Abstract
The study's objective was to identify factors associated with differences in the rate of viral suppression among minority women with HIV/AIDS in care in the Miami-Dade County Ryan White Program (RWP). A retrospective cohort study was conducted using social characteristics and laboratory data of minority women enrolled in the Miami-Dade County RWP in 2017. Viral suppression was defined as <200 copies/mL using the last viral load test of 2017. Multilevel logistic regression models were used to estimate adjusted odds ratio (aOR) and 95% confidence intervals (CIs). Of the 1,550 racial/ethnic minority women in the study population, 43.1% were African American, 31.3% were Hispanic, and 25.6% were Haitian. The proportion of women virally suppressed was lower among African Americans (80.8%) than among Hispanics (86.4%) and Haitians (85.1%). Viral suppression rates were significantly lower among women aged 18–34 years (aOR: 0.41, CI: 0.27–0.64) and 35–49 years (0.63, 0.45–0.90) vs. ≥50 years, born in the United States (0.48, 0.30–0.78), having a household income of <100% the federal poverty level (0.54, 0.30–0.95), previously diagnosed with AIDS (0.60, 0.44–0.81), reporting problematic drug use (0.23, 0.08–0.69), and living in a residentially unstable neighborhood (0.77, 0.64–0.93). Race/ethnicity was not associated with viral suppression after adjusting for other factors. Factors associated with lack of viral suppression were similar among minority racial/ethnic groups. Interventions at the individual level focusing on young, U.S. born individuals, and those who report drug use, and at the neighborhood level for those living in residentially unstable neighborhoods are needed to improve viral suppression outcomes.
Keywords: race/ethnicity, minority women, viral suppression, HIV
Introduction
Viral suppression, which requires consistent adherence to antiretroviral therapy (ART), significantly improves the health outcomes of people diagnosed with HIV (PWH) and also reduces the likelihood of transmitting the virus to an uninfected individual.1,2 In 2018, among PWH age 13 years and older in the United States, women had a lower percentage of viral suppression (63.0%) than men (65.1%).3 However, both of these estimates were far below the National HIV/AIDS Strategy (NHAS) goal of having at least 80% of PWH achieve viral suppression by 2020.4 Women are less likely to be engaged in care, prescribed or use ART, and achieve viral suppression when compared to men.5–7
Among HIV-infected women, disparities in viral suppression exist by race/ethnicity.8,9 For instance, a study conducted using nationally representative data revealed a lower prevalence of viral suppression among minority women [Black (55%) and Hispanic (58%) women] compared with White women (62%).10 Similarly, another study among six U.S. cities reported comparable findings, with Black women having lower prevalence of viral suppression (63%) than Hispanics (73%) and Whites (78%).11 Minority women face numerous barriers to care including poverty, education, language barriers, as well access to insurance and care, when compared with White women.8,10
The Ryan White Program (RWP) provides HIV medical care, support services, and medications to more than half a million uninsured PWH in the United States, most of whom are racial/ethnic minorities.12 In Miami-Dade County, nearly 10,000 PWH receive RWP services.13 Identifying factors associated with HIV viral suppression among minority women in the RWP will help in developing tailored interventions to help minority women experiencing significant psychosocial barriers and ultimately contribute to the attainment of the national goal of decreasing HIV-related disparities. Considering the individual and community benefits of viral suppression and the disparities observed among minority women, the objective of this study was to assess the prevalence of and identify factors associated with viral suppression among minority women enrolled in the Miami-Dade County Ryan White Program.
Methods
Study design and study population
A retrospective cohort study was conducted using administrative data on minority women (Hispanic, African American, and Haitian) enrolled in the Miami-Dade County RWP in 2017. Minority women were considered enrolled if they received care from 1 of the 13 RWP agencies providing medical case management during 2017. These agencies provide numerous services and vary by size and percentage of women clients served. Administrative data included sociodemographic, self-reported psychosocial and health status factors, and clinical laboratory assessments. Medical case managers conducted comprehensive health assessments twice a year for RWP clients. We used data from the first health assessment in 2017 for each client, and if there was no health assessment in 2017, we used the last health assessment in 2016. Minority women with closed files due to death in 2017, relocation out of the county, incarceration, loss to follow-up in care, or financial ineligibility for the RWP were excluded from the analysis. Additionally, non-Hispanic White women and women of other race/ethnicities were removed from the analysis due to small numbers (n = 59).
Outcome and predictor variables
The outcome of interest, viral suppression, was defined as having <200 copies/mL using the last viral load test in 2017, and treated as a binary (yes, no) variable.14 Predictor variables (demographic, vulnerable/enabling characteristics, need factors, health care environment, and neighborhood factors) for this study were chosen based on the Behavioral Model for Vulnerable Population adapted for HIV by Christopoulos et al.15 and the availability of these variables in the RWP data (Table 1). Race/ethnicity variables were grouped as African Americans (excluding Hispanic and Haitian Blacks), Hispanics, and Haitians based on their response for ethnicity, preferred language, and country of origin. For individuals whose ethnicity category was inconsistent with their preferred language (e.g., Haitian ethnicity but Spanish as preferred language) (n = 8), race/ethnicity was determined using both preferred language and country of birth. One individual had more than one ethnicity reported, and that person's preferred language was used to determine the race/ethnicity classification. Finally, women were categorized as U.S. born (those born in any of the 50 states) and non-U.S. born individuals [those born in another country or U.S. territories such as Puerto Rico (n = 20) and U.S. Virgin Islands (n = 2)] as some studies suggest similar health outcomes are observed among those born in U.S. territories and outside the United States.16
Table 1.
Characteristics of Minority Women of the Miami-Dade County Ryan White Program, 2017
| Total, n | African American n (%) |
Hispanic n (%) |
Haitian n (%) |
p | |
|---|---|---|---|---|---|
| Total | 1,550 | 668 (43.1) | 485 (31.3) | 397 (25.6) | |
| Demographic characteristics | |||||
| Age group (years) | <.0001 | ||||
| 18–34 | 240 | 124 (18.6) | 66 (13.6) | 50 (12.6) | |
| 35–49 | 580 | 218 (32.6) | 219 (45.2) | 143 (36.0) | |
| ≥50 | 730 | 326 (48.8) | 200 (41.2) | 204 (51.4) | |
| Born in the United States | <.0001 | ||||
| Yes | 654 | 579 (86.7) | 53 (10.9) | 22 (5.5) | |
| No | 896 | 89 (13.3) | 432 (89.1) | 375 (94.5) | |
| Household income, percent of FPL | .6230 | ||||
| ≥200% | 190 | 89 (13.3) | 59 (12.2) | 42 (10.6) | |
| 100%–199% | 555 | 229 (34.3) | 181 (37.3) | 145 (36.5) | |
| <100% | 805 | 350 (52.4) | 245 (50.5) | 210 (52.9) | |
| Number of minors in household | .0011 | ||||
| None | 1,106 | 507 (75.9) | 346 (71.3) | 253 (63.7) | |
| One | 240 | 83 (12.4) | 78 (16.1) | 79 (19.9) | |
| Two | 134 | 48 (7.2) | 46 (9.5) | 40 (10.1) | |
| Three or more | 70 | 30 (4.5) | 15 (3.1) | 25 (6.3) | |
| Need characteristics | |||||
| Diagnosis of AIDS at any time | .0587 | ||||
| Yes | 773 | 338 (50.6) | 222 (45.8) | 213 (53.7) | |
| No | 777 | 330 (49.4) | 263 (54.2) | 184 (46.4) | |
| Total | 1,550 | 668 (43.1) | 485 (31.3) | 397 (25.6) | |
| Vulnerable/enabling variables | |||||
| Drug use | <.0001 | ||||
| Used drugs in the last 12 months | 87 | 64 (9.6) | 22 (4.5) | 1 (0.3) | |
| Did not use drugs in the last 12 months | 1,463 | 604 (90.4) | 463 (95.5) | 396 (99.8) | |
| Problematic drug use (drug use resulted in problems with daily activities or legal issue or hazardous situation) | .0021 | ||||
| Yes | 40 | 25 (3.7) | 14 (2.9) | 1 (0.3) | |
| No | 1,510 | 643 (96.3) | 471 (97.1) | 396 (99.8) | |
| Alcohol/drugs affects adherence | <.0001 | ||||
| Yes | 52 | 40 (6.0) | 12 (2.5) | 0 (0.00) | |
| No | 1,498 | 628 (94.0) | 473 (97.5) | 397 (100.00) | |
| Would like substance use treatment now | <.0001 | ||||
| Yes | 31 | 26 (3.9) | 4 (0.8) | 1 (0.3) | |
| No | 1,519 | 642 (96.1) | 481 (99.2) | 396 (99.8) | |
| Feeling depressed or anxious | <.0001 | ||||
| Yes | 272 | 158 (23.7) | 85 (17.5) | 29 (7.3) | |
| No | 1,278 | 510 (76.4) | 400 (82.5) | 368 (92.7) | |
| Receives or needs mental health services | <.0001 | ||||
| Yes | 289 | 175 (26.2) | 96 (19.8) | 18 (4.5) | |
| No | 1,261 | 493 (73.8) | 389 (80.2) | 379 (95.5) | |
| Total | 1,550 | 668 (43.1) | 485 (31.3) | 397 (25.6) | |
| Ever experienced domestic violence | <.0001 | ||||
| Yes | 123 | 75 (11.2) | 42 (8.7) | 6 (1.5) | |
| No | 1,427 | 593 (88.8) | 443 (91.3) | 391 (98.5) | |
| Has a social support system to depend on | <.0001 | ||||
| Yes | 1,330 | 593 (88.8) | 384 (79.2) | 353 (88.9) | |
| No | 220 | 75 (11.2) | 101 (20.8) | 44 (11.1) | |
| Disclosure of HIV status to partner or an adult in households | .0010 | ||||
| No adult in household and no partner | 513 | 215 (32.2) | 162 (33.4) | 136 (34.3) | |
| At least one adult in household or partner, but partner and adult do not know status | 201 | 85 (12.7) | 44 (9.1) | 72 (18.1) | |
| Adult in household or a partner knows status | 836 | 368 (55.1) | 279 (57.5) | 189 (47.6) | |
| Work-related barriers to attending care appointments | <.0001 | ||||
| No | 709 | 250 (37.4) | 253 (52.2) | 206 (51.9) | |
| Yes | 48 | 25 (3.7) | 15 (3.1) | 8 (2.0) | |
| Not working | 793 | 393 (58.8) | 217 (44.7) | 183 (46.1) | |
| Client has access to transportation to appointments | .1432 | ||||
| Yes | 1,429 | 607 (90.9) | 448 (92.4) | 374 (94.2) | |
| No | 121 | 61 (9.1) | 37 (7.6) | 23 (5.8) | |
| Total | 1,550 | 668 (43.1) | 485 (31.3) | 397 (25.6) | |
| Client getting food he/she needs | .2614 | ||||
| Yes | 1,530 | 657 (98.4) | 478 (98.6) | 395 (99.5) | |
| No | 20 | 11 (1.7) | 7 (1.4) | 2 (0.5) | |
| Homeless | |||||
| Yes | 69 | 47 (7.0) | 13 (2.7) | 9 (2.3) | <.0001 |
| No | 1,481 | 621 (93.0) | 472 (97.3) | 388 (97.7) | |
| Infected perinatally with HIVa | |||||
| Yes | 23 | 13 (2.0) | 4 (0.8) | 6 (1.5) | .2982 |
| No | 1,519 | 655 (98.1) | 478 (99.2) | 386 (98.5) | |
| Health care environment | |||||
| Number of Ryan White clients that client's clinician cares for | <.0001 | ||||
| 1–9 | 66 | 46 (6.9) | 11 (2.3) | 9 (2.3) | |
| 10–29 | 58 | 37 (5.5) | 17 (3.5) | 4 (1.0) | |
| 30–99 | 416 | 137 (20.5) | 113 (23.3) | 166 (41.8) | |
| 100–199 | 494 | 252 (37.7) | 98 (20.2) | 144 (36.3) | |
| ≥200 | 426 | 153 (22.9) | 218 (45.0) | 55 (13.9) | |
| Unknown | 90 | 43 (6.4) | 28 (5.8) | 19 (4.8) | |
| Neighborhood environment | |||||
| Neighborhood low SES index | |||||
| Median | 1.37 | 0.71 | 0.73 | <.0001 | |
| Interquartile range | (0.47; 1.68) | (−0.14; 1.35) | (0.67; 1.04) | ||
| Neighborhood residential instability/homicide index | |||||
| Median | 0.45 | 0.09 | 0.33 | <.0001 | |
| Interquartile range | (0.05; 1.28) | (−0.47; 0.88) | (−0.29; 0.65) | ||
Bold text represents p-values < .05.
The number of missing values for Hispanic by variable is: perinatal infection, n = 3. The number of missing values for Haitian by variable is: perinatal infection, n = 5.
FPL, federal poverty level; SES, socioeconomic status.
Twenty-five neighborhood level variables were extracted by ZIP code tabulated areas (ZCTA) to determine 5-year estimates (2013–2017) from the 2017 American Community Survey (details are provided elsewhere).17,18 ZCTAs are generalized area representations of ZIP Code service areas, and for this study serve as approximations of neighborhoods.19 Data on homicides were extracted from Simply Analytics, for each ZIP code in Miami-Dade County.20 Due to a large number of neighborhood variables, we employed a data reduction technique to create indices. First, we conducted a reliability analysis, where eight variables were removed to improve the Cronbach's alpha from 0.8084 to 0.9386. Second, exploratory factor analysis with and without varimax rotation was conducted; a two-factor loading pattern was selected: low socioeconomic status (SES) (loadings ranged between 0.45 and 0.93), and residential instability/homicide (loadings ranged between 0.66 and 0.72). Based on cut points from the literature, we removed variables with factor loadings <0.4.21 Lastly, confirmatory factor analysis was conducted to calculate scores for each factor as the linear combination of the standardized values of the variables. The indices created were later merged with the RWP data set.
Statistical analysis
All analyses were conducted in SAS version 9.4.22 Using chi-square tests for categorical variables and Wilcoxon signed-rank test for continuous variables, we compared the differences in client characteristics by race/ethnicity stratified by viral suppression status. Multilevel logistic regression models using medical case management groups as a random effect were computed using the PROC GLIMMIX procedure in SAS. Medical case management groups were used to assess site variations that were observed in HIV care outcomes for racial/ethnic groups during a prior preliminary analysis. We assessed collinearity among all variables in the model, and preferred language was removed due to collinearity with race/ethnicity (type II tolerance = 0.03). The model was not stratified by race/ethnicity due to small cell numbers. Additionally, we tested for interactions of race/ethnicity with all variables that were significant in the model. This study was approved by the Florida International University Institutional Review Board.
Results
Characteristics of study sample
Of the 1,550 racial/ethnic minority women included in the study, 43.1% were African American, 31.3% were Hispanic, and 25.6% were Haitian. The demographic characteristics of the women in each racial/ethnic group are reported in Table 1. Among the three racial/ethnic groups in the study, Hispanic women were the youngest (18–49 years old). African Americans were more likely to be born in the United States, while Haitians and Hispanics were more likely to be born outside of the United States. History of drug use during the 12 months before the health assessment, problematic drug or alcohol use, and previous experience of domestic violence were more commonly reported among African Americans compared to other groups. Hispanic women were more likely to report disclosure of their HIV status to a partner or an adult in the household. Additionally, African American women were more likely to live in neighborhoods with lower SES and higher residential instability, as indicated by the higher indices (Table 1). Across the three racial/ethnic groups, there were no significant differences in household income, AIDS diagnosis at any time, transportation to appointments, getting needed food, and perinatal infection (Table 1).
Prevalence of and racial/ethnic differences in client characteristics by viral suppression in minority women
Overall, 83.7% of minority women were virally suppressed at their last viral load test of the year. Prevalence of viral suppression (i.e., having viral load <200 copies/mL) differed significantly among the three racial ethnic groups, with Hispanics (86.4%) and Haitians (85.1%) having a higher prevalence than African Americans (80.8%) (p value: .0276). Among the three racial/ethnic groups, being born in the United States, drug use in the last 12 months, and problematic drug use were significantly associated with viral suppression (Table 2). Age, household income, and neighborhood residential instability were significantly associated with viral suppression among African Americans and Hispanics but not among Haitians. Among African Americans and Haitians, wanting substance use treatment was significantly associated with viral suppression. Previous diagnosis of AIDS and homelessness were significantly associated with viral suppression among African Americans. Feeling depressed or anxious, needing mental health services, work-related barriers to attending care appointments, and getting needed food were significantly associated with viral suppression among Hispanics. Perinatal infection was only significantly associated with viral suppression among Haitians.
Table 2.
Viral Suppression in HIV Care by Characteristics for Minority Women, Miami-Dade County Ryan White Program, 2017
| Characteristics | African American (n = 668) |
Hispanic (n = 485) |
Haitian (n = 397) |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Virally suppressed, n (%) 540 (80.8) |
Not virally suppressed, n (%) 128 (19.2) |
p | Virally suppressed, n (%) 419 (86.4) |
Not virally suppressed, n (%) 66 (13.6) |
p | Virally suppressed, n (%) 338 (85.1) |
Not virally suppressed, n (%) 59 (14.9) |
p | |
| Demographic characteristics | |||||||||
| Age | .0017 | .0016 | .0552 | ||||||
| 18–34 | 87 (70.2) | 37 (29.8) | 49 (74.2) | 17 (25.8) | 38 (76.0) | 12 (24.0) | |||
| 35–49 | 176 (80.7) | 42 (19.3) | 187 (85.4) | 32 (14.6) | 119 (83.2) | 24 (16.8) | |||
| ≥50 | 277 (85.0) | 49 (15.0) | 183 (91.5) | 17 (8.5) | 181 (88.7) | 23 (11.3) | |||
| Born in the United States | .0014 | .0040 | .0035 | ||||||
| Yes | 457 (78.9) | 122 (21.1) | 39 (73.6) | 14 (26.42) | 14 (63.64) | 8 (36.4) | |||
| No | 83 (93.3) | 6 (6.7) | 380 (88.0) | 52 (12.0) | 324 (86.4) | 51 (13.6) | |||
| Household income, percent of FPL | .0128 | .0107 | .1658 | ||||||
| ≥200% | 77 (86.5) | 12 (13.5) | 56 (94.9) | 3 (5.1) | 39 (92.9) | 3 (7.1) | |||
| 100%–199% | 195 (85.2) | 34 (14.9) | 162 (89.5) | 19 (10.5) | 126 (86.9) | 19 (13.1) | |||
| <100% | 268 (76.6) | 82 (23.4) | 201 (82.0) | 44 (18.0) | 173 (82.4) | 37 (17.6) | |||
| Number of minors in household | .2064 | .1423 | .2969 | ||||||
| None | 415 (81.9) | 92 (18.2) | 304 (87.9) | 42 (12.1) | 219 (86.6) | 34 (13.4) | |||
| One | 63 (75.9) | 20 (24.1) | 68 (87.2) | 10 (12.8) | 65 (82.3) | 14 (17.7) | |||
| Two | 41 (85.4) | 7 (14.6) | 36 (78.3) | 10 (21.7) | 31 (77.5) | 9 (22.5) | |||
| Three or more | 21 (70.0) | 9 (30.0) | 11 (73.3) | 4 (26.7) | 23 (92.0) | 2 (8.0) | |||
| Need characteristics | |||||||||
| Diagnosis of AIDS at any time | .0093 | .3138 | .5070 | ||||||
| Yes | 260 (76.9) | 78 (23.1) | 188 (84.7) | 34 (15.3) | 179 (84.0) | 34 (16.0) | |||
| No | 280 (84.9) | 50 (15.2) | 231 (87.8) | 32 (12.2) | 159 (86.4) | 25 (13.6) | |||
| Vulnerable/enabling variables | |||||||||
| Drug use | .0035 | .0001 | .0166 | ||||||
| Used drugs in the last 12 months | 43 (67.2) | 21 (32.8) | 13 (59.1) | 9 (40.9) | 0 (0.00) | 1 (100.00) | |||
| Did not use drugs in the last 12 months | 497 (82.3) | 107 (17.7) | 406 (87.7) | 57 (12.3) | 338 (85.4) | 58 (14.7) | |||
| Problematic drug use (drug use resulted in problems with daily activities or legal issue or hazardous situation) | <.0001 | <.0001 | .0166 | ||||||
| Yes | 12 (48.0) | 13 (52.0) | 6 (42.9) | 8 (57.1) | 0 (0.00) | 1 (100.00) | |||
| No | 528 (82.1) | 115 (17.9) | 413 (87.7) | 58 (12.3) | 338 (85.4) | 58 (14.7) | |||
| Alcohol/drugs affect adherence or not | .3332 | .2438 | |||||||
| Yes | 30 (75.0) | 10 (25.0) | 9 (75.0) | 3 (25.0) | |||||
| No | 510 (81.2) | 118 (18.8) | 410 (86.7) | 63 (13.3) | 338 (85.1) | 59 (14.9) | |||
| Would like substance use treatment now | .0004 | .5046 | .0166 | ||||||
| Yes | 14 (53.9) | 12 (46.2) | 3 (75.0) | 1 (25.0) | 0 (0.00) | 1 (100.00) | |||
| No | 526 (81.9) | 116 (18.1) | 416 (86.5) | 65 (13.5) | 338 (85.4) | 58 (14.7) | |||
| Feeling depressed or anxious | .1854 | .0033 | .4776 | ||||||
| Yes | 122 (77.2) | 36 (22.8) | 65 (76.5) | 20 (23.5) | 26 (89.7) | 3 (10.3) | |||
| No | 418 (82.0) | 92 (18.0) | 354 (88.5) | 46 (11.5) | 312 (84.8) | 56 (15.2) | |||
| Receives or needs mental health services | .0950 | .0083 | .8256 | ||||||
| Yes | 134 (76.6) | 41 (23.4) | 75 (78.1) | 21 (21.9) | 15 (83.3) | 3 (16.7) | |||
| No | 406 (82.4) | 87 (17.7) | 344 (88.4) | 45 (11.6) | 323 (85.2) | 56 (14.8) | |||
| Ever experienced domestic violence | .8448 | .1220 | .9003 | ||||||
| Yes | 60 (80.0) | 15 (20.0) | 33 (78.6) | 9 (21.4) | 5 (83.3) | 1 (16.7) | |||
| No | 480 (80.9) | 113 (19.1) | 386 (87.1) | 57 (12.9) | 333 (85.2) | 58 (14.8) | |||
| Has a social support system to depend on | .2585 | .3708 | .5114 | ||||||
| Yes | 483 (81.5) | 110 (18.6) | 329 (85.7) | 55 (14.3) | 302 (85.6) | 51 (14.5) | |||
| No | 57 (76.0) | 18 (24.0) | 90 (89.1) | 11 (10.9) | 36 (81.8) | 8 (18.2) | |||
| Disclosure of status to adults in households | .0970 | .1826 | .4501 | ||||||
| No adults in household | 165 (76.7) | 50 (23.3) | 138 (85.2) | 24 (14.8) | 164 (86.8) | 25 (13.2) | |||
| Adults in household, but none know status | 74 (87.1) | 11 (12.9) | 42 (95.5) | 2 (4.6) | 58 (80.6) | 14 (19.4) | |||
| At least one adult in household knows status | 301 (81.8) | 67 (18.2) | 239 (85.7) | 40 (14.3) | 116 (85.3) | 20 (14.7) | |||
| Work related barriers to attending care appointments | .1920 | .0187 | .7938 | ||||||
| No | 209 (83.6) | 41 (16.4) | 229 (90.5) | 24 (9.5) | 173 (84.0) | 33 (16.0) | |||
| Yes | 22 (88.0) | 3 (12.0) | 13 (86.7) | 2 (13.3) | 7 (87.5) | 1 (12.5) | |||
| Not working | 309 (78.6) | 84 (21.4) | 177 (81.6) | 40 (18.4) | 158 (86.3) | 25 (13.7) | |||
| Client has access to transportation to appointments | .4302 | .0479 | .3394 | ||||||
| Yes | 493 (81.2) | 114 (18.8) | 391 (87.3) | 57 (12.7) | 320 (85.6) | 54 (14.4) | |||
| No | 47 (77.1) | 14 (23.0) | 28 (75.7) | 9 (24.3) | 18 (78.3) | 5 (21.7) | |||
| Client getting food he/she needs | .9336 | .0230 | .5536 | ||||||
| Yes | 531 (80.8) | 126 (19.2) | 415 (86.8) | 63 (13.2) | 336 (85.1) | 59 (14.9) | |||
| No | 9 (81.8) | 2 (18.2) | 4 (57.1) | 3 (42.9) | 2 (100.00) | 0 (0.00) | |||
| Homeless | .0021 | .0674 | .1151 | ||||||
| Yes | 30 (63.8) | 17 (36.2) | 9 (69.2) | 4 (30.8) | 6 (66.67) | 3 (33.3) | |||
| No | 510 (82.1) | 111 (17.9) | 410 (86.9) | 62 (13.1) | 332 (85.6) | 56 (14.4) | |||
| Infected perinatally with HIV | .0742 | .5046 | .0003 | ||||||
| Yes | 8 (61.5) | 5 (38.5) | 3 (75.0) | 1 (25.0) | 2 (33.3) | 4 (66.7) | |||
| No | 532 (81.2) | 123 (18.8) | 416 (86.5) | 65 (13.5) | 336 (85.9) | 55 (14.1) | |||
| Health care environment | |||||||||
| Number of Ryan White clients that client's clinician cares for | .5767 | .2797 | .9295 | ||||||
| 1–9 | 33 (71.7) | 13 (28.3) | 8 (72.7) | 3 (27.3) | 8 (88.9) | 1 (11.1) | |||
| 30–99 | 109 (79.6) | 28 (20.4) | 100 (88.5) | 13 (11.5) | 139 (83.7) | 27 (16.3) | |||
| 100–199 | 206 (81.8) | 46 (18.3) | 81 (82.7) | 17 (17.4) | 123 (85.4) | 21 (14.6) | |||
| ≥200 | 126 (82.4) | 27 (17.7) | 194 (89.0) | 24 (11.0) | 47 (85.5) | 8 (14.6) | |||
| Unknown | 34 (79.1) | 9 (20.9) | 22 (78.6) | 6 (21.4) | 17 (89.5) | 2 (10.5) | |||
| Neighborhood environment | |||||||||
| Neighborhood low SES index | |||||||||
| Median | 1.37 | 1.37 | .4529 | 0.71 | 0.97 | .1353 | 0.73 | 0.73 | .5608 |
| Interquartile range | (0.47, 1.68) | (0.46, 1.68) | (−0.14, 1.35) | (−0.09, 1.51) | (0.67, 1.04) | (0.47, 1.04) | |||
| Neighborhood residential instability/homicide index | |||||||||
| Median | 0.45 | 0.80 | .0018 | 0.09 | 0.65 | .0489 | 0.33 | 0.65 | .0749 |
| Interquartile range | (−0.04, 1.28) | (0.33, 1.28) | (−0.52, 0.87) | (−0.31, 1.28) | (−0.29, 0.65) | (−0.29, 0.75) | |||
Bold text represents p-values < .05.
Multivariate analysis of factors associated with viral suppression among minority women
In the multilevel logistic regression model using medical case management site as a random effect, younger age [18–34 vs. ≥50 years: adjusted odds ratio (aOR): 0.41, 95% confidence interval (CI): 0.27–0.64]; (35–49 vs. ≥50 years: aOR: 0.63, 95% CI: 0.45–0.90), being born in the United States (aOR: 0.48, 95% CI: 0.30–0.78), having a household income of <100% federal poverty level (FPL) vs. ≥200% FPL (aOR: 0.54: 95% CI: 0.30–0.95), being previously diagnosed with AIDS (aOR 0.60: 95% CI: 0.44–0.81), problematic drug use (aOR 0.23: 95% CI: 0.08–0.69), and living in a residentially unstable neighborhood (aOR: 0.77, 95% CI: 0.64–0.93) were significantly associated with lower odds of viral suppression (Table 3). In the unadjusted analysis, Hispanics had a significantly higher odds of being virally suppressed when compared to African Americans (odds ratio: 1.47, 95% CI: 1.06–2.04), while odds among Haitians was not significantly different. However, race/ethnicity was not associated with viral suppression adjusting for other covariates. Additionally, interaction was not observed between race/ethnicity and all other variables.
Table 3.
Crude and Adjusted Odds Ratios of Viral Suppression in HIV Care by Characteristics for Minority Women, Miami-Dade County Ryan White Program, 2017 (N = 1,547)
| Characteristic | Crude OR and 95% CI | aOR and 95% CI |
|---|---|---|
| Demographic characteristics | ||
| Race/ethnicity | ||
| African American | Ref. | Ref. |
| Hispanic | 1.47 (1.06–2.04) | 0.77 (0.46–1.28) |
| Haitian | 1.31 (0.93–1.85) | 0.66 (0.38–1.13) |
| Age (years) | ||
| 18–34 | 0.36 (0.25–0.53) | 0.41 (0.27–0.64) |
| 35–49 | 0.67 (0.49–0.93) | 0.63 (0.45–0.90) |
| ≥50 | Ref. | Ref. |
| U.S.-born | ||
| Yes | 0.49 (0.38–0.65) | 0.48 (0.30–0.78) |
| No | Ref. | Ref. |
| Household income, percent of FPL | ||
| ≥200% | Ref. | Ref. |
| 100%–199% | 0.71 (0.41–1.21) | 0.80 (0.46–1.42) |
| <100% | 0.42 (0.25–0.69) | 0.54 (0.30–0.95) |
| Number of minors in household | ||
| None | Ref. | Ref. |
| One | 0.79 (0.55–1.14) | 0.79 (0.52–1.19) |
| Two | 0.74 (0.46–1.17) | 0.78 (0.47–1.31) |
| Three or more | 0.64 (0.35–1.17) | 0.94 (0.48–1.81) |
| Need characteristics | ||
| Diagnosis of AIDS at any time | ||
| Yes | 0.65 (0.49–0.86) | 0.60 (0.44–0.81) |
| No | Ref. | Ref. |
| Vulnerable/enabling variables | ||
| Drug use | ||
| Used drugs in the last 12 months | 0.33 (0.21–0.53) | 1.01 (0.33–3.05) |
| Did not use drugs in the last 12 months | Ref. | Ref. |
| Drug use resulted in problems with daily activities or legal issue or hazardous situation | ||
| Yes | 0.15 (0.08–0.29) | 0.23 (0.08–0.69) |
| No | Ref. | Ref. |
| Drug use affect adherence or not | ||
| Yes | 0.61 (0.32–1.17) | 1.54 (0.53–4.48) |
| No | Ref. | |
| Would like substance use treatment now | ||
| Yes | 0.24 (0.11–0.49) | 0.85 (0.32–2.26) |
| No | Ref. | Ref. |
| Feeling depressed or anxious | ||
| Yes | 0.67 (0.48–0.94) | 0.82 (0.54–1.25) |
| No | Ref. | Ref. |
| Receives or needs mental health services | ||
| Yes | 0.62 (0.45–0.86) | 0.91 (0.59–1.39) |
| No | Ref. | Ref. |
| Ever experienced domestic violence | ||
| Yes | 0.78 (0.48–1.24) | 1.26 (0.73–2.18) |
| No | Ref. | Ref. |
| Disclosure of status to adults in households | ||
| No adults in household | 0.84 (0.63–1.13) | 0.78 (0.56–1.07) |
| Adults in household, but none know status | 1.19 (0.76–1.85) | 0.99 (0.62–1.59) |
| At least one adult in household knows status | Ref. | Ref. |
| Has a social support system to depend on | ||
| Yes | Ref. | Ref. |
| No | 0.97 (0.66–1.42) | 1.03 (0.68–1.57) |
| Work related barriers to attending care appointments | ||
| No | Ref. | Ref. |
| Yes | 1.01 (0.91–1.13) | 0.89 (0.36–2.23) |
| Not working | 0.95 (0.92–0.99) | 0.91 (0.64–1.28) |
| Client has access to transportation to appointments | ||
| Yes | 0.63 (0.40–0.99) | 0.79 (0.48–1.32) |
| No | Ref. | Ref. |
| Client getting food he/she needs | ||
| Yes | 0.61 (0.22–1.68) | 0.90 (0.28–2.95) |
| No | Ref. | Ref. |
| Homeless | ||
| Yes | 0.36 (0.21–0.60) | 0.99 (0.49–1.99) |
| No | Ref. | Ref. |
| Infected perinatally with HIV | ||
| Yes | 0.22 (0.09–0.50) | 0.50 (0.19–1.32) |
| No | Ref. | Ref. |
| Health care environment | ||
| Number of Ryan White clients that client's clinician cares for | ||
| 1–9 | 0.89 (0.81–0.98) | 0.56 (0.28–1.10) |
| 10–29 | 1.00 (0.90–1.11) | 1.19 (0.50–2.86) |
| 30–99 | 0.98 (0.93–1.03) | 0.75 (0.50–1.14) |
| 100–199 | 0.97 (0.92–1.02) | 0.82 (0.55–1.23) |
| ≥200 | Ref. | Ref. |
| Unknown | 0.95 (0.87–1.03) | 1.16 (0.59–2.26) |
| Neighborhood environment | ||
| Neighborhood low SES index | 0.86 (0.72–1.04) | 1.05 (0.85–1.29) |
| Neighborhood residential instability/homicide index | 0.70 (0.59–0.83) | 0.77 (0.64–0.93) |
Bolding of odds ratios and 95% confidence intervals indicate statistically significance.
OR are adjusting for all factors listed in this table.
Ref. = referent group.
Missing values = 3. Three records were excluded due to missing values in the variable perinatal infection.
aOR, adjusted odds ratio; CI, confidence interval; OR, odds ratio.
Discussion
In this study, an estimated 83.7% of minority women receiving RWP services in 2017 had viral suppression; however, we observed differences in the percentages of viral suppression by race/ethnicity. Although women in each of the minority groups met the national HIV prevention goal of having at least 80% of PWH achieving viral suppression,4 Hispanic and Haitian women largely surpassed this national goal. Consistent with previous studies,11,23 African American women had a lower percentage of viral suppression than Hispanic women. Our unadjusted analysis showed an association between race/ethnicity and viral suppression, which disappeared as we controlled for psychosocial and neighborhood factors. This finding suggests that observed racial/ethnic differences are likely due to confounding by factors such as lower age, place of birth, poverty, problematic substance use, and the neighborhood environment as opposed to belonging to a particular racial/ethnic group.
One of these factors is place of birth. U.S. born women were less likely to be virally suppressed compared to foreign-born women in our study. Similar results were obtained from a study by Myers et al. reporting a higher proportion of viral suppression among foreign-born adults receiving care compared to their U.S. counterparts.24 A possible explanation for this difference might be that foreign-born people, at least those that were able to obtain access to the Ryan White Program, possess some coping skills that facilitate better adherence to ART and retention in care.24 Higher odds of viral suppression among foreign-born might be reflective of the resilience and self-efficacy observed among some immigrants.25 However, in contrast to our results, some studies have reported no difference in viral suppression among foreign-born and U.S. born adults in care.26,27
Additionally, differences in viral suppression status were observed by race/ethnicity and U.S. birth status. In our study, African American women were more likely to be born in the United States and have demonstrated lower percentage of viral suppression. This finding is also consistent with a study done among women living with HIV in Florida.28 Poor retention in care and viral suppression were observed among U.S. born Blacks compared to their foreign counterparts, while, Hispanics were more likely to be foreign-born and have higher odds of being linked to care, retained in care and achieve viral suppression.16,29 Given the population demographic of Miami-Dade County, a majority Hispanic and immigrant population, foreign-born Hispanic's health outcomes may be attributed to decreased language and culture barriers in accessing care within Miami-Dade County. Further studies need to be conducted to assess factors that affect U.S. nativity, racial/ethnic differences, and viral suppression.
In the multivariable model, we found that younger minority women had reduced odds of being virally suppressed compared to their older counterparts, and this is similar to results of previous studies of lower rates of viral suppression among younger women.9,23,30 This suggests the need for interventions that address barriers that are particularly relevant to younger women living with HIV in these populations. Prior studies have established that women of lower income have reduced odds of achieving viral suppression.8,31 This is consistent with our result of women whose household income was below 100% of the FPL being less likely to achieve viral suppression. In line with previous research,32,33 our findings show reduced odds of achieving viral suppression among women who reported problems with drug or alcohol use. Lower rates of viral suppression among those with substance abuse may be due to poor engagement in continual care or lack of adherence to medication.34 Primary prevention of drug use should be prioritized to attain the goal of having 80% of people on ART achieve viral suppression.4
In this study, we found that women living in a residentially unstable neighborhood were less likely to be virally suppressed, which is consistent with previous studies highlighting the adverse health effect of living in an unstable neighborhood.35 Chandran et al., reported that women living in neighborhoods with increased residential stability had a higher likelihood of being engaged in care and increased antiretroviral adherence due to social support,36 which consequently improves their chances of being virally suppressed. Finally, in contrast to a study done in Florida reporting the association between lack of AIDS diagnosis and likelihood of not being virally suppressed,29 our findings showed that women diagnosed with AIDS at any time, have significantly reduced odds of being virally suppressed.
Our study has some limitations. First, this study was carried out among minority women accessing care in the RWP. Therefore, our findings may not be generalizable to minority women accessing care outside of RWP or women living with HIV who are not in care. Second, the RWP dataset is composed of administrative data and not originally collected for research purposes. Thus, some of the measures (e.g., drug use) are not collected using validated instruments, and there was likely a lack of uniformity in how these questions were asked because case managers from multiple case management systems administered the questions for the purpose of needs assessment and not research.
Conclusion
Suboptimal viral suppression was observed among minority women, particularly those who are born in the United States, in the RWP. Both individual (young age, being born in the United States, low income, previous AIDS diagnosis, and problematic drug use) and neighborhood (residential instability) factors were associated with lack of viral suppression. Interventions targeted to improve viral suppression among younger women and those with drug use problems should be encouraged. Also, studies examining the mechanism of neighborhood residential instability and viral suppression should be conducted.
Acknowledgments
The authors wish to acknowledge Carla Valle-Schwenk, Ryan White Program Administrator, and the entire Ryan White Part A Program in the Miami-Dade County Office of Management and Budget, for their continued assistance, cooperation, and facilitation in the implementation of this study.
Authors' Contributions
S.O.C.-A.: Conceptualization, Data Analysis, Writing-Original, Review, and Editing, Final Approval. R.D. and S.O.G.: Data Analysis, Writing-Original, Review, and Editing, Final Approval. K.P.F. and D.M.S.: Data Analysis, Writing-Review and Editing, Final Approval. P.B. and R.A.L.: Data Curation, Writing-Review and Editing, Final Approval. M.J.T.: Conceptualization, Data Analysis, Writing-Review and Editing, Final Approval.
Disclaimer
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.
Funding Information
The project described was supported by Award Numbers R01MD013563 and in part by R01MD012421, U54MD012393, K01MD013770, and F31MD015234 from the National Institute on Minority Health and Health Disparities at the National Institutes of Health.
References
- 1. Bavinton BR, Pinto AN, Phanuphak N, et al. : Viral suppression and HIV transmission in serodiscordant male couples: An international, prospective, observational, cohort study. Lancet HIV 2018;5:e438–e447 [DOI] [PubMed] [Google Scholar]
- 2. 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]
- 3. Centers for Disease Control and Prevention: Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 dependent areas, 2018. HIV Surveillance Supplemental Report 2020;25(No. 2). Available at www.cdc.gov/hiv/library/reports/hiv-surveillance.html. (2020), accessed April08, 2021 [Google Scholar]
- 4. National HIV/AIDS Strategy for the United States: 2017. Progress Report. Available at https://www.hiv.gov/federal-response/national-hiv-aids-strategy/nhas-update (2017), accessed March12, 2020
- 5. Matson TE, McGinnis KA, Rubinsky AD, et al. : Gender and alcohol use: Influences on HIV care continuum in a national cohort of patients with HIV. AIDS 2018;32:2247–2253 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Meyer JP, Zelenev A, Wickersham JA, Williams CT, Teixeira PA, Altice FL: Gender disparities in HIV treatment outcomes following release from jail: Results from a multicenter study. Am J Public Health 2014;104:434–441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Sayles JN, Rurangirwa J, Kim M, Kinsler J, Oruga R, Janson M: Operationalizing treatment as prevention in Los Angeles County: Antiretroviral therapy use and factors associated with unsuppressed viral load in the Ryan White system of care. AIDS Patient Care STDS 2012;26:463–470 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Nwangwu-Ike N, Frazier EL, Crepaz N, Tie Y, Sutton MY: Racial and ethnic differences in viral suppression among HIV-positive women in care. JAIDS 2018;79:e56–e68 [DOI] [PubMed] [Google Scholar]
- 9. Nwangwu-Ike N, Hernandez AL, An Q, Huang T, Hall HI: The epidemiology of human immunodeficiency virus infection and care among adult and adolescent females in the United States, 2008–2012. Womens Health Issues 2015;25:711–719 [DOI] [PubMed] [Google Scholar]
- 10. Beer L, Mattson CL, Bradley H, Skarbinski J: Understanding cross-sectional racial, ethnic, and gender disparities in antiretroviral use and viral suppression among HIV patients in the United States. Medicine 2016;95:e3171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Geter A, Sutton MY, Armon C, Buchacz K, HIV Outpatient Study Investigators: Behavior: Disparities in viral suppression and medication adherence among women in the USA, 2011–2016. AIDS 2019;23:3015–3023 [DOI] [PubMed] [Google Scholar]
- 12. Health Resources and Services Administration (HRSA): Ryan White and Global HIV/AIDS Program. Available at https://hab.hrsa.gov/about-ryan-white-hivaids-program/about-ryan-white-hivaids-program (2019), accessed April21, 2020
- 13. Miami-Dade County: Ryan White Program. Available at https://www.miamidade.gov/global/service.page?Mduid_service=ser1482944607068715 (2020), accessed April21, 2020
- 14. Centers for Disease Control and Prevention: HIV Treatment as Prevention Technical Fact Sheet. National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention. Division of HIV/AIDS Prevention. Available at https://www.cdc.gov/hiv/pdf/risk/art/cdc-hiv-art-viral-suppression.pdf (2018), accessed April21, 2020
- 15. Christopoulos KA, Das M, Colfax GN: Linkage and retention in HIV care among men who have sex with men in the United States. Clin Infect Dis 2011;52(suppl_2):S214–S222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Demeke HB, Luo Q, Luna-Gierke RE, et al. : HIV Care outcomes among Hispanics/Latinos with diagnosed HIV in the United States by place of birth-2015–2018, medical monitoring project. Int J Environ Res Public Health 2020;17:171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Trepka MJ, Sheehan DM, Dawit R, et al. : Differential role of psychosocial, health care system and neighborhood factors on the retention in HIV care of women and men in the Ryan White Program. J Int Assoc Provid AIDS Care 2020;19:2325958220950087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. American Community Survey: 2011. –2017 American Community Survey 5-Year Estimates. Washington, DC, USA. Available at https://factfinder.census.gov (2017), accessed March18, 2019
- 19. Bureau: USC. Zip Code Tabulation Areas (ZCTAs). Available at https://www.census.gov/programs-surveys/geography/guidance/geo-areas/zctas.html (2017), accessed July15, 2020
- 20. Simply Analytics: Simply Analytics Data. Available at https://simplyanalytics.com/ (2019), accessed December19, 2019
- 21. Peterson RA: A meta-analysis of variance accounted for and factor loadings in exploratory factor analysis. Market Lett 2000;11:261–275 [Google Scholar]
- 22. SAS version 9.4 Software: SAS Institute, Inc., Cary, North Carolina
- 23. Geter A, Sutton MY, Armon C, et al. : Trends of racial and ethnic disparities in virologic suppression among women in the HIV Outpatient Study, USA, 2010–2015. PLoS One 2018;13:e0189973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Myers TR, Lin X, Skarbinski J: Antiretroviral therapy and viral suppression among foreign-born HIV-infected persons receiving medical care in the United States: A complex sample, cross-sectional survey. Medicine 2016;95:e3051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Poon KK, Dang BN, Davila JA, Hartman C, Giordano TP: Treatment outcomes in undocumented Hispanic immigrants with HIV infection. PLoS One 2013;8:e60022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Levison JH, Regan S, Khan I, Freedberg KA: Foreign-born status as a predictor of engagement in HIV care in a large US metropolitan health system. AIDS Care 2017;29:244–251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Parrish DD, Blevins M, Stinnette SE, et al. : Virologic, immunologic and clinical responses in foreign-born versus US-born HIV-1 infected adults initiating antiretroviral therapy: An observational cohort study. PLoS One 2012;7:e52336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Dawit R, Sheehan DM, Gbadamosi SO, et al. : Identifying patterns of retention in care and viral suppression using latent class analysis among women living with HIV in Florida 2015–2017. AIDS Care 2021;33:131–135 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Sheehan DM, Fennie KP, Mauck DE, Maddox LM, Lieb S, Trepka MJ: Retention in HIV care and viral suppression: Individual-and neighborhood-level predictors of racial/ethnic differences, Florida, 2015. AIDS Patient Care STDS 2017;31:167–175 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Crawford TN, Harris LM, Peyrani P: Examining age as a moderating effect on the relationship between alcohol use and viral suppression among women living with HIV. Women Health 2019;59:789–800 [DOI] [PubMed] [Google Scholar]
- 31. Ludema C, Cole SR, Eron Jr JJ, et al. : Impact of health insurance, ADAP, and income on HIV viral suppression among US women in the Women's Interagency HIV Study, 2006–2009. J Acquir Immune Defic Syndr 2016;73:307–312 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Cohen MH, Cook JA, Grey D, et al. : Medically eligible women who do not use HAART: The importance of abuse, drug use, and race. Am J Public Health 2004;94:1147–1151 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Sullivan KA, Messer LC, Quinlivan EB: Substance abuse, violence, and HIV/AIDS (SAVA) syndemic effects on viral suppression among HIV positive women of color. AIDS Patient Care STDS 2015;29(S1):S42–S48 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Olatosi BA, Probst JC, Stoskopf CH, Martin AB, Duffus WA: Patterns of engagement in care by HIV-infected adults: South Carolina, 2004–2006. AIDS 2009;23:725–730 [DOI] [PubMed] [Google Scholar]
- 35. Matheson FI, Moineddin R, Dunn JR, Creatore MI, Gozdyra P, Glazier RH: Urban neighborhoods, chronic stress, gender and depression. Soc Sci Med 2006;63:2604–2616 [DOI] [PubMed] [Google Scholar]
- 36. Chandran A, Edmonds A, Benning L, et al. : Longitudinal associations between neighborhood factors and HIV care outcomes in the WIHS. AIDS Behav 2020;24:2811–2818 [DOI] [PMC free article] [PubMed] [Google Scholar]