by the Health Resources and Services Administration from 50 different sources. From these data, we calculated the county-specific average number of short-term general hospitals per 100 square miles during 2006–2010, the number of hospitals with HIV/AIDS services per 100 square miles during 2010, and the average number of actively practicing medical doctors and doctors of osteopathy per 100 square miles during 2010–2011. The number of hospitals and physicians per square mile were considered because studies indicate that transportation tends to be a barrier to obtaining health care among people living with HIV infection.23,24 The distance between the health care provider and any given person's residence was not available, so the density of providers and hospitals was used as a proxy for travel distance as a barrier. The county-specific average number of HIV tests at publically funded counseling and testing sites during 2007–2011 was obtained from the Florida DOH,25 as was the location of publically funded sites (Florida DOH, personal communication, August 15, 2013). If there were multiple test sites at a given physical location (e.g., both an anonymous HIV test site and a STD clinic test site located in a given health department building), the location was counted as one physical site in the calculation of density of counseling and testing sites (number of counseling and testing sites per 100 square miles). The distribution of values for ZCTA- and county-level factors was categorized into quartiles.

Late HIV diagnosis was defined as an AIDS diagnosis within 3 months of an HIV diagnosis. Other studies have variously defined a late diagnosis as a time period of 1, 3, or 12 months between HIV and AIDS diagnosis.26 A 3-month time frame was chosen for this study based on a recommendation of the National HIV/AIDS Strategy to have people linked to care within 3 months27 and a recommendation stemming from a population-based study similar to the current one that used surveillance data in another state.28

Analytic plan

Associations between late HIV diagnosis and individual demographic characteristics, ZCTA-level socioeconomic, and county-level health care access factors were tested using the chi-square test for categorical variables and two-sample Wilcoxon Rank Sum tests for continuous variables. We considered any risk factor associated with screening in the bivariate analysis based on an α=0.25 as a candidate for inclusion in the multivariable logistic regression model.29 Significant independent variables were included in multilevel logistic models, with the dependent variable being late diagnosis. We arrived at a parsimonious model through backward selection (of demographic socioeconomic, and health care access variables), removing variables that did not significantly change the fit of the model based on the log likelihood ratio test.29 SAS Proc GLIMMIX was used for multilevel modeling. The model included three levels: individual, ZCTA, and county. In the model, the ZCTA was nested within the county. Clustering within ZCTA and county were accounted for through random intercepts. We conducted all analyses using SAS 9.2 (SAS Institute SAS Software, version 9.2. Cary, NC, 2002). The institutional review boards of Florida International University and the Florida Department of Health approved the study protocol.

Results

There were 27,424 records with a new HIV diagnosis among people 13 years of age or older between 2007 and 2011. Of these, 991 (3.6%) had an invalid zip code (post office box, nonexistent zip code, or no match with a ZCTA), and 848 (3.1%) were diagnosed in a correctional facility. Of the remaining 25,585 cases, 746 (2.9%) resided in a rural ZCTA at the time of diagnosis, and 24,839 (97.1%) resided in an urban ZCTA. Overall, 27.6% had a late HIV diagnosis. Those residing in a rural area were more likely to have a late HIV diagnosis (35.8%) than those in urban areas (27.4%) (p<0.0001) (Table 1). After controlling for individual level factors of age, sex, race/ethnicity, HIV transmission mode, country of birth, and diagnosis year, HIV-positive individuals residing in rural areas had 1.39 greater odds of a late HIV diagnosis [95% confidence interval (CI) 1.17–1.66, p<0.0002].

Table 1.

Characteristics of People with Late HIV Diagnosis (AIDS Diagnosis Within 3 Months of HIV Diagnosis) vs. Those Without Late Diagnosis (No AIDS Diagnosis within 3 Months after HIV Diagnosis), Stratified by Urban vs. Rural Area of Residence, Florida, 2007–2011 (n=25,585)

  Rural Urban
Characteristic Total, n AIDS diagnosis within 3 months of HIV diagnosis), n (%) No AIDS diagnosis within 3 months of HIV diagnosis), n (%) pValue Total, n AIDS diagnosis within 3 months of HIV diagnosis), n (%) No AIDS diagnosis within 3 months of HIV diagnosis), n (%) pValue
Totala 746 267 (35.8) 479 (64.2)   24,839 6,793 (27.4) 18,046 (72.7)  
Year of HIV diagnosis       0.11       0.08
 2007 181 53 (29.3) 128 (70.7)   5,740 1,496 (26.1) 4,244 (73.9)  
 2008 160 57 (35.6) 103 (64.4)   5,467 1,539 (28.2) 3,928 (71.9)  
 2009 138 55 (39.9) 83 (60.1)   4,832 1,363 (28.2) 3,469 (71.8)  
 2010 149 63 (42.3) 86 (57.7)   4,371 1,188 (27.2) 3,183 (72.8)  
 2011 118 39 (33.1) 79 (67.0)   4,429 1,207 (27.3) 3,222 (72.8)  
Sex at birth       <0.0001       0.80
 Male 499 204 (40.9) 295 (59.1)   18,034 4,924 (27.3) 13,110 (72.7)  
 Female 247 63 (25.5) 184 (74.5)   6,805 1,869 (27.5) 4,936 (72.5)  
Race/ethnicity       0.36       <0.0001
 Hispanics 88 38 (43.2) 50 (56.8)   5,442 1,433 (26.3) 4,009 (73.7)  
 Non-Hispanic blacks 370 131 (35.4) 239 (64.6)   11,923 3,424 (28.7) 8,499 (71.3)  
 Non-Hispanic whites 282 95 (33.7) 187 (66.3)   6,985 1,785 (25.6) 5,200 (74.5)  
 Other/unknown 6 3 (50.0) 3 (50.0)   489 151 (30.9) 338 (69.1)  
Age group at diagnosis       <0.0001       <0.0001
 13–19 years 41 5 (12.2) 36 (87.8)   1,031 111 (10.8) 920 (89.2)  
 20–39 years 317 80 (25.2) 237 (74.8)   11,548 2,414 (20.9) 9,134 (79.1)  
 40–59 years 336 161 (47.9) 175 (52.1)   10,767 3,663 (34.0) 7,104 (66.0)  
 60 years or older 52 21 (40.4) 31 (59.6)   1,493 605 (40.5) 888 (59.5)  
Country of birth       0.64       <0.0001
 United States 632 224 (35.4) 408 (64.6)   17,450 4,600 (26.4) 12,850 (73.6)  
 Not United States 114 43 (37.7) 71 (62.3)   7,389 2,193 (29.7) 5,196 (70.3)  
Mode of transmission       0.11       <0.0001
 Men who have sex with men (MSM) 277 98 (35.4) 179 (64.6)   11,548 2,611 (22.6) 8,937 (77.4)  
 Injection drug use (IDU) & MSM/IDU 71 26 (36.6) 45 (63.4)   1,631 481 (29.5) 1,150 (70.5)  
 Heterosexual 293 95 (32.4) 198 (67.6)   8,613 2,684 (31.2) 5,929 (68.8)  
 Other/unknown 105 48 (45.7) 57 (54.3)   3,047 1,017 (33.4) 2,030 (66.6)  
Zip code tabulation area level variables
Percent of population below poverty line (average 2007–2011), quartiles       0.88       0.241
 <8.7 21 6 (28.6) 15 (71.4)   2,087 553 (26.5) 1,534 (73.5)  
 8.7–12.9 163 61 (37.4) 102 (62.6)   4,694 1,291 (27.5) 3,403 (72.5)  
 13.0–19.3 189 67 (35.5) 122 (64.6)   7,320 1,952 (26.7) 5,368 (73.3)  
 ≥19.4 373 133 (35.7) 240 (64.3)   10,738 2,997 (27.9) 7,741 (72.1)  
Percent of population 16 and older who is unemployed (average 2007–2011), quartiles       0.95       <0.0001
 <4.2 199 71 (35.7) 128 (64.3)   2,707 652 (24.1) 2,055 (75.9)  
 4.2–5.5 166 58 (34.9) 108 (65.1)   3,889 1,012 (26.0) 2,877 (74.0)  
 5.6–7.2 239 89 (37.2) 150 (62.8)   5,719 1,583 (27.7) 4,136 (72.3)  
 ≥7.3 142 49 (34.5) 93 (65.5)   12,524 3,546 (28.3) 8,978 (71.7)  
Percent of population 18 years and older that is a high school graduate (average 2007–2011), quartiles       0.09       <0.0001
 ≥92.1 25 9 (36.0) 16 (64.0)   3,017 723 (24.0) 2,294 (76.0)  
 86.9–92.0 147 56 (38.1) 91 (61.9)   4,672 1,326 (28.4) 3,346 (71.6)  
 80.4–86.8 185 52 (28.1) 133 (71.9)   7,586 2,050 (27.0) 5,536 (73.0)  
 <80.4 389 150 (38.6) 239 (61.4)   9,564 2,694 (28.2) 6,870 (71.8)  
County level variables
Number of hospitals with HIV/AIDS services during 2010       0.74       <0.0001
 0 600 213 (35.5) 387 (64.5)   3,041 950 (31.2) 2,091 (68.8)  
 ≥1 146 54 (37.0) 92 (63.0)   21,798 5,843 (26.8) 15,955 (73.2)  
Number of hospitals per 100 square miles (average 2006–2010), quartiles       0.02       <0.0001
 <0.1374 112 32 (28.6) 80 (71.4)   548 173 (31.6) 375 (68.4)  
 0.1374–0.1966 280 89 (31.8) 191 (68.2)   538 168 (31.2) 370 (68.8)  
 0.1967–0.4104 300 126 (42.0) 174 (58.0)   4,286 1,270 (29.6) 3,016 (70.4)  
 ≥0.4105 54 20 (37.0) 34 (63.0)   19,467 5,182 (26.6) 14,285 (73.4)  
Number of doctors per 100 square miles (average 2010–2011)       0.23       <0.0001
 <2.128 123 34 (27.6) 89 (72.4)   26 7 (26.9) 19 (73.1)  
 2.128–23.378 495 185 (37.4) 310 (62.6)   133 42 (31.6) 91 (68.4)  
 23.379–92.539 87 33 (37.9) 54 (62.1)   2,898 991 (34.2) 1,907 (65.8)  
 ≥92.540 41 15 (36.6) 26 (63.4)   21,782 5,753 (26.4) 16,029 (73.6)  
Number of HIV tests per 1000 pop (average 2007–2011)       0.14       <0.0001
 <14.021 125 50 (40.0) 75 (60.0)   1,609 525 (32.6) 1,084 (67.4)  
 14.021–19.303 220 65 (29.6) 155 (70.5)   1,635 493 (30.2) 1,142 (69.9)  
 19.304–24.800 165 62 (37.6) 103 (62.4)   12,613 3,336 (26.5) 9,277 (73.6)  
 ≥24.801 236 90 (38.1) 146 (61.9)   8,982 2,439 (27.2) 6,543 (72.9)  
Number of CT sites per 100 square miles (2012)       0.32       <0.0001
 <0.1779 198 61 (30.8) 137 (69.2)   277 96 (34.7) 181 (65.3)  
 0.1779–0.3416 244 89 (36.5) 155 (63.5)   746 228 (30.6) 518 (69.4)  
 0.3417–0.7555 255 100 (39.2) 155 (60.8)   1,956 683 (34.9) 1,273 (65.1)  
 ≥0.7556 49 17 (34.7) 32 (65.3)   21,860 5,786 (26.5) 16,074 (73.5)  
Open in a new tab
a

The p-value for rural-urban difference was (p<0.0001).

To identify differences in characteristics associated with late diagnosis, the subsequent analyses were stratified according to rural or urban area. In the bivariate analysis (Table 1), in rural areas, people with a late HIV diagnosis compared to those whose diagnosis was not late were more likely to be male, 40 years of age or older, and live in counties with a higher hospital density. In urban areas, those with a late HIV diagnosis were more likely to be non-Hispanic black, other/unknown race/ethnicity, be 40 years of age or older, not born in the US, and have the HIV transmission modes of heterosexual sex or other/unknown. In urban areas, the ZCTA-level factors of unemployment and high school graduation rate were associated with late diagnosis. Also in urban areas, the county-level factors of no hospitals with HIV/AIDS services, low hospital density, low physician density, low HIV counseling and testing rates, and low HIV counseling and testing site density were associated with late HIV diagnosis.

In the multilevel logistic regression model in rural areas, male sex, being 40–59 years of age vs. the referent group (13–19 years) and being 60 or older were significantly associated with a late HIV diagnosis [adjusted odds ratio (aOR) 1.85, 95% CI 1.29–2.65; aOR 7.47, 95% CI 2.78–20.04; and aOR 5.20, 95% CI: 1.70–15.85, respectively] (Table 2). In urban areas, eight variables met the inclusion criterion, and all but year of HIV diagnosis were associated with late HIV diagnosis, including male sex (aOR 1.36, 95% CI: 1.25–1.47), being age 60 or older (aOR 5.26, 95% CI: 4.20–6.60), being 40–59 years of age (aOR 4.25, 95% CI: 3.47–5.21), being 20–39 years of age (aOR 2.23, 95% CI: 1.82–2.74), being Hispanic (aOR 1.14, 95% CI: 1.04–1.25), non-Hispanic black (aOR 1.17, 95% CI: 1.08–1.27), or other/unknown race/ethnicity (aOR 1.33, 95% CI: 1.08–1.64), not born in the US (1.14, 95% CI: 1.07–1.23), living in a ZCTA with the lowest quartile high school graduation rates vs. highest (the referent group) (aOR 1.17, 95% CI: 1.04–1.31), second lowest quartile (aOR 1.14, 95% CI: 1.02–1.28), third lowest quartile (aOR 1.21, 95% CI: 1.08–1.36) or in a county with the second highest quartile for doctor density compared to the highest (aOR 1.30, 95% CI: 1.13–1.50). There were, however, only 159 cases (0.6% of all urban cases) in the two lowest quartiles of doctor density. Relative to the mode of transmission of heterosexual sex, there was a lower adjusted odds of late HIV diagnosis among those with a transmission mode of MSM (0.65, 95% CI: 0.59–0.70) and transmission mode of injection drug use IDU (0.83, 95% CI: 0.73–0.93).

Table 2.

Adjusted Odds Ratios and 95% Confidence Intervals for Late HIV Diagnosis (AIDS Diagnosis Within 3 Months of HIV Diagnosis) Among People Reported with HIV Infection, by Rural/Urban Residential Status, 2007–2011, Florida

Characteristic Rural (n=746) Adjusted odds ratioa(95% confidence interval) Urban (n=24,839) Adjusted odds ratioa (95% confidence interval)
Year of HIV diagnosis
 2007 0.82 (0.49–1.40) 0.90 (0.83–0.99)
 2008 1.11 (0.65–1.89) 1.01 (0.92–1.11)
 2009 1.53 (0.88–2.65) 1.03 (0.94–1.13)
 2010 1.59 (0.93–2.71) 1.00 (0.91–1.10)
 2011 Ref. Ref.
Sex
 Male 1.85 (1.29–2.65) 1.36 (1.25–1.47)
 Female Ref. Ref.
Age group at diagnosis
 13–19 years Ref. Ref.
 20–39 years 2.65 (0.99–7.15) 2.23 (1.82–2.74)
 40–59 years 7.47 (2.78–20.04) 4.25 (3.47–5.21)
 60 years or older 5.20 (1.70–15.85) 5.26 (4.20–6.60)
Race/ethnicity
 Hispanic b 1.14 (1.04–1.25)
 Non-Hispanic black   1.17 (1.08–1.27)
 Other   1.33 (1.08–1.64)
 Non-Hispanic white   Ref.
Country of birth
 United States b Ref.
 Not United States   1.14 (1.07–1.23)
Mode of HIV transmission
 Men who have sex with men (MSM) b 0.65 (0.59–0.70)
 Injection drug use (IDU) & MSM/IDU   0.83 (0.73–0.93)
 Heterosexual   Ref.
 Other/unknown   1.00 (0.91–1.10)
Percent of population in ZCTA 18 years and older that is a high school graduate (average 2007–2011), quartiles    
 <80.4 b 1.17 (1.04–1.31)
 80.4–86.8   1.14 (1.02–1.28)
 86.9–92.0   1.21 (1.08–1.36)
 ≥92.1   Ref.
Number of doctors in county per 100 square miles (average 2010–2011)
 <2.128 b 1.02 (0.41–2.56)
 2.128–23.378   1.34 (0.88–2.05)
 23.379–92.539   1.30 (1.13–1.50)
 ≥92.540   Ref.
Open in a new tab
a

Odds ratios adjusted for all variables in column. bVariable not shown because α>0.25 in bivariate analysis or not retained in the final backwards selection model.

Discussion

In this study, we found that late HIV diagnosis occurred in about a quarter of the cases and was more common in rural areas than in urban areas. Furthermore, factors associated with late HIV diagnosis differed between rural and urban areas. In rural areas, none of the ZCTA-level socioeconomic factors or county-level physician, hospital, or public counseling and testing factors were associated with delayed diagnosis; while in urban areas, low ZCTA-level high school graduation and physician density were associated with late diagnosis. The individual-level characteristics of male sex and older age were associated with late diagnosis in both rural and urban areas. However, only in urban areas were the individual-level characteristics Hispanic and non-Hispanic black race/ethnicity, being foreign born, and heterosexual mode of HIV transmission associated with late diagnosis. While our study confirms the findings on the overall percentage of late diagnoses, the rural/urban differences in the percentage of late diagnoses and factors associated with late diagnosis are noteworthy.

With respect to the proportion of late diagnosis, the finding that 27.6% of all reported HIV diagnoses were late is similar to the 25.5% reported from a 2010 CDC study that used the same definition of late diagnosis (AIDS diagnosis within 3 months of HIV diagnosis) in 14 jurisdictions (12 states, the District of Columbia, and San Francisco).30 In additional analyses using the 12-month definition of late diagnosis (diagnosis of AIDS within 12 months of HIV diagnosis), we found that 32.7% of diagnoses were late, which was very similar to what was estimated for the US (32.3%) and for Florida (30.1%).31 Using the 1-month definition, 21.8% were diagnosed late, which is similar to the 23.6% reported in a population-based study in the Bronx after a major enhancement of HIV testing.32 It should be noted that not all of these observed diagnosis delays may be preventable. An estimated 3.6–13.0% of HIV cases that were diagnosed with AIDS within 1 year in New York City were cases of accelerated progression of HIV, as opposed to being cases of late HIV diagnosis.33

The finding that rural residence was associated with late diagnosis is consistent with data from a national representative sample that indicated rural residents are less likely to have ever been tested for HIV.12 This finding is also consistent with an earlier (2001–2005) population-based study in South Carolina that found rural residence was associated with late diagnosis13 and a study of veterans which found a later entry into care for rural residents relative to urban residents.14 Rural residents have also been less likely to receive other types of screening, such as breast and colorectal cancer screening, suggesting that this pattern is not unique to HIV care.34,35 The statistically significant rural-urban difference was also seen when using the 1- and 12-month definitions of late diagnosis in our study. The rural resident diagnosis disadvantage was somewhat unexpected, given the results of an earlier study in Florida finding no survival disadvantage between people in rural and urban areas diagnosed with AIDS between 1993–2007.36 However, that study used an earlier time period and examined survival from the time of AIDS and not the time of HIV diagnosis. Furthermore, survival is affected by care and treatment in addition to early diagnosis.

Proposed reasons for why rural residents may be more likely to be diagnosed late are less perceived risk, more stigma, fear of loss of confidentiality, lower education, poverty, and less access to screening services and health care.13,24,37–39 Even if counseling and testing sites are close to a rural resident's home, rural residents may prefer to be tested in a different community to ensure anonymity, which can result in delayed testing. Rural providers have suggested that integrating HIV testing into other health services; rapid oral testing and community health education; outreach; and making HIV testing more convenient, accessible, and free are potential ways to decrease barriers.24

With respect to community-level factors, we found that neither the density of public HIV counseling and testing sites nor public HIV counseling and testing rates was associated with late diagnosis in rural or urban areas in the multivariate analyses. We also found that physician density, hospital density, or having a hospital with HIV/AIDS services in the county was not associated with late diagnosis in rural areas. This suggests that health care resource disparities are not the underlying reason for late HIV diagnosis in rural areas. However, our variables related to HIV counseling and testing sites, HIV counseling and testing rates, physicians, and hospitals were at the county level. There is likely significant heterogeneity with regard to where these resources are distributed within counties such that the density of counseling and testing sites may not reflect well what a given person has access to. For example, a county may have several public HIV counseling and testing sites, but if they are clustered in one region of the county and the person diagnosed with HIV resides in a different part of the county, that person may not have had access to those sites. Additionally, our HIV counseling and testing data were only for testing at public counseling and testing sites and do not include testing by private providers. In a separate analysis, we found that the public HIV counseling and testing rate was actually slightly higher in counties classified as rural using the state of Florida statute's definition of rural counties40 than in urban counties (median 20.4 tests vs. 18.0 tests per 1,000 population, respectively) (Wilcoxon Rank Sum P=0.09), suggesting that disparities in testing may be in private as opposed to public counseling and testing.

The individual-level factors that we found associated with late diagnosis are largely consistent with other studies. Late diagnosis has generally been found to be more common among men and older adults in multiple studies in the US, Australia, Europe, and Asia.13,28,41–51 Health care providers may be less likely to consider HIV infection in an older adult, and older adults may perceive themselves as at lower risk than younger adults.51 Rapid disease progression, which would appear as a late diagnosis, has been reported in older adults.44,47 The gender differences may be related to women having more frequent health care encounters, including family planning or antenatal care visits, that offer HIV screening.52

Late diagnosis was more common among non-Hispanic blacks, Hispanics, and those in the “other” race/ethnic group in urban but not in rural areas. Other population-based studies conducted during the last 10 years, found Hispanic ethnicity to be associated with late diagnosis.11,13,28,53 Non-Hispanic black race/ethnicity was associated with late testing in population-based studies in South Carolina13 and California,53 but not Harris County, Texas.28 Our findings of a lack of association in rural areas may be due to the relatively small number of cases in rural areas or a weaker/nonexistent association in rural areas. None of the above cited studies examined rural areas separately from urban areas. Therefore, the association of race/ethnicity with delayed diagnosis in rural areas should be examined in other states.

As with race/ethnicity, being born outside the US was associated with delayed diagnosis in urban but not rural areas, which may be due to the relatively small number of foreign-born people in rural areas. Studies in the US and in Europe have consistently found that foreign birth is associated with delayed diagnosis.41,42,53–59 Having an undocumented immigration status was associated with not ever being tested for HIV in a study in South Flordia.52 Reasons identified in other studies for foreign birth being associated with delayed diagnosis include stigma,56,60,61 language barriers,58,61 and less knowledge about HIV.56,61

Late HIV/AIDS diagnosis in urban areas was associated with heterosexual compared with MSM and IDU modes of HIV transmission. Late HIV/AIDS diagnosis has been associated with heterosexual transmission in several other studies in the US, Australia, and Europe.28,45,49,53,57,62,63 This may be due to people with high risk heterosexual activity not perceiving themselves to be at high risk of HIV.

There are several additional limitations that need to be noted for this study. First, there are relatively few rural HIV cases that may have limited our power to find factors associated with late diagnosis in rural areas. This is partially due to only 7.0% of Florida's population living in rural areas.40 Second, we may have incomplete ascertainment of AIDS diagnoses. The CD4 values were missing for 30.3% of cases. Although these people were probably less likely to have AIDS than the group as a whole because they presumably were not sick enough to seek care, some may have met the AIDS case definition within 3 months of their HIV diagnosis. A missing CD4 count or percent test was less common in rural areas (22.1%) than in urban areas (30.6%) (p<0.0001), so it is possible that some of the rural/urban difference was related to better identification of AIDS cases in rural areas. Third, because HIV infection reporting was not mandated in Florida prior to July 1997, and because reporting was not implemented retroactively, some cases might appear to be diagnosed with HIV at a more recent time than actually happened. This could bias the time to AIDS diagnosis or death by making it appear to be shorter in such cases. However, all cases in this investigation were diagnosed from 2007 to 2011. The likelihood that they were actually diagnosed more than 10–15 years prior to this study is probably small and unlikely to differ by rural/urban status. Fourth, we had no measures of individual reasons for not having an HIV test such as lack of perceived risk or fear of the test result. Rural/urban differences in these individual reasons should be further explored to find potential points for intervention in improving HIV testing rates. Fifth, although we found no association between community-level socioeconomic factors and late diagnosis in rural areas, we cannot conclude that individual level SES is not important in predicting HIV counseling and testing behavior or access to HIV counseling and testing because we had no measures of individual level socioeconomic status. Finally, the rural/urban categorization was at the ZCTA level, and the health care provider and counseling and testing variables were at the county level. It would have been preferable to have data on health care provider and counseling and testing at the ZCTA level too, but they were only available at the county level.

In summary, the findings of this study indicate that late HIV diagnosis is relatively common, particularly among older and among male rural residents. The gender difference was particularly large in rural areas and merits further investigation.

These findings suggest that in rural areas in Florida enhanced efforts be made to target older individuals and men in screening programs. Additional studies of delayed diagnosis are needed in rural areas of other states, particularly states with more small and isolated rural areas, to determine if similar characteristics are associated with delayed diagnosis. If possible, such studies should analyze actual travel distances from rural residence to closest counseling and testing facilities. In urban areas late HIV diagnosis also was associated with belonging to a racial/ethnic minority groups and being foreign born. Studies of psychosocial barriers and structural barriers to HIV counseling and testing in rural and urban areas in the groups more likely to have an HIV diagnosis should be pursued. Lastly, when planning, implementing, and evaluating HIV and AIDS prevention, screening, and care efforts, it is important to consider rural/urban differences.

Acknowledgments

The project described was supported by Award Numbers R01MD004002 and P20MD002288 from the National Institute on Minority Health and Health Disparities at the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Minority Health and Health Disparities or the National Institutes of Health.

Author Disclosure Statement

No competing financial interests exist.

References

  • 1.Centers for Disease Control and Prevention. Guidelines for prevention and treatment of opportunistic infections in HIV-infected adults and adolescents, MMWR Morb Mortal Wkly Recomm Rep 2009;58:1–207 Available at: http://www.cdc.gov/mmwr/pdf/rr/rr5804.pdf Accessed September18, 2013 [PubMed] [Google Scholar]
  • 2.Thompson MA, Aberg JA, Hoy JF, et al. Antiretroviral treatment of adult HIV infection. JAMA 2012;308:387–402 [DOI] [PubMed] [Google Scholar]
  • 3.Attia S, Egger M, Muller M, Zwahlen M, Low N. Sexual transmission of HIV according to viral load and antiretroviral therapy: Systematic review and meta-analysis. AIDS 2009;23:1397–1404 [DOI] [PubMed] [Google Scholar]
  • 4.Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med 2011;365:493–505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Donnell D, Baeten JM, Kiarie J, et al. Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: A prospective cohort analysis. Lancet 2010;375:2092–2098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Das M, Chu PL, Santos G-M, et al. Decreases in community viral load are accompanied by reductions in new HIV infections in San Francisco. PLoS One 2010;5:e11068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Montaner JS, Lima VD, Barrios R, et al. Association of highly active antiretroviral therapy coverage, population viral load, and yearly new HIV diagnoses in British Columbia, Canada: A population-based study. Lancet 2010;376:532–539 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hall HI, Holtgrave DR, Maulsby C. HIV transmission rates from persons living with HIV who are aware and unaware of their infection. AIDS 2012;26:893–896 [DOI] [PubMed] [Google Scholar]
  • 9.Crepaz N, Lyles CM, Wolitski RJ, et al. Do prevention interventions reduce HIV risk behaviours among people living with HIV? A meta-analytic review of controlled trials. AIDS 2006;20:143–157 [DOI] [PubMed] [Google Scholar]
  • 10.Krentz HB, Gill MJ. The direct medical costs of late presentation (<350/mm3) of HIV infection over a 15-year period. AIDS Res Treat 2012;2012:757135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hall HI, Frazier EL, Rhodes P, et al. Differences in human immunodeficiency virus care and treatment among subpopulations in the United States. JAMA Intern Med 2013;173:1337–1344 [DOI] [PubMed] [Google Scholar]
  • 12.Ohl ME, Perencevich E. Frequency of human immunodeficiency virus (HIV) testing in urban vs. rural areas of the United States: Results from a nationally-representative sample. BMC Public Health 2011;11:681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Weis KE, Liese AD, Hussey J, Gibson JJ, Duffus WA. Associations of rural residence with timing of HIV diagnosis and stage of disease at diagnosis, South Carolina 2001–2005. J Rural Health 2010;26:105–112 [DOI] [PubMed] [Google Scholar]
  • 14.Ohl M, Tate J, Duggal M, et al. Rural residence is associated with delayed care entry and increased mortality among veterans with human immunodeficiency virus infection. Med Care 2010;48:1064–1070 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Centers for Disease Control and Prevention. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome, MMWR Morb Mortal Wkly Recomm Rep 1999;48:1–28 Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/rr4813a1.htm Accessed September18, 2013 [PubMed] [Google Scholar]
  • 16.Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Recomm Rep 1992;41:1–19, December18, 1992. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/00018871.htm Accessed September18, 2013 [PubMed] [Google Scholar]
  • 17.Centers for Disease Control and Prevention. Revised surveillance case definitions for HIV infection among adults, adolescents, and children aged <18 months and for HIV infection and AIDS among children aged 18 months to <13 years—United States, 2008. MMWR Morb Mortal Wkly Recomm Rep 2008;57:1–8 Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5710a1.htm Accessed September18, 2013 [PubMed] [Google Scholar]
  • 18.US Census Bureau. American FactFinder. Washington, DC: Government Printing Office, No date. Available at: http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml Accessed August5, 2013
  • 19.US Census Bureau. ZIP™ Code Tabulation Areas (ZCTA™). Washington, DC: Government Printing Office, No date; Available at: http://www.census.gov/geo/reference/zctas.html Accessed September18, 2013 [Google Scholar]
  • 20.Hart LG, Larsen EH, Lishner DM. Rural definitions for health policy and research. Am J Public Health 2005;95:1149–1155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.WWAMI Rural Health Research Center. Rural Urban Commuting Areas (RUCA). No date. Available at http://http://depts.washington.edu/uwruca/ruca-data.php Accessed September18, 2013
  • 22.Health Resources and Services Administration Health Professions. Area Health Resource File, 2012–2013. Rockville, MD: US Department of Health and Human Services; Available at: http://arf.hrsa.gov/ Accessed June19, 2013 [Google Scholar]
  • 23.Schur CL, Berk ML, Dunbar JR, Shapiro ME, Cohn SE, Bozzette SA. Where to seek care: An examination of people in rural areas with HIV/AIDS. J Rural Health 2002;18:337–347 [DOI] [PubMed] [Google Scholar]
  • 24.Sutton M, Anthony M, Vila C, McLellan-Lemal E, Weidle PJ. HIV testing and HIV/AIDS treatment services in rural counties in 10 southern states: Service provider perspectives. J Rural Health 2010;26:240–247 [DOI] [PubMed] [Google Scholar]
  • 25.Florida Department of Health. 2012 Data from the Counseling and Testing Database. Tallahassee, FL: US Department of Health and Human Services; 2013. Available at http://www.doh.state.fl.us/disease_ctrl/aids/trends/ct/2012Data/2012data.html Accessed August5, 2013 [Google Scholar]
  • 26.Mukolo A, Villegas R, Aliyu M, Wallston KA. Predictors of late presentation for HIV diagnosis: A literature review and suggested way forward. AIDS behavior 2013;17:5–30 [DOI] [PubMed] [Google Scholar]
  • 27.Centers for Disease Control and Prevention. High-Impact HIV Prevention CDC's Approach to reducing HIV infections in the United States. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2011. Available at: http://www.cdc.gov/hiv/pdf/policies_NHPC_Booklet.pdf Accessed September23, 2013 [Google Scholar]
  • 28.Yang B, Chan SK, Mohammad N, et al. Late HIV diagnosis in Houston/Harris County, Texas, 2000–2007. AIDS Care 2010;22:766–774 [DOI] [PubMed] [Google Scholar]
  • 29.Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York, NY: John Wiley and Sons, Inc, 2000 [Google Scholar]
  • 30.Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 US dependent areas—2010. HIV Surveillance Report, 2013, Vol.18 (No. 2, part B). Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2013. Available at: http://www.cdc.gov/hiv/pdf/statistics_2010_HIV_Surveillance_Report_vol_18_no_2.pdf Accessed September18, 2013 [Google Scholar]
  • 31.Centers for Disease Control and Prevention. Vital Signs: HIV testing and diagnosis among adults—United States, 2001–2009, MMWR Morb Mortal Wkly Rep 2010;59(47)1550–1555 Available at: http://www.cdc.gov/mmwr/pdf/wk/mm5947.pdf Accessed October1, 2013 [PubMed] [Google Scholar]
  • 32.Myers JE, Braunstein SL, Shepard CW, et al. Assessing the impact of a community-wide HIV testing scale-up initiative in a major urban epidemic. J Acquir Immune Defic Syndr 2012;61:23–31 [DOI] [PubMed] [Google Scholar]
  • 33.Sabharwal CJ, Sepkowitz K, Mehta R, Shepard C, Bodach S, Torian L, Begier EM. Impact of accelerated progression to AIDS on public health monitoring of late HIV diagnosis. AIDS Patient Care STDS 2011;25:143–151 [DOI] [PubMed] [Google Scholar]
  • 34.Coughlin SS, Thompson TD, Colorectal cancer screening practices among men and women in rural and nonrural areas of the United States, 1999. J Rural Health 2004;20:118–124 [DOI] [PubMed] [Google Scholar]
  • 35.Doescher MP, Jackson JE. Trends in cervical and breast cancer screening practices among women in rural and urban areas of the United States J Public Health Manag Practice 2009;15:200–209 [DOI] [PubMed] [Google Scholar]
  • 36.Trepka MJ, Niyonsenga T, Maddox LM, Lieb S. Rural AIDS diagnoses in Florida: Changing demographics and factors associated with survival. J Rural Health 2013;29:266–280 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Crosby RA, Yarber Wl, DiClemente RJ, Wingood G, Meyerson B. HIV-associated histories, perceptions, and practices among low-income African American women: Does rural residence matter? Am J Public Health 2003;92:4:655–659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Sison N, Yolken A, Poceta J, Mena L, Chan PA, Barnes A, Smith E, Nunn A. Healthcare provider attitudes, practices, and recommendations for enhancing routine HIV testing and linkage to care in the Mississippi Delta Region. AIDS Patient Care STDS 2013;27:511–517 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Zukoski AP, Thorburn S. Experiences of stigma and discrimination among adults living with HIV in a low HIV-prevalence context: A qualitative analysis. AIDS Patient Care STDS 2009;23:267–276 [DOI] [PubMed] [Google Scholar]
  • 40.Florida Department of Health Office of Rural Health. Rural Health. n.d. Available at: http://www.doh.state.fl.us/workforce/ruralhealth/ruralhealthhome.html Accessed October14, 2013
  • 41.Adler A, Mounier-Jack S, Coker RJ. Late diagnosis of HIV in Europe: Definitional and public health challenges. AIDS Care 2009;21:284–293 [DOI] [PubMed] [Google Scholar]
  • 42.Castilla J, Sobrino P, de la Fuente L, et al. Late diagnosis of HIV infection in the era of highly active antiretroviral therapy: Consequences for AIDS incidence. AIDS 2002;16:1945–1951 [DOI] [PubMed] [Google Scholar]
  • 43.Chin T, Hicks C, Samsa G, McKellar M. Diagnosing HIV infection in primary care settings: Missed opportunities. AIDS Patient Care STDs 2013;27:392–397 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Hall HI, McDavid K, Ling Q, Sloggett A. Determinants of progression to AIDS or death after HIV diagnosis, United States, 1996 to 2001. Ann Epidemiol 2006;16:824–833 [DOI] [PubMed] [Google Scholar]
  • 45.Lemoh C, Guy R, Yohannes K, et al. Delayed diagnosis of HIV infection in Victoria 1994 to 2006. Sex Health 2009;6:117–122 [DOI] [PubMed] [Google Scholar]
  • 46.Mojumdar K, Vajpayee M, Chauhan NK, Mendiratta S. Late presenters to HIV care and treatment, identification of associated risk factors in HIV-1 infected Indian population. BMC Public Health 2010;10:416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Smith RD, Delpech VC, Brown AE, Rice BD. HIV transmission and high rates of late diagnoses among adults aged 50 years and over. AIDS 2010;24:2109–2115 [DOI] [PubMed] [Google Scholar]
  • 48.Thanawuth N, Chongsuvivatwong V. Late HIV diagnosis and delay in CD4 count measurement among HIV-infected patients in southern Thailand. AIDS Care 2008;20:43–50 [DOI] [PubMed] [Google Scholar]
  • 49.Wolbers M, Bucher HC, Furrer H, et al. Delayed diagnosis of HIV infection and late initiation of antiretroviral therapy in the Swiss HIV Cohort Study. HIV Med 2008;9:397–405 [DOI] [PubMed] [Google Scholar]
  • 50.Wong K, Lee S, Low K, Wan W. Temporal trend and factors associated with late HIV diagnosis in Hong Kong, a low HIV prevalence locality. AIDS Patient Care STDS 2003;17:461–469 [DOI] [PubMed] [Google Scholar]
  • 51.Grabar S, Weiss L, Costagliola D. HIV infection in older patients in the HAART era. J Antimicrob Chemother 2006;57:4–7 [DOI] [PubMed] [Google Scholar]
  • 52.Fernandez MI, Collazo JB, Bowen GS, et al. Predictors of HIV testing and intention to test among Hispanic farmworkers in south Florida. J Rural Health 2005;21:56–64 [DOI] [PubMed] [Google Scholar]
  • 53.Tang JJ, Levy V, Hernandez MT. Who are California's late HIV testers? An analysis of state AIDS surveillance data, 2000–2006. Public Health Rep 2011;126:338–343 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Crawford T, Caldwell G, Bush HM, Browning S, Thornton A. Foreign born status and HIV/AIDS: A comparative analysis of HIV/AIDS characteristics among foreign and U.S. born individuals. J Immigr Minor Health 2012;14:82–88 [DOI] [PubMed] [Google Scholar]
  • 55.Espinoza L, Hall I, Selik RM, Hu X. Characteristics of HIV infection among Hispanics, United States 2003–2006. J Acquir Immune Defic Syndr 2008;49:94–101 [DOI] [PubMed] [Google Scholar]
  • 56.Levy V, Prentiss D, Balmas G, et al. Factors in the delayed HIV presentation of immigrants in northern California: Implications for voluntary counseling and testing programs. J Immigr Minor Health 2007;9:49–54 [DOI] [PubMed] [Google Scholar]
  • 57.Schwarcz S, Hsu L, Dilley JW, et al. Late diagnosis of HIV infection: Trends, prevalence, and characteristics of persons whose HIV diagnosis occurred within 12 months of developing AIDS. J Acquir Immune Defic Syndr 2006;43:491–494 [DOI] [PubMed] [Google Scholar]
  • 58.Wohl AR, Tejero J, Frye DM. Factors associated with late HIV testing for Latinos diagnosed with AIDS in Los Angeles. AIDS Care 2009;21:1203–1210 [DOI] [PubMed] [Google Scholar]
  • 59.Yazdanpanah Y, Lange J, Cairns G. Earlier testing for HIV—How do we prevent late presentation? Antivir Ther 2010;15:17–24 [DOI] [PubMed] [Google Scholar]
  • 60.Ojikutu B, Nnaji C, Sithole J, Schneider KL, Higgins-Biddle M, Cranston K, Earls F. All black people are not alike: Differences in HIV testing patterns, knowledge, and experience of stigma between U.S.-born and non-U.S.-born blacks in Massachusetts. AIDS Patient Care STDS 2013;27:45–54 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Dang BN, Giordano TP, Kim JH. Sociocultural and structural barriers to care among undocumented Latino immigrants with HIV infection. J Immigr Minor Health 2012;14:124–131 [DOI] [PubMed] [Google Scholar]
  • 62.Longo B, Pezzotti P, Boros S, Urciuoli R, Rezza G. Increasing proportion of late testers among AIDS cases in Italy, 1996–2002. AIDS Care 2005;17:834–841 [DOI] [PubMed] [Google Scholar]
  • 63.Manavi K, McMillan A, Ogilvie M, Scott G. Heterosexual men and women with HIV test positive at a later stage of infection than homo- or bisexual men. Int J STD AIDS 2004;15:811–814 [DOI] [PubMed] [Google Scholar]

Articles from AIDS Patient Care and STDs are provided here courtesy of Mary Ann Liebert, Inc.

RESOURCES