Abstract
The Department of Health and Human Services recommends drug resistance testing at linkage to HIV care. Because receipt and timing of testing are not well characterized, we examined testing patterns among persons with diagnosed HIV who are linked to care. Using surveillance data in six jurisdictions for persons aged ≥13 years with HIV infection diagnosed in 2013, we assessed the proportion receiving testing, and among these, the proportion receiving testing at linkage. Multivariable log-binomial regression modeling estimated associations between selected characteristics and receipt of testing (1) overall, and (2) at linkage among those tested. Of 9,408 persons linked to care, 66% received resistance testing, among whom 68% received testing at linkage. Less testing was observed among male persons who inject drugs (PWID), compared with men who have sex with men (adjusted prevalence ratio [aPR]: 0.88; 95% confidence interval [CI]: 0.81–0.97) and persons living in areas with population <500,000 compared with those in areas with population ≥2,500,000 (aPR: 0.88; CI: 0.84–0.93). In certain jurisdictions, testing was lower for persons with initial CD4 counts ≥500 cells/mm3, compared with those with CD4 counts <200 cells/mm3 (aPR range: 0.80–0.85). Of those tested, testing at linkage was lower among male PWID (aPR: 0.85; CI: 0.75–0.95) and, in some jurisdictions, persons with CD4 counts ≥500 cells/mm3 (aPR range: 0.63–0.73). Two-thirds of persons with diagnosed HIV who were linked to care received resistance testing, and most received testing at linkage as recommended. Improving receipt and timing of testing among male PWID, persons in less populous settings, and in all jurisdictions, regardless of CD4 count, may improve care outcomes.
Keywords: HIV, resistance testing, linkage
Background
Antiretroviral therapy (ART) has dramatically reduced morbidity and mortality among people living with diagnosed HIV infection around the world (Palella et al., 1998). However, sub-optimal adherence to ART can result in resistance to medications, thereby limiting the effectiveness of prescribed ART (Deeks et al., 2009) and perpetuating the transmission of a drug-resistant virus. Major viral drug resistance mutations have been reported among people with diagnosed HIV infection (Buchacz et al., 2015; Little et al., 2002; Palella et al., 1998; Ross et al., 2007; Weinstock et al., 2004; Wheeler et al., 2010; Yanik et al., 2012). Prevalence of drug resistance among persons with diagnosed HIV infection who have not initiated ART has been estimated to be 6–15% based on studies published in the previous decade (Buchacz et al., 2015; Little et al., 2002; Torian & Forgione, 2013; Weinstock et al., 2004). Transmitted drug resistance is more commonly detected among persons with recent HIV infection (Yanik et al., 2012) and may be higher among certain demographic and transmission risk groups, such as whites and gay, bisexual, and other men who have sex with men (MSM) (Buchacz et al., 2015; Ross et al., 2007; Weinstock et al., 2004; Yanik et al., 2012). People who take ART medications to which their HIV infection is resistant may not achieve viral suppression (Gill et al., 2010; Kantor et al., 2015; Little et al., 2002). Thus, identifying drug resistance before initiation of treatment may help ensure selection of appropriate ART regimens, viral suppression, and improved overall survival. Given the importance of early identification of drug resistance, the U.S. Department of Health and Human Services (HHS) began recommending drug resistance testing at entry to care among all persons who have received a diagnosis of HIV infection in 2007 (DHHS, Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents). However, the extent to which these recommendations are followed, including the timing with which testing occurs with respect to linkage to HIV care, is not fully known.
This analysis used data from the Centers for Disease Control and Prevention’s (CDC) U.S. National HIV Surveillance System (NHSS) to estimate the prevalence and timing of HIV drug resistance testing among persons with diagnosed HIV infection who have been linked to care, and to investigate demographic and geographic factors associated with (1) receipt of drug resistance testing, and (2) receipt of testing at linkage among those tested. These findings address knowledge gaps related to timing of HIV drug resistance testing and identify key populations in which testing might be less prevalent. These results may inform targeted interventions to increase resistance testing, thus improving HIV outcomes in the United States.
Methods
Population
We analyzed data from NHSS, a population-based surveillance system that collects demographic, risk, and clinical information on persons with diagnosed HIV infection in the United States. All fifty states, the District of Columbia, and six U.S. territories collect information on all persons with diagnosed HIV infection in the United States; this HIV case information is then reported to CDC. As a part of surveillance, health care providers and laboratories report demographic data, risk information, and CD4 and viral load test results to state and local HIV surveillance programs according to local public health disease reporting requirements. As a part of routine HIV care, healthcare providers also order HIV drug resistance tests at entry to care. Nucleotide sequence data generated through these tests are reported to state and local health departments in 27 jurisdictions that conduct Molecular HIV Surveillance, a component of NHSS. All HIV surveillance data are subsequently reported to CDC without personal identifiers.
For this analysis, NHSS data reported to CDC through December 2015 were restricted to persons aged ≥13 years with HIV infection diagnosed in 2013 who were linked to care within 3 months after diagnosis. Data were also restricted to persons residing in a jurisdiction with (1) complete laboratory reporting of CD4 and viral load tests to CDC (CDC, 2015) and (2) a drug resistance test reported within 12 months of HIV diagnosis for ≥50% of persons. The six jurisdictions included in the analysis were Los Angeles County, Michigan, New York, South Carolina, Texas, and Washington.
Measures
We defined linkage to care as having evidence of HIV laboratory testing (at least one CD4, viral load, or nucleotide sequence test) with a specimen collection date within 3 months after diagnosis. Among those who were linked to HIV care, we examined the proportion of persons who received HIV drug resistance testing, and calculated the number of months between linkage to care and HIV drug resistance testing for those who received testing. Persons who received HIV drug resistance testing within one month of linkage were considered to have received testing at linkage to care according to the HHS guidelines.
We assessed resistance testing by age at HIV diagnosis, race/ethnicity, sex, transmission category, and initial CD4 count at or after HIV diagnosis. Age at diagnosis was grouped as 13–24, 25–34, 35–44, 45–54, and ≥55 years. Race/ethnicity categories included: black/African American (henceforth referred to as black), Hispanic/Latino, and white; due to small numbers, we collapsed the American Indian/Alaska Native, Asian, Native Hawaiian/other Pacific Islander, multiple races, and other groups into a single category titled “other.” We combined sex and transmission category into a single variable with the following categories: males whose HIV infection was attributed to male-to-male sexual contact (men who have sex with men, or MSM), males whose HIV infection was attributed to both male-to-male sexual contact and injection drug use (MSM who inject drugs), males whose HIV infection was attributed to injection drug use (male persons who inject drugs, or male PWID), females whose HIV infection was attributed to injection drug use (female PWID), males whose HIV infection was attributed to heterosexual contact (heterosexual males), and females whose HIV infection was attributed to heterosexual contact (heterosexual females). When missing, information on transmission category was imputed based on methodology used to generate NHSS surveillance reports (Harrison, Kajese, Hall, & Song, 2008). Initial CD4 counts were categorized as <200, 200–349, 350–499, and ≥500 cells/mm3. Population of area of residence at diagnosis was derived from OMB standards in defining metropolitan statistical areas (MSAs) and Decennial 2010 Census estimates of the population of the MSA where the person resided at diagnosis (CDC, 2016; OMB, 2010).
Analytic methods
Descriptive statistics
Of those linked to HIV care within 3 months after diagnosis, we assessed the proportion of persons who received HIV drug resistance testing, and among these, the proportion who received testing at linkage (i.e., within one month of linkage). We used Mantel-Haenszel chi-square tests to identify differences in drug resistance testing patterns by demographic, geographic, and clinical characteristics, including age, race/ethnicity, transmission category, initial CD4 count at or after HIV diagnosis, area of residence at diagnosis, and population of area of residence at diagnosis.
Modeling
We used multivariable log-binomial regression models to estimate associations between these selected characteristics and (1) receipt of HIV drug resistance testing, as well as (2) receipt of HIV drug resistance testing at linkage among those who were tested, where two-sided P < .05 indicated statistical significance. Both models included age, race/ethnicity, transmission category, initial CD4 count at or after HIV diagnosis, area of residence at diagnosis, and population of area of residence at diagnosis as covariates. We also considered a two-way statistical interaction between initial CD4 count and area of residence at diagnosis to assess any differences in provider practices, by jurisdiction, with respect to resistance testing based on initial CD4 count. No other interaction terms were of interest, and thus, were not considered in this analysis. Both models excluded persons with transmission category of “other” and those who lived in an area of unknown population size at HIV diagnosis due to small numbers in these categories. We estimated adjusted prevalence ratios (aPRs) and corresponding 95% confidence intervals (CIs) and highlighted in the text statistically significant covariates with a PR ≥1.1 or ≤0.9 for at least one estimate. All statistical analyses were conducted using SAS 9.3 (Cary, NC).
Results
Descriptive statistics
Of 40,203 adults and adolescents who received a diagnosis of HIV infection in 2013, 11,261 (28%) resided in one of the six jurisdictions eligible for inclusion. Among these, 9408 (84%) persons were linked to HIV care within three months of HIV diagnosis and were included in the analysis. Overall, just over half of persons were under the age of 35 years at diagnosis, 37% were black, and 69% were MSM (Table 1). A majority of persons included in the analysis resided in Texas (37%) or New York (30%) at diagnosis. Most people (66%) resided in areas with populations ≥2,500,000.
Table 1.
Receipt of drug resistance testing among adults and adolescents with HIV infection diagnosed in 2013 who were linked to HIV care within three months after diagnosis, overall and by selected characteristics – National HIV Surveillance System, six jurisdictions.
| Total linked within 3 months |
Drug resistance testing | Drug resistance testing at linkage |
||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|
||||||
| n | (%) | n | (%) | p | n | (%) | p | |
| Overall | 9408 | (100) | 6229 | (66) | 4264 | (68) | ||
| Age, in years | 0.059 | 0.017 | ||||||
| 13–24 | 2038 | (22) | 1393 | (68) | 918 | (66) | ||
| 25–34 | 2943 | (31) | 1928 | (66) | 1310 | (68) | ||
| 35–44 | 1987 | (21) | 1309 | (66) | 917 | (70) | ||
| 45–54 | 1601 | (17) | 1068 | (67) | 752 | (70) | ||
| ≥55 | 839 | (9) | 531 | (63) | 367 | (69) | ||
| Race/ethnicity | <0.001 | 0.009 | ||||||
| Black/African American | 3436 | (37) | 2170 | (63) | 1423 | (66) | ||
| Hispanic/Latino | 3103 | (33) | 2136 | (69) | 1492 | (70) | ||
| White | 2238 | (24) | 1490 | (67) | 1058 | (71) | ||
| Other | 631 | (7) | 433 | (69) | 291 | (67) | ||
| Transmission category* | <0.001 | 0.472 | ||||||
| Male to male sexual contact | 6449 | (69) | 4381 | (68) | 3023 | (69) | ||
| Injection drug use (male) | 311 | (3) | 186 | (60) | 112 | (60) | ||
| Injection drug use (female) | 242 | (3) | 156 | (65) | 107 | (68) | ||
| Male to male sexual contact and injection drug use | 284 | (3) | 187 | (66) | 124 | (66) | ||
| Heterosexual contact (male) | 592 | (6) | 381 | (64) | 259 | (68) | ||
| Heterosexual contact (female) | 1518 | (16) | 930 | (61) | 634 | (68) | ||
| Other | 13 | 7 | 6 | |||||
| Initial CD4 count (cells/mm3) | <0.001 | <0.001 | ||||||
| Missing | 120 | 32 | 28 | |||||
| <200 | 2434 | (26) | 1719 | (71) | 1225 | (71) | ||
| 200–349 | 1819 | (20) | 1294 | (71) | 905 | (70) | ||
| 350–499 | 2009 | (22) | 1340 | (67) | 944 | (70) | ||
| ≥500 | 3026 | (33) | 1844 | (61) | 1162 | (63) | ||
| Area of residence at diagnosis | <0.001 | 0.567 | ||||||
| Los Angeles County | 1349 | (14) | 926 | (69) | 588 | (63) | ||
| Michigan | 643 | (7) | 495 | (77) | 363 | (73) | ||
| New York | 2861 | (30) | 2106 | (74) | 1549 | (74) | ||
| South Carolina | 636 | (7) | 370 | (58) | 162 | (44) | ||
| Texas | 3512 | (37) | 2078 | (59) | 1447 | (70) | ||
| Washington | 407 | (4) | 254 | (62) | 155 | (61) | ||
| Population of area at diagnosis† | <0.001 | 0.877 | ||||||
| Unknown | 25 | 9 | 7 | |||||
| <500,000 | 1226 | (13) | 720 | (59) | 478 | (66) | ||
| 500,000–2,499,999 | 1987 | (21) | 1251 | (63) | 883 | (71) | ||
| ≥2,500,000 | 6170 | (66) | 4249 | (69) | 2896 | (68) | ||
Other transmission category omitted due to small numbers.
Unknown population of area at diagnosis omitted due to small numbers.
Of those linked to care, 6229 (66%) received an HIV drug resistance test (Table 1). Prevalence of drug resistance testing varied significantly by race/ethnicity (P < 0.001), with the lowest level of testing among blacks; transmission category (P < 0.001), with less testing among male PWID and heterosexual females; area of residence at diagnosis (P < 0.001), with lower levels of testing in South Carolina, Texas, and Washington; initial CD4 count (P < 0.001), with less testing among persons with CD4 counts ≥500 cells/mm3; and population of area of residence at diagnosis (P < 0.001), with the lowest testing among persons residing in areas with population <2,500,000. Among the 6229 persons who received an HIV drug resistance test, 4264 (68%) received a test at linkage to care. Patterns in drug resistance testing at linkage differed by age (P = 0.017), with lower prevalence of testing among persons 13–24 years; race/ethnicity (P = 0.009), with lower prevalence of testing among blacks, and initial CD4 count at or after HIV diagnosis (P <0.001), with less prevalent testing among persons with a CD4 count ≥500 cells/mm3.
Modeling
Drug resistance testing
In the multivariable analysis, persons less likely to receive HIV drug resistance testing included persons aged 55 years and older (aPR: 0.90; 95% CI: 0.85, 0.96), compared with those 13–24 years of age; male PWID (aPR: 0.88; 95% CI: 0.81, 0.97), compared with MSM; and those living in areas with populations <500,000 (aPR: 0.88; 95% CI: 0.84, 0.93), compared with those living in more populous areas (Table 2). Lower prevalence of testing was also observed among persons with an initial CD4 count ≥500 cells/mm3, compared with those with a CD4 count <200 cells/mm3, but only in selected jurisdictions, including Michigan (aPR: 0.85; 95% CI: 0.77, 0.94), New York (aPR: 0.84; 95% CI: 0.80, 0.89), and Texas (aPR: 0.80; 95% CI: 0.75, 0.86). Less testing was also observed among persons with CD4 counts 350–499 cells/mm3 in Texas (aPR: 0.85; 95% CI: 0.79, 0.92).
Table 2.
Factors associated with receipt of drug resistance testing among adults and adolescents with HIV diagnosed in 2013 who were linked to HIV care within three months after diagnosis, overall and by selected characteristics – National HIV Surveillance System, six jurisdictions.
| Characteristics | Resistance testing | No resistance testing | aPR* (95% CI) | p |
|---|---|---|---|---|
| Age, in years | ||||
| 13–24 | 1379 | 619 | Ref | |
| 25–34 | 1916 | 980 | 0.94 (0.91, 0.97) | 0.002 |
| 35–44 | 1303 | 659 | 0.95 (0.91, 0.99) | 0.008 |
| 45–54 | 1062 | 516 | 0.95 (0.91, 0.99) | 0.024 |
| ≥55 | 521 | 297 | 0.90 (0.85, 0.96) | <0.001 |
| Race/ethnicity | ||||
| Black/African American | 2147 | 1227 | 0.93 (0.89, 0.96) | <0.001 |
| Hispanic/Latino | 2127 | 933 | 1.02 (0.98, 1.05) | 0.403 |
| White | 1477 | 715 | Ref | |
| Other | 430 | 196 | 0.99 (0.93, 1.05) | 0.710 |
| Transmission category† | ||||
| Male to male sexual contact | 4351 | 1993 | Ref | |
| Injection drug use (male) | 184 | 120 | 0.88 (0.81, 0.97) | 0.007 |
| Injection drug use (female) | 156 | 80 | 0.95 (0.87, 1.03) | 0.220 |
| Male to male sexual contact and injection drug use | 187 | 95 | 0.99 (0.91, 1.07) | 0.790 |
| Heterosexual contact (male) | 378 | 207 | 0.96 (0.91, 1.02) | 0.186 |
| Heterosexual contact (female) | 925 | 575 | 0.94 (0.90, 0.98) | 0.003 |
| Initial CD4 counts in each area of residence at diagnosis (cells/mm3)‡ | ||||
| Los Angeles County | ||||
| <200 | 224 | 103 | Ref | |
| 200–349 | 196 | 61 | 1.10 (0.99, 1.21) | 0.068 |
| 350–499 | 199 | 81 | 1.02 (0.92, 1.13) | 0.682 |
| ≥500 | 305 | 153 | 0.96 (0.87, 1.06) | 0.435 |
| Michigan | ||||
| <200 | 153 | 30 | Ref | |
| 200–349 | 97 | 30 | 0.93 (0.83, 1.04) | 0.183 |
| 350–499 | 89 | 17 | 1.01 (0.91, 1.12) | 0.898 |
| ≥500 | 151 | 56 | 0.85 (0.77, 0.94) | 0.001 |
| New York | ||||
| <200 | 543 | 148 | Ref | |
| 200–349 | 459 | 124 | 0.98 (0.93, 1.04) | 0.479 |
| 350–499 | 471 | 153 | 0.94 (0.89, 0.99) | 0.032 |
| ≥500 | 617 | 301 | 0.84 (0.80, 0.89) | <0.001 |
| South Carolina | ||||
| <200 | 114 | 75 | Ref | |
| 200–349 | 74 | 38 | 1.05 (0.89, 1.25) | 0.547 |
| 350–499 | 82 | 57 | 0.95 (0.79, 1.13) | 0.538 |
| ≥500 | 99 | 87 | 0.87 (0.73, 1.03) | 0.110 |
| Texas | ||||
| <200 | 619 | 323 | Ref | |
| 200–349 | 411 | 247 | 0.94 (0.87, 1.02) | 0.129 |
| 350–499 | 439 | 323 | 0.85 (0.79, 0.92) | <0.001 |
| ≥500 | 589 | 516 | 0.80 (0.75, 0.86) | <0.001 |
| Washington | ||||
| <200 | 60 | 32 | Ref | |
| 200–349 | 54 | 24 | 1.06 (0.86, 1.30) | 0.604 |
| 350–499 | 59 | 34 | 0.95 (0.77, 1.18) | 0.659 |
| ≥500 | 80 | 59 | 0.88 (0.72, 1.08) | 0.229 |
| Population of area at diagnosis§ | ||||
| <500,000 | 713 | 490 | 0.88 (0.84, 0.93) | <0.001 |
| 500,000–2,499,999 | 1240 | 716 | 0.94 (0.90, 0.99) | 0.008 |
| ≥2,500,000 | 4229 | 1864 | Ref |
PR: adjusted prevalence ratio.
Other transmission category omitted due to small numbers.
Unknown CD4 counts omitted due to small numbers.
Unknown population of area at diagnosis omitted due to small numbers.
Drug resistance testing at linkage
Even among those who received drug resistance testing, male PWIDs were significantly less likely to receive testing at linkage to care (aPR: 0.85; 95% CI: 0.75, 0.95), compared with MSM (Table 3). Compared with persons with initial CD4 counts <200 cells/mm3, prevalence of HIV drug resistance testing at linkage was significantly lower among persons with initial CD4 counts ≥500 cells/mm3 in certain jurisdictions, including Los Angeles County (aPR: 0.73; 95% CI: 0.64, 0.83), South Carolina (aPR: 0.69; 95% CI: 0.51, 0.94), and Washington (aPR: 0.63, 95% CI: 0.47, 0.83). Levels of drug resistance testing at linkage were also lower among those who received a diagnosis in New York but the adjusted prevalence ratio was not ≤0.9 (aPR: 0.92; 95% CI: 0.86, 0.99). Further, point estimates indicated lower levels of testing among persons with higher CD4 counts in selected jurisdictions, such as Los Angeles County, South Carolina, and Washington, but not all of these associations were statistically significant. This may be due to limited sample size in these jurisdictions. Finally, those living in areas with a population size between 500,000–2,499,999 were more likely to receive testing at linkage (aPR: 1.12; 95% CI: 1.07, 1.17), compared with those in areas with population ≥2,500,000.
Table 3.
Factors associated with receipt of drug resistance testing at linkage to HIV care among adults and adolescents with HIV diagnosed in 2013 who were linked to HIV care within three months after diagnosis and received drug resistance testing at or after linkage, overall and by selected characteristics – National HIV Surveillance System, six jurisdictions.
| Resistance testing | No resistance testing | aPR* (95% CI) | p | |
|---|---|---|---|---|
| Age, in years | ||||
| 13–24 | 907 | 472 | Ref | |
| 25–34 | 1301 | 615 | 1.03 (0.98, 1.08) | 0.304 |
| 35–44 | 911 | 392 | 1.05 (1.00, 1.11) | 0.059 |
| 45–54 | 747 | 316 | 1.04 (0.98, 1.10) | 0.200 |
| ≥55 | 358 | 163 | 1.00 (0.93, 1.07) | 0.997 |
| Race/ethnicity | ||||
| Black/African American | 1405 | 742 | 0.93 (0.89, 0.98) | 0.003 |
| Hispanic/Latino | 1484 | 643 | 0.96 (0.92, 1.00) | 0.057 |
| White | 1046 | 431 | Ref | |
| Other | 289 | 141 | 0.92 (0.86, 0.99) | 0.029 |
| Transmission category† | ||||
| Male to male sexual contact | 2997 | 1354 | Ref | |
| Injection drug use (male) | 111 | 73 | 0.85 (0.75, 0.95) | 0.006 |
| Injection drug use (female) | 107 | 49 | 0.97 (0.87, 1.08) | 0.565 |
| Male to male sexual contact and injection drug use | 123 | 64 | 0.96 (0.87, 1.06) | 0.422 |
| Heterosexual contact (male) | 257 | 122 | 0.97 (0.90, 1.04) | 0.370 |
| Heterosexual contact (female) | 629 | 296 | 0.99 (0.94, 1.04) | 0.688 |
| Initial CD4 counts in each area of residence at diagnosis (cells/mm3)‡ | ||||
| Los Angeles County | ||||
| <200 | 164 | 60 | Ref | |
| 200–349 | 126 | 70 | 0.88 (0.77, 1.00) | 0.053 |
| 350–499 | 136 | 63 | 0.94 (0.83, 1.06) | 0.310 |
| ≥500 | 161 | 144 | 0.73 (0.64, 0.83) | <0.001 |
| Michigan | ||||
| <200 | 114 | 40 | Ref | |
| 200–349 | 71 | 26 | 1.02 (0.87, 1.18) | 0.841 |
| 350–499 | 62 | 27 | 0.94 (0.79, 1.11) | 0.442 |
| ≥500 | 112 | 39 | 1.02 (0.89, 1.17) | 0.750 |
| New York | ||||
| <200 | 409 | 135 | Ref | |
| 200–349 | 348 | 114 | 1.01 (0.94, 1.08) | 0.832 |
| 350–499 | 358 | 113 | 1.01 (0.94, 1.09) | 0.721 |
| ≥500 | 426 | 191 | 0.92 (0.86, 0.99) | 0.029 |
| South Carolina | ||||
| <200 | 60 | 54 | Ref | |
| 200–349 | 30 | 44 | 0.75 (0.54, 1.04) | 0.087 |
| 350–499 | 34 | 48 | 0.76 (0.56, 1.04) | 0.08<5 |
| ≥500 | 37 | 62 | 0.69 (0.51, 0.94) | 0.019 |
| Texas | ||||
| <200 | 432 | 187 | Ref | |
| 200–349 | 292 | 119 | 1.02 (0.95, 1.11) | 0.565 |
| 350–499 | 318 | 121 | 1.06 (0.98, 1.14) | 0.165 |
| ≥500 | 386 | 203 | 0.95 (0.88, 1.02) | 0.171 |
| Washington | ||||
| <200 | 44 | 16 | Ref | |
| 200–349 | 38 | 16 | 0.97 (0.77, 1.22) | 0.813 |
| 350–499 | 35 | 24 | 0.81 (0.63, 1.05) | 0.112 |
| ≥500 | 37 | 43 | 0.63 (0.47, 0.83) | 0.001 |
| Population of area of residence at diagnosis§ | ||||
| <500,000 | 471 | 242 | 1.04 (0.99, 1.11) | 0.145 |
| 500,000–2,499,999 | 872 | 367 | 1.12 (1.07, 1.17) | <0.001 |
| ≥2,500,000 | 2880 | 1348 | Ref |
aPR: adjusted prevalence ratio.
Other transmission category omitted due to small numbers.
Unknown CD4 counts omitted due to small numbers.
Unknown population of area at diagnosis omitted due to small numbers.
Discussion
This analysis examined the prevalence of HIV drug resistance testing among persons with diagnosed HIV infection in selected jurisdictions in the United States and sought to identify factors associated with (1) drug resistance testing, and (2) receipt of testing at linkage among those tested. Overall, nearly two-thirds of persons with diagnosed HIV infection who were linked to care received drug resistance testing, among whom almost 70% received testing at linkage. The prevalence of drug resistance testing was lower among male PWID, persons living in less populous areas, and those with higher initial CD4 counts in certain jurisdictions included in this analysis. Among persons who received drug resistance testing, testing at linkage was lower among male PWID and persons with higher initial CD4 counts in selected jurisdictions. Based on these findings, implementation of timely resistance testing among these groups may improve HIV care outcomes.
Identifying drug resistance at the time of diagnosis facilitates selection of ART medications to which the HIV strain is sensitive. This can reduce the time to viral suppression and transmission of drug-resistant HIV strains to others. Improving drug resistance testing is particularly important for PWID, among whom prevalence of risk behaviors associated with HIV transmission, such as sharing injection equipment and engaging in condomless sex, may be quite high (Spiller et al., 2015). Reasons for lack of drug resistance testing or delayed testing in this group are unknown, although a recent report showed that nearly a third of PWID in selected U.S. cities reported not having health insurance and nearly 80% were living at or below the federal poverty level (Spiller et al., 2015). Lack of health insurance access and lower socioeconomic status may be potential barriers to receiving medical care and laboratory testing in a timely manner within this population (Moneyham et al., 2010; Mueller, Patil, & Boilesen, 1998; Public Financing and Delivery of HIV/AIDS Care: Securing the Legacy of Ryan White, 2005). There may also be low awareness of free or subsidized HIV care through the Ryan White HIV/AIDS Program or through federally qualified health centers.
In addition, previous work has shown that non-HIV specialists are less likely to order resistance testing compared with HIV specialists (Ocfemia et al., 2016). Since there may be greater injection drug use in smaller, less populous areas (Zibbell et al., 2015), fewer and less specialized healthcare services in these settings may necessitate greater efforts to engage HIV-infected PWID in medical care. This may particularly be true since overall healthcare use may be lower in less populous settings (Mueller et al., 1998; Reschovsky & Staiti, 2005), potentially due to fewer available resources for comprehensive care, costs associated with healthcare, and lack of insurance coverage (Lu, Samuels, Kletke, & Whitler, 2010; Reschovsky & Staiti, 2005). With HIV-infected PWID being less likely to be engaged in care and virally suppressed (Bradley et al., 2014), prioritizing drug resistance testing at linkage and implementing measures to retain persons in care over time in this population may improve provision of appropriate ART regimens. This is especially important in light of the recent HIV outbreak among PWID in southern Indiana (Conrad et al., 2015), given the high HIV transmission potential through injection drug use and the substantial rise in the number of deaths attributed to opioid use in the United States (Rudd, Aleshire, Zibbell, & Gladden, 2016).
Despite HHS recommendations on drug resistance testing at entry to care (DHHS, Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents.), these data showed significantly lower levels of receipt and timing of drug resistance testing among persons with higher initial CD4 counts at or after HIV diagnosis, but only in some jurisdictions. These differences may indicate geographic differences in care practices related to drug resistance testing. For instance, providers in certain jurisdictions may prioritize resistance testing for persons with low initial CD4 counts. Further, in persons for whom HIV treatment may need to be delayed due to medical reasons such as TB co-infection (DHHS, Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents), providers may also decide to postpone resistance testing. Even if treatment is delayed, however, resistance testing should be conducted at entry to care, as drug resistance mutations may decay in the absence of drug pressure, potentially affecting treatment success if mutations are present at low levels that are no longer detectable (Johnson et al., 2008). Care practices related to receipt and timing of testing may also be influenced by variation in cost coverage of drug resistance testing, especially in regards to differential Medicaid expansion by state. Nevertheless, emphasizing the need to initiate drug resistance testing at entry to care across all jurisdictions, regardless of initial CD4 count, is vital to ensuring successful treatment outcomes.
There were also lower levels of resistance testing observed, regardless of initial CD4 count, in certain jurisdictions, specifically Los Angeles, South Carolina, and Washington. These differences by jurisdiction may be explained by other factors not accounted for in this analysis, including variation in cost coverage of drug resistance testing or differential completeness of laboratory reporting of viral genetic sequences to state and local health departments, and thus, to CDC. In any case, increasing testing in all jurisdictions will help in identifying and addressing drug resistance in a timely manner. In addition, state and local health departments should continue working to increase completeness of laboratory reporting of nucleotide sequence data and improve the quality of reported data on drug resistance testing.
This analysis is subject to several limitations. First, because these data were limited to adults and adolescents with HIV infection diagnosed in 2013, temporal trends in drug resistance testing could not be assessed. Only jurisdictions with complete laboratory reporting and adequate reporting of drug resistance testing were included in this analysis, so the results may not be generalizable to all people with diagnosed HIV infection in the United States. The results presented here may be an underestimate of prevalence of HIV drug resistance testing if laboratory reporting of resistance test results was incomplete. Finally, recommendations to start HIV treatment regardless of CD4 count and conduct resistance testing at entry to care are recent with respect to the analysis period and may have impacted a healthcare provider’s decision to order resistance testing. Thus, future analyses using more recent data may yield higher estimates for prevalence of resistance testing.
In summary, these findings demonstrate that, in the six jurisdictions included in the analysis, a majority of persons with diagnosed HIV infection who are linked to care received drug resistance testing; the timing of drug resistance testing suggests that most providers order testing at linkage to care as recommended. Differences in receipt and timing of testing underscores the need to increase drug resistance testing among male PWID, those who live in less populous areas, and in all jurisdictions, regardless of initial CD4 count. Factors influencing lack of or delayed resistance testing should be further explored and addressed, with interventions tailored based on varying characteristics and needs, and accounting for differences in care practices and cost coverage by jurisdiction. Addressing such barriers to testing may increase the number of persons who receive appropriate treatment, thus reducing disparities in important HIV outcomes over time.
Acknowledgments
The authors acknowledge the work of state and local health departments in collecting and submitting HIV sequencing data to CDC. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Footnotes
Disclosure statement
No potential conflict of interest was reported by the authors.
References
- Bradley H, Hall HI, Wolitski RJ, Van Handel MM, Stone AE, LaFlam M, Valleroy LA. Vital signs: HIV diagnosis, care, and treatment among persons living with HIV–United States, 2011. Morbidity and Mortality Weekly Report. 2014;63(47):1113–1117. [PMC free article] [PubMed] [Google Scholar]
- Buchacz K, Young B, Palella FJ, Jr, Armon C, Brooks JT HIV Outpatient Study (HOPS) investigators. Trends in use of genotypic resistance testing and frequency of major drug resistance among antiretroviral-naive persons in the HIV outpatient study, 1999–2011. Journal of Antimicrobial Chemotherapy. 2015;70(8):2337–2346. doi: 10.1093/jac/dkv120. [DOI] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance dataUnited States and 6 dependent areas, 2014. HIV Surveillance Supplemental Report 2016. 2015;21(4) http://www.cdc.gov/hiv/library/reports/surveillance. [Google Scholar]
- Centers for Disease Control and Prevention. Diagnoses of HIV infection in the United States and dependent areas, 2014. HIV Surveillance Report, 2014. 2016;26 http://www.cdc.gov/hiv/library/reports/surveillance/ [Google Scholar]
- Conrad C, Bradley HM, Broz D, Buddha S, Chapman EL, Galang RR, Duwve JM. Community outbreak of HIV infection linked to injection drug use of Oxymorphone–Indiana, 2015. Morbidity and Mortality Weekly Report. 2015;64(16):443–444. [PMC free article] [PubMed] [Google Scholar]
- Deeks SG, Gange SJ, Kitahata MM, Saag MS, Justice AC, Hogg RS North American AIDS Cohort Collaboration on Research and Design. Trends in multidrug treatment failure and subsequent mortality among antiretroviral therapy-experienced patients with HIV infection in North America. Clinical Infectious Diseases. 2009;49(10):1582–1590. doi: 10.1086/644768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Department of Health and Human Services. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Retrieved from https://aidsinfo.nih.gov/contentfiles/lvguidelines/adultandadolescentgl.pdf. [PubMed]
- Gill VS, Lima VD, Zhang W, Wynhoven B, Yip B, Hogg RS, Harrigan PR. Improved virological outcomes in British Columbia concomitant with decreasing incidence of HIV type 1 drug resistance detection. Clinical Infectious Diseases. 2010;50(1):98–105. doi: 10.1086/648729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harrison KM, Kajese T, Hall HI, Song R. Risk factor redistribution of the national HIV/AIDS surveillance data: An alternative approach. Public Health Representative. 2008;123(5):618–627. doi: 10.1177/003335490812300512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Institute of Medicine. Public financing and delivery of HIV/AIDS care: Securing the legacy of Ryan White. Washington, DC: The National Academies Press; 2005. [Google Scholar]
- Johnson JA, Li JF, Wei X, Lipscomb J, Irlbeck D, Craig C, Heneine W. Minority HIV-1 drug resistance mutations are present in antiretroviral treatment-naive populations and associate with reduced treatment efficacy. PLoS Medicine. 2008;5(7):e158. doi: 10.1371/journal.pmed.0050158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kantor R, Smeaton L, Vardhanabhuti S, Hudelson SE, Wallis CL, Tripathy S AIDS Clinical Trial Group (ACTG) A5175 Study Team. Pretreatment HIV drug resistance and HIV-1 subtype C are independently associated with virologic failure: Results from the multinational PEARLS (ACTG A5175) clinical trial. Clinical Infectious Diseases. 2015;60(10):1541–1549. doi: 10.1093/cid/civ102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Little SJ, Holte S, Routy JP, Daar ES, Markowitz M, Collier AC, Richman DD. Antiretroviral-drug resistance among patients recently infected with HIV. New England Journal of Medicine. 2002;347(6):385–394. doi: 10.1056/NEJMoa013552. [DOI] [PubMed] [Google Scholar]
- Lu N, Samuels ME, Kletke PR, Whitler ET. Rural-urban differences in health insurance coverage and patterns among working-age adults in Kentucky. The Journal of Rural Health. 2010;26(2):129–138. doi: 10.1111/j.1748-0361.2010.00274.x. [DOI] [PubMed] [Google Scholar]
- Moneyham L, McLeod J, Boehme A, Wright L, Mugavero M, Seal P, Kempf MC. Perceived barriers to HIV care among HIV-infected women in the deep south. Journal of the Association of Nurses in AIDS Care. 2010;21(6):467–477. doi: 10.1016/j.jana.2010.03.003. [DOI] [PubMed] [Google Scholar]
- Mueller KJ, Patil K, Boilesen E. The role of uninsurance and race in healthcare utilization by rural minorities. Health Services Research. 1998;33(3 Pt 1):597–610. [PMC free article] [PubMed] [Google Scholar]
- Ocfemia M, Oster A, Valverde E, Yungfeng T, Beer L, Hernandez A, Weiser J. HIV drug resistance testing among patients new to HIV care in the United States; Conference on Retroviruses and Opportunistic Infections; Boston, Massachusetts. 2016. [Google Scholar]
- Office of Management and Budget. Standards for defining metropolitan and micropolitan statistical areas. Federal Register. 2010;75(123):37246–39052. http://go.usa.gov/vSpG. [Google Scholar]
- Palella FJ, Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, Holmberg SD. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV outpatient study investigators. New England Journal of Medicine. 1998;338(13):853–860. doi: 10.1056/NEJM199803263381301. [DOI] [PubMed] [Google Scholar]
- Reschovsky JD, Staiti AB. Access and quality: Does rural America lag behind? Health Affairs. 2005;24(4):1128–1139. doi: 10.1377/hlthaff.24.4.1128. [DOI] [PubMed] [Google Scholar]
- Ross L, Lim ML, Liao Q, Wine B, Rodriguez AE, Weinberg W, Shaefer M. Prevalence of antiretroviral drug resistance and resistance-associated mutations in antiretroviral therapy-naive HIV-infected individuals from 40 United States cities. HIV Clinical Trials. 2007;8(1):1–8. doi: 10.1310/hct0801-1. [DOI] [PubMed] [Google Scholar]
- Rudd RA, Aleshire N, Zibbell JE, Gladden RM. Increases in drug and opioid overdose deaths - United States, 2000–2014. MMWR. Morbidity and Mortality Weekly Report. 2016;64(50–51):1378–1382. doi: 10.15585/mmwr.mm6450a3. [DOI] [PubMed] [Google Scholar]
- Spiller MW, Broz D, Wejnert C, Nerlander L, Paz-Bailey G. HIV infection and HIV-associated behaviors among persons who inject drugs–20 cities, United States, 2012. MMWR. Morbidity and Mortality Weekly Report. 2015;64(10):270–275. [PMC free article] [PubMed] [Google Scholar]
- Torian LV, Forgione LA. Transmitted antiretroviral drug resistance in New York city, 2006–2010: The first five years of routine genotype surveillance. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2013;63(3):e119–e122. doi: 10.1097/QAI.0b013e31828d2fc1. [DOI] [PubMed] [Google Scholar]
- Weinstock HS, Zaidi I, Heneine W, Bennett D, Garcia-Lerma JG, Douglas JM, Jr, Kaplan JE. The epidemiology of antiretroviral drug resistance among drug-naive HIV-1-infected persons in 10 US cities. The Journal of Infectious Diseases. 2004;189(12):2174–2180. doi: 10.1086/420789. [DOI] [PubMed] [Google Scholar]
- Wheeler WH, Ziebell RA, Zabina H, Pieniazek D, Prejean J, Bodnar UR Variant, Atypical, and Resistant HIV Surveillance Group. Prevalence of transmitted drug resistance associated mutations and HIV-1 subtypes in new HIV-1 diagnoses, U.S.-2006. AIDS (London, England) 2010;24(8):1203–1212. doi: 10.1097/QAD.0b013e3283388742. [DOI] [PubMed] [Google Scholar]
- Yanik EL, Napravnik S, Hurt CB, Dennis A, Quinlivan EB, Sebastian J, Eron JJ. Prevalence of transmitted antiretroviral drug resistance differs between acutely and chronically HIV-infected patients. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2012;61(2):258–262. doi: 10.1097/QAI.0b013e3182618f05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zibbell JE, Iqbal K, Patel RC, Suryaprasad A, Sanders KJ, Moore-Moravian L, Holtzman D. Increases in hepatitis C virus infection related to injection drug use among persons aged ≤30 years - Kentucky, Tennessee, Virginia, and West Virginia, 2006–2012. MMWR. Morbidity and Mortality Weekly Report. 2015;64(17):453–458. [PMC free article] [PubMed] [Google Scholar]