Why was the cohort set up?
The National Institutes of Health established the Women’s Interagency HIV Study (WIHS) in 1993 to study the impact and progression of HIV infection among women in response to the rising number of AIDS cases and the relative paucity of clinical, behavioural and epidemiological data in this population. Women now comprise more than 50% of people with HIV (PWH) worldwide.1 The WIHS is the largest and oldest ongoing prospective cohort study of women with and at risk for HIV infection in the world, and remains the leading study to document the experience of women with HIV (WWH) in the United States.
The cohort’s scope of data collection and research has expanded considerably since the two previous publications that described the methodology and profile of the study in its early years.2,3 Current areas of research include, but are not limited to: HIV-related comorbidities; epidemiological methods; genetics; frailty and ageing; behavioural and social determinants of health; pharmacology; and studies related to pursuit of an HIV cure. The purpose of this paper is to update researchers on the substantial expansion of the cohort, its methods and the available data.
Who is in the cohort?
The six original cohort sites were in Brooklyn, NY; the Bronx/Manhattan, NY; Washington, DC; Chicago, IL; San Francisco, CA; and Los Angeles, CA. Since 1997, the WIHS Data Management and Analysis Center (WDMAC) has been located in Baltimore, MD. Four southern sites were added in 2013: Chapel Hill, NC; Atlanta, GA; Birmingham, AL/Jackson, MS; and Miami, FL. The Los Angeles site discontinued active follow-up in 2013.
The cohort enrolled 4982 women who at baseline were: either HIV-seropositive (3677); or HIV-seronegative (1305; 26 of them acquired HIV during follow-up) with a history of sexually transmitted infections (STIs) or behavioural or demographic characteristics that increased their risk of acquiring HIV. Enrolment occurred during four waves: 1994-95 (2054 HIV+; 569 HIV-); 2001–02 (737 HIV+, 406 HIV-), 2011-12 (276 HIV+; 95 HIV-); and 2013-15 (610 HIV+; 235 HIV-). Eligibility criteria were similar throughout the study, but varied from wave to wave to increase representativeness of the cohort.2,3
The most recent enrolment wave during 2013–15 (Wave 4) enrolled women between the ages of 25 and 60 years. WWH were required to have documentation of a reactive HIV serology and confirmatory test. Women who used anti-retroviral therapy (ART) were eligible only if all of their regimens met criteria for highly active antiretroviral therapy (HAART), as defined by the US consensus treatment recommendations.4 Those who used non-HAART regimens were eligible only if use occurred during pregnancy or HIV pre- or post-exposure prophylaxis. Documentation of CD4+ T cell counts and HIV RNA quantification before initiation of HAART was required.
HIV-seronegative women were eligible for Wave 4 if they had at least one high-risk exposure in the preceding 5 years [e.g. STI diagnosis; sex without a condom with three or more men; sex with a condom with six or more men; trading sex; sex with an HIV-seropositive man; injection drug use (IDU) or use of crack cocaine, cocaine, heroin or methamphetamine; or any partner who had any of the previously mentioned risk characteristics]. Wave 3 enrolled women between the ages of 30 and 55 years, but eligibility criteria were otherwise similar to those of Wave 2.3 During Wave 4, investigators used word of mouth and flyers to recruit participants from community-based organizations and medical, STI and family planning clinics.
Since the beginning of the study in 1993, 1268 participants have died, 130 have withdrawn from the study, 806 were discontinued for administrative reasons (such as termination of a site’s participation or decreased funding that required sites to decrease enrolment), and 415 have been lost to follow-up. (Table 1) As of October 2016, the WIHS is actively following 2363 women.
Table 1.
Disposition of participants in the Women’s Interagency HIV Study, by enrolment wave and baseline HIV statusa
| Wave 1 |
Wave 2 |
Wave 3 |
Wave 4 |
All Waves |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | |
| N = 569 | N = 2054 | N = 406 | N = 737 | N = 95 | N = 276 | N = 235 | N = 610 | N = 1305 | N = 3677 | |
| Activeb | 207 (36) | 508 (25) | 217 (53) | 354 (48) | 78 (82) | 209 (76) | 214 (91) | 576 (94) | 716 (55) | 1647 (45) |
| Deceasedc | 95 (17) | 1016 (49) | 16 (4) | 113 (15) | 2 (2) | 13 (5) | 5 (2) | 8 (1) | 118 (9) | 1150 (31) |
| Withdrew from the study | 26 (5) | 62 (3) | 15 (4) | 21 (3) | 3 (3) | 1 (< 1) | 1 (< 1) | 1 (< 1) | 45 (3) | 85 (2) |
| Administrative disenrolmentd | 144 (25) | 319 (16) | 116 (29) | 177 (24) | 6 (6) | 38 (14) | 3 (1) | 3 (< 1) | 269 (21) | 537 (15) |
| Lost to follow-upe | 97 (17) | 149 (7) | 42 (10) | 72 (10) | 6 (6) | 15 (5) | 12 (5) | 22 (4) | 157 (12) | 258 (7) |
Dates of enrolment were: Wave 1 = 1994–95; Wave 2 = 2001–02; Wave 3 = 2011–12; Wave 4 = 2013–15.
Participants were considered ‘active’ if they attended at least one visit between October 2015 and September 2016. Among active participants who were HIV-seronegative at baseline, 14 seroconverted during WIHS follow-up (11 from Wave 1, two from Wave 2 and one from Wave 3).
Among deceased participants who were HIV-seronegative at baseline, eight seroconverted during WIHS follow-up (six from Wave 1 and two from Wave 2) and two seroconverted after last WIHS visit (one from Wave 1 and one from Wave 2).
Examples of reasons for administrative disenrollment include termination of a site’s participation or decreased funding that required sites to decrease enrollment. Among administratively disenrolled participants who were HIV-seronegative at baseline, 1 seroconverted during WIHS follow-up (Wave 2).
Among lost to follow-up participants who were HIV-seronegative at baseline, one seroconverted during WIHS follow-up Wave 1.
A unique aspect of the WIHS is its enrolment of demographically similar HIV-seronegative women. The age and racial/ethnic distributions of seronegative participants are similar to those of WWH in the cohort (Black 72%, White 11%, Hispanic 14% and Other 3%) who, in turn, are generally representative by race/ethnicity of WWH in the USA (Black 59%, White 17%, Hispanic 19% and Other 5%). Tables 2 and 3 outline some key demographic, behavioural and clinical characteristics of active participants as of 2016 (visit 44). Most participants are poor; 64% of HIV-seropositive and 56% of HIV-seronegative participants report an annual household income of $18 000 or less. About one-third (33% and 31%, respectively, of HIV+ and HIV- participants) have attained less than a high school education. Almost one-third (31%) of HIV+ participants have reported a history of clinical AIDS; as expected, women enrolled earlier are more likely to have had an AIDS condition than those enrolled later (57%, 30%, 18% and 14%, respectively, for Waves 1, 2, 3 and 4). Almost all (97%) of WWH have received some antiretroviral therapy (ART), and 89% reported taking ART at the most recent visit. Nevertheless, a substantial minority (31%) continue to have unsuppressed HIV viral loads, reflecting recognized deficiencies in the continuum of HIV care.5
Table 2.
Demographic and behavioural characteristics of currently active participants in the Women’s Interagency HIV Study, by enrolment wavea and HIV status at last WIHS visit
| Wave 1 |
Wlave 2 |
Wave 3 |
Wave 4 |
All Waves |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | |
| N = 196 | N = 519 | N = 215 | N = 356 | N = 77 | N = 210 | N = 214 | N = 576 | N = 702 | N = 1661 | |
| Demographicsb | ||||||||||
| Median age (IQR) | 56 (49, 61) | 56 (52, 61) | 43 (37, 50) | 48 (42, 52) | 53 (47, 58) | 49 (42, 53) | 46 (37, 53) | 47 (39, 53) | 49 (41, 55) | 51 (44, 56) |
| Race/ethnicity | ||||||||||
| Black, non-Hispanic | 128 (65) | 308 (59) | 135 (63) | 245 (69) | 65 (84) | 163 (78) | 178 (83) | 477 (83) | 506 (72) | 1193 (72) |
| White, non-Hispanic | 19 (10) | 91 (18) | 13 (6) | 21 (6) | 2 (3) | 18 (9) | 19 (9) | 52 (9) | 53 (8) | 182 (11) |
| Hispanic/Latina | 38 (19) | 102 (20) | 54 (25) | 72 (20) | 4 (5) | 21 (10) | 12 (6) | 40 (7) | 108 (15) | 235 (14) |
| Otherc | 11 (6) | 18 (3) | 13 (6) | 18 (5) | 6 (8) | 8 (3) | 5 (2) | 7 (1) | 35 (5) | 51 (3) |
| Heterosexual | 163 (83) | 452 (87) | 170 (79) | 320 (90) | 66 (86) | 184 (88) | 178 (83) | 536 (93) | 577 (82) | 1492 (90) |
| Less than high school education | 61 (31) | 170 (33) | 75 (35) | 120 (34) | 21 (28) | 77 (37) | 62 (29) | 178 (31) | 219 (31) | 545 (33) |
| Married or partnered | 55 (28) | 129 (25) | 68 (32) | 105 (29) | 20 (26) | 56 (27) | 54 (25) | 136 (24) | 197 (28) | 426 (26) |
| Unstable housing | 6 (3) | 16 (3) | 5 (2) | 12 (3) | 8 (10) | 8 (4) | 14 (7) | 29 (5) | 33 (5) | 65 (4) |
| Annual household income ≤ $18 000 | 100 (51) | 291 (56) | 104 (48) | 203 (57) | 58 (75) | 141 (67) | 133 (62) | 424 (74) | 395 (56) | 1059 (64) |
| Employed | 67 (34) | 145 (28) | 92 (43) | 152 (43) | 25 (32) | 68 (32) | 87 (41) | 208 (36) | 271 (39) | 573 (35) |
| Health insurance | ||||||||||
| Any | 169 (86) | 490 (94) | 185 (86) | 341 (96) | 73 (95) | 201 (96) | 131 (61) | 544 (94) | 558 (79) | 1576 (95) |
| Medicaid | 103 (53) | 331 (64) | 120 (56) | 242 (68) | 55 (71) | 149 (71) | 71 (33) | 272 (47) | 349 (50) | 994 (60) |
| Medicare | 47 (24) | 159 (31) | 17 (8) | 57 (16) | 15 (19) | 33 (16) | 24 (11) | 99 (17) | 103 (15) | 348 (21) |
| Commercial/private | 48 (24) | 131 (25) | 52 (24) | 74 (21) | 11 (14) | 35 (17) | 42 (20) | 100 (17) | 153 (22) | 340 (20) |
| Military/student/other | 68 (35) | 174 (34) | 86 (40) | 114 (32) | 20 (26) | 70 (33) | 24 (11) | 29 (5) | 198 (28) | 387 (23) |
| AIDS Drug Assistance Program | 0 (0) | 85 (16) | 0 (0) | 68 (19) | 0 (0) | 39 (19) | 0 (0) | 273 (47) | 0 (0) | 465 (28) |
| Behaviouralb | ||||||||||
| Injection drug use | ||||||||||
| Ever, reported at baseline | 48 (24) | 160 (31) | 23 (11) | 34 (10) | 19 (25) | 21 (10) | 14 (7) | 40 (7) | 104 (15) | 255 (15) |
| During WIHS follow-up | 24 (12) | 60 (12) | 19 (9) | 9 (3) | 5 (6) | 9 (4) | 3 (1) | 5 (1) | 51 (7) | 83 (5) |
| Past 6 months | 3 (2) | 3 (1) | 3 (1) | 4 (1) | 2 (3) | 3 (1) | 0 (0) | 1 (< 1) | 8 (1) | 11 (1) |
| Use of non-injection illicit drugs | ||||||||||
| Ever, reported at baseline | 136 (69) | 390 (75) | 160 (74) | 224 (63) | 68 (88) | 148 (70) | 157 (73) | 370 (64) | 521 (74) | 1132 (68) |
| During WIHS follow-up | 129 (66) | 310 (60) | 156 (73) | 174 (49) | 53 (69) | 116 (55) | 104 (49) | 232 (40) | 442 (63) | 832 (50) |
| Past 6 months | 45 (23) | 87 (17) | 68 (32) | 71 (20) | 32 (42) | 63 (30) | 69 (32) | 127 (22) | 214 (30) | 348 (21) |
| Marijuana use | ||||||||||
| Ever, reported at baseline | 128 (65) | 361 (70) | 146 (68) | 207 (58) | 65 (84) | 130 (62) | 142 (66) | 317 (55) | 481 (69) | 1015 (61) |
| During WIHS follow-up | 113 (58) | 272 (52) | 137 (64) | 150 (42) | 47 (61) | 94 (45) | 92 (43) | 194 (34) | 389 (55) | 710 (43) |
| Past 6 months | 37 (19) | 78 (15) | 56 (26) | 62 (17) | 21 (27) | 55 (26) | 62 (29) | 108 (19) | 176 (25) | 303 (18) |
| Crack, cocaine or heroin use | ||||||||||
| Ever, reported at baseline | 112 (57) | 320 (62) | 87 (40) | 137 (38) | 64 (83) | 114 (54) | 113 (53) | 278 (48) | 376 (54) | 849 (51) |
| During WIHS follow-up | 96 (49) | 197 (38) | 82 (38) | 80 (22) | 40 (52) | 61 (29) | 49 (23) | 114 (20) | 267 (38) | 452 (27) |
| Past 6 months | 16 (8) | 25 (5) | 17 (8) | 19 (5) | 20 (26) | 16 (8) | 19 (9) | 46 (8) | 72 (10) | 106 (6) |
| Methamphetamines, past 6 months | 0 (0) | 1 (< 1) | 3 (1) | 3 (1) | 3 (4) | 4 (2) | 0 (0) | 2 (< 1) | 6 (1) | 10 (1) |
| Other drugs, past 6 months | 4 (2) | 0 (0) | 4 (2) | 5 (1) | 1 (1) | 2 (1) | 4 (2) | 2 (< 1) | 13 (2) | 9 (1) |
| Alcohol use, past 6 months | ||||||||||
| Complete abstention | 92 (47) | 310 (60) | 76 (35) | 211 (59) | 39 (51) | 101 (48) | 84 (39) | 318 (55) | 291 (41) | 940 (57) |
| > 0–7 drinks/week | 50 (26) | 137 (26) | 91 (42) | 23 (26) | 22 (29) | 71 (34) | 82 (38) | 199 (35) | 245 (35) | 500 (30) |
| > 7–12 drinks/week | 15 (8) | 15 (3) | 15 (7) | 15 (4) | 5 (6) | 8 (4) | 11 (5) | 17 (3) | 46 (7) | 55 (3) |
| > 12 drinks/week | 22 (11) | 26 (5) | 17 (8) | 29 (8) | 10 (13) | 20 (10) | 35 (16) | 39 (7) | 84 (12) | 114 (7) |
| Current smoker | 74 (38) | 160 (31) | 87 (40) | 113 (32) | 44 (57) | 99 (47) | 102 (48) | 253 (44) | 307 (44) | 625 (38) |
| Male sex partners | ||||||||||
| Median lifetime, reported at baseline (IQR) | 10 (5, 30) | 12 (5, 50) | 10 (5, 40) | 10 (5, 25) | 15 (8, 50) | 12 (5, 30) | 17 (10, 52) | 10 (5, 30) | 12 (6, 40) | 10 (5, 30) |
| At least one, past 6 months | 75 (38) | 195 (38) | 135 (63) | 219 (62) | 43 (56) | 121 (58) | 166 (78) | 366 (64) | 419 (60) | 901 (54) |
| More than one, past 6 months | 3 (2) | 13 (3) | 29 (13) | 20 (6) | 11 (14) | 15 (7) | 57 (27) | 43 (7) | 100 (14) | 91 (5) |
| Any female sex partners | ||||||||||
| Ever | 48 (24) | 115 (22) | 67 (31) | 71 (20) | 27 (35) | 58 (28) | 82 (38) | 139 (24) | 224 (32) | 383 (23) |
| Past 6 months | 6 (3) | 11 (2) | 13 (6) | 10 (3) | 3 (4) | 5 (2) | 12 (6) | 10 (2) | 34 (5) | 36 (2) |
| Vaginal sex | ||||||||||
| Past 6 months | 71 (36) | 189 (36) | 134 (62) | 219 (62) | 42 (55) | 119 (57) | 165 (77) | 360 (63) | 412 (59) | 887 (53) |
| Always used condoms | 17 (24) | 121 (64) | 26 (19) | 126 (58) | 5 (12) | 66 (55) | 42 (25) | 197 (55) | 90 (22) | 510 (58) |
| Anal sex | ||||||||||
| Past 6 months | 4 (2) | 8 (2) | 11 (5) | 13 (4) | 2 (3) | 11 (5) | 9 (4) | 20 (3) | 26 (4) | 52 (3) |
| Always used condoms | 3 (75) | 5 (63) | 3 (27) | 6 (46) | 0 (0) | 3 (27) | 2 (22) | 5 (25) | 8 (31) | 19 (37) |
| Sex with known HIV+ partner | ||||||||||
| During WIHS follow-upd | 23 (12) | 187 (36) | 49 (23) | 160 (45) | 14 (18) | 87 (41) | 28 (13) | 206 (36) | 114 (16) | 640 (39) |
| Past 6 months | 2 (1) | 41 (8) | 7 (3) | 54 (15) | 3 (4) | 32 (15) | 16 (7) | 110 (19) | 28 (4) | 237 (14) |
| Transactional sex | ||||||||||
| Ever | 54 (28) | 187 (36) | 60 (28) | 108 (30) | 44 (57) | 75 (36) | 105 (49) | 183 (32) | 263 (37) | 553 (33) |
| Past 6 months | 3 (2) | 2 (< 1) | 6 (3) | 1 (< 1) | 3 (4) | 4 (2) | 8 (4) | 10 (2) | 20 (3) | 17 (1) |
| Ever experienced sexual abuse | 86 (44) | 257 (50) | 63 (29) | 89 (25) | 36(47) | 89 (42) | 94 (44) | 197 (34) | 279 (40) | 632 (38) |
| Ever experienced physical violence | 111 (57) | 309 (60) | 95 (44) | 146 (41) | 51 (66) | 130 (62) | 127 (59) | 260 (45) | 384 (55) | 845 (51) |
Dates of enrolment were: Wave 1 = 1994–95; Wave 2 = 2001–02; Wave 3 = 2011–12; Wave 4 = 2013–15.
N (column %), unless otherwise noted.
Includes: American Indian, Alaskan Native, Asian, Native Hawaiian, Pacific Islander and multiple races/ethnicities.
Data are available since 1 October 2005.
Table 3.
Clinical characteristics of participants in the Women’s Interagency HIV Study, by enrolment wavea and HIV status at latest WIHS visit
| Wave 1 |
Wave 2 |
Wave 3 |
Wave 4 |
Overall |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | HIV+ | ||
| N = 196 | N = 519 | N = 215 | N = 356 | N = 77 | N = 210 | N = 214 | N = 576 | N = 702 | N = 1661 | ||
| Clinicalb | |||||||||||
| Median CD4 cells/mm3 (IQR) | 1049 | 619 | 1018 | 604 | 1000 | 683 | 991 | 635 | 1011 | 628 | |
| (832, 1293) | (393, 850) | (785, 1280) | (411, 797) | (798, 1259) | (515, 934) | (816, 1267) | (436, 865) | (815, 1280) | (435, 853) | ||
| HIV RNA | |||||||||||
| Not detected: < 20 copies/ml | N/A | 351 (68) | N/A | 238 (67) | N/A | 143 (68) | N/A | 420 (73) | N/A | 1152 (69) | |
| Median copies/ml among detected (IQR) | N/A | 142 | N/A | 545 | N/A | 309 | N/A | 383 | N/A | 265 | |
| (42, 2760) | (74, 23000) | (49, 3030) | (51, 9540) | (53, 7000) | |||||||
| Antiretroviral drug use | |||||||||||
| Ever | N/A | 510 (98) | N/A | 347 (97) | N/A | 198 (94) | N/A | 555 (96) | N/A | 1610 (97) | |
| At last visit | N/A | 478 (92) | N/A | 308 (87) | N/A | 178 (85) | N/A | 522 (91) | N/A | 1486 (89) | |
| 100% adherent | N/A | 231 (48) | N/A | 140 (45) | N/A | 96 (54) | N/A | 260 (50) | N/A | 727 (49) | |
| ≥ 95% adherent | N/A | 417 (87) | N/A | 261 (85) | N/A | 155 (87) | N/A | 445 (85) | N/A | 1278 (86) | |
| Body mass index (IQR) | 32 | 29 | 31 | 30 | 30 | 30 | 34 | 32 | 32 | 30 | |
| (27, 38) | (24, 34) | (26, 37) | (25, 35) | (26, 36) | (25, 37) | (28, 38) | (27, 40) | (27, 37) | (25, 36) | ||
| Blood pressure (IQR) | |||||||||||
| Systolic | 128 | 120 | 122 | 120 | 122 | 119 | 122 | 120 | 124 | 120 | |
| (116, 144) | (111, 135) | (111, 136) | (110, 131) | (111, 134) | (110, 134) | (112, 137) | (110, 136) | (112, 138) | (110, 134) | ||
| Diastolic | 76 | 74 | 76 | 74 | 75 | 76 | 74 | 75 | 75 | 75 | |
| (70, 84) | (68, 81) | (69, 84) | (68, 81) | (69, 83) | (70, 83) | (69, 83) | (69, 82) | (69, 84) | (69, 82) | ||
| Hypertensionc | 118 (60) | 298 (57) | 82 (38) | 142 (40) | 43 (56) | 92 (44) | 97 (45) | 304 (53) | 340 (48) | 836 (50) | |
| LDL cholesterol | |||||||||||
| Unknown | 28 (14) | 79 (15) | 27 (13) | 50 (14) | 9 (12) | 24 (11) | 8 (4) | 23 (4) | 72 (10) | 176 (11) | |
| ≤ 3.36 mmol/l | 126 (64) | 356 (69) | 159 (74) | 261 (73) | 61 (79) | 156 (74) | 178 (83) | 461 (80) | 524 (75) | 1234 (74) | |
| > 3.36 mmol/l | 42 (21) | 84 (16) | 29 (13) | 45 (13) | 7 (9) | 30 (14) | 28 (13) | 92 (16) | 106 (15) | 251 (15) | |
| HDL cholesterol | |||||||||||
| Unknown | 28 (14) | 79 (15) | 27 (13) | 50 (14) | 9 (12) | 24 (11) | 8 (4) | 23 (4) | 72 (10) | 176 (11) | |
| < 1.03 mmol/l | 24 (12) | 52 (10) | 26 (12) | 59 (17) | 8 (10) | 25 (12) | 36 (17) | 104 (18) | 536 (76) | 1245 (75) | |
| ≥ 1.03 mmol/l | 144 (73) | 388 (75) | 162 (75) | 247 (69) | 60 (78) | 161 (77) | 170 (79) | 449 (78) | 94 (13) | 240 (14) | |
| Estimated GFR: MDRD | 88 | 80 | 98 | 94 | 87 | 89 | 100 | 94 | 95 | 89 | |
| (75, 103) | (64, 99) | (85, 111) | (77, 109) | (75, 104) | (74, 104) | (84, 117) | (76, 111) | (81, 111) | (72, 106) | ||
| Diabetes mellitus | 64 (33) | 147 (28) | 36 (17) | 66 (19) | 19 (25) | 38 (18) | 39 (18) | 89 (15) | 158 (23) | 340 (20) | |
| History of adverse cardiovascular eventd | 47 (24) | 129 (25) | 20 (9) | 37 (10) | 17 (22) | 19 (9) | 24 (11) | 58 (10) | 108 (15) | 243 (15) | |
| History of cancer | 16 (8) | 47 (9) | 6 (3) | 21 (6) | 2 (3) | 5 (2) | 2 (1) | 10 (2) | 26 (4) | 83 (5) | |
| Baseline HCV infection status | |||||||||||
| Negative | 151 (77) | 335 (65) | 194 (90) | 309 (87) | 65 (84) | 172 (82) | 187 (87) | 504 (88) | 597 (85) | 1320 (79) | |
| Resolved infection, Ab + RNA- | 10 (5) | 51 (10) | 8 (4) | 13 (4) | 0 (0) | 2 (1) | 0 (0) | 0 (0) | 18 (3) | 66 (4) | |
| Ab + RNA unknown | 1 (1) | 6 (1) | 0 (0) | 2 (1) | 11 (14) | 33 (16) | 27 (13) | 70 (12) | 39 (6) | 111 (7) | |
| Active infection, RNA+ | 33 (17) | 126 (24) | 13 (6) | 32 (9) | 1 (1) | 2 (1) | 0 (0) | 0 (0) | 47 (7) | 160 (10) | |
| History of clinical AIDS | N/A | 295 (57) | N/A | 108 (30) | N/A | 37 (18) | N/A | 82 (14) | N/A | 522 (31) | |
| Results of last Pap smear | |||||||||||
| Normal or unspecified ASCUS | 179 (91) | 457 (88) | 195 (91) | 322 (90) | 74 (96) | 197 (94) | 203 (95) | 536 (93) | 651 (93) | 1512 (91) | |
| LSIL | 0 (0) | 15 (3) | 2 (1) | 16 (4) | 2 (3) | 5 (2) | 1 (< 1) | 25 (4) | 5 (1) | 61 (4) | |
| HSIL or carcinoma in situ | 2 (1) | 8 (2) | 1 (< 1) | 5 (1) | 1 (1) | 3 (1) | 2 (1) | 13 (2) | 6 (1) | 29 (2) | |
| High risk HPV during WIHS follow-upe | 15 (8) | 93 (18) | 11 (5) | 27 (8) | N/A | N/A | N/A | N/A | 26 (6) | 120 (14) | |
| Sexually transmitted infectionf | |||||||||||
| Baseline | 108 (55) | 359 (69) | 124 (58) | 217 (61) | 56 (73) | 144 (69) | 153 (72) | 397 (69) | 441 (63) | 1117 (67) | |
| During WIHS follow-up | 91 (46) | 352 (68) | 98 (46) | 189 (53) | 26 (34) | 59 (28) | 65 (30) | 156 (27) | 280 (40) | 756 (46) | |
| Past 6 months | 1 (1) | 13 (3) | 3 (1) | 9 (3) | 3 (4) | 6 (3) | 13 (6) | 30 (5) | 20 (3) | 58 (3) | |
| Pregnancy | |||||||||||
| Ever during WIHS | 74 (38) | 135 (26) | 118 (55) | 122 (34) | 3 (4) | 14 (7) | 13 (6) | 20 (3) | 208 (30) | 291 (18) | |
| Hormonal contraceptive use | 2 (1) | 8 (2) | 15 (7) | 24 (7) | 3 (4) | 11 (5) | 18 (8) | 36 (6) | 38 (5) | 79 (5) | |
| Menopause status at last visitg | |||||||||||
| Premenopausal | 26 (13) | 47 (9) | 108 (50) | 122 (34) | 18 (23) | 72 (34) | 97 (45) | 214 (37) | 249 (35) | 455 (27) | |
| Early perimenopause | 16 (8) | 31 (6) | 24 (11) | 38 (11) | 8 (10) | 26 (12) | 22 (10) | 52 (9) | 70 (10) | 147 (9) | |
| Late perimenopause | 7 (4) | 19 (4) | 12 (6) | 26 (7) | 4 (5) | 8 (4) | 7 (3) | 30 (5) | 30 (4) | 83 (5) | |
| Postmenopausal (natural or surgical) | 129 (66) | 392 (76) | 56 (26) | 160 (45) | 46 (60) | 94 (45) | 84 (39) | 277 (48) | 315 (45) | 923 (56) | |
| At risk for depressionh | 53 (27) | 145 (28) | 50 (23) | 80 (22) | 27 (35) | 70 (33) | 80 (37) | 195 (34) | 210 (30) | 490 (30) | |
LDL, low-density lipoprotein; HDL, high-density lipoprotein; ASCUS, atypical squamous cells of uncertain significance; GFR, glomerular filtration rate (ml/min/1.73 m2); HSIL, high-grade squamous epithelial lesions; MDRD, modified diet in renal disease; LSIL, low-grade squamous epithelial lesions; NA, not available.
Dates of enrolment were: Wave 1 = 1994-95; Wave 2 = 2001-02; Wave 3 = 2011-12; Wave 4 = 2013-15.
N (column %), unless otherwise noted.
Confirmed through self-report: on medication or systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg.
Myocardial infarction, stroke or transient ischaemic attack, hospitalization for congestive heart failure or chest pain/angina, or surgery on heart vessels.
High-risk HPV types 16, 18, 31, 45.
Self report of gonorrhoea, syphilis, chlamydia, pelvic inflammatory disease, genital herpes, genital warts, or trichomonal vaginitis.
Stages of Reproductive Aging Workshop (STRAW) + 10 criteria.
Centers for Epidemiologic Studies – Depression (CESD) score of 16 or greater.
Despite the high prevalence of IDU reported by participants at study entry, current IDU is less common; only 1% reported IDU in the 6 months preceding their most recent study visit. By contrast, higher proportions (6% HIV+, 10% HIV-) reported smoking or snorting crack cocaine, cocaine or heroin in the past 6 months. Almost one-third of women (30% HIV+, 30% HIV-) screened positive for risk of depression at their last visit, and 38% of HIV+ and 40% of HIV- women reported a history of sexual abuse. Cigarette smoking is common (38%, HIV+, 44% HIV-), as is obesity, with median body mass index (BMI) of 30 (HIV+) and 32 (HIV-). Substantial proportions of women have diabetes (20% HIV+, 23% HIV-), hypertension (50% HIV+, 48% HIV-) and a history of cardiovascular disease (15% HIV+, 15% HIV-).
How often have they been followed up? What has been measured?
Participants attend WIHS visits every 6 months for data collection. Each of these visits includes a scripted interview that covers general medical and medication history, substance use, sexual history, health care use and psychosocial/behavioural data. A physical examination, including standardized blood pressure assessment, anthropometric measures and gynaecological examination with cervical cytology, is conducted. Frailty assessments and ankle-brachial index measures to determine the extent of peripheral arterial disease are performed annually or every 2 years, respectively, on women 40 years of age or older. Colposcopy and biopsy are performed, as clinically indicated. Women who need additional evaluation and/or treatment of health conditions are referred to appropriate providers. In addition to clinical data, the WIHS collects haematological, metabolic, immunological, genetic and virological data. The following sections describe several special study modules that operate at varying intervals.
Neurocognition
Incorporation of neurocognitive testing has facilitated HIV-associated neurocognitive disorder (HAND) research.6 Every 2 years since April 2009, all English-speaking participants have performed neurological and neurocognitive function tests, including: psychomotor speed and attention (Symbol Digit Substitution Test), gross motor speed (timed gait), fine motor speed and coordination (Grooved Pegboard), visuo-motor speed and executive function (Trail Making Test Part A and B), working memory/executive function (Letter Number Sequencing), executive function (Stroop, Letter Fluency; FAS), semantic fluency (animals) and verbal memory (Hopkins Verbal Memory Test-Revised and Brief Visual Memory Test). Participants who speak only Spanish undergo more limited testing. The Wide Range Achievement Test-3 (WRAT-3) was administered once during each participant’s first neurocognitive test.
Geocoding
Because neighbourhood characteristics affect health,7 the WIHS has used ArcGIS™ (Esri, US) to geocode the residence of consenting participants annually since 2013. ArcGIS matches each participant’s geocoded location to a Federal Information Processing Standard (FIPS) code that identifies geographical locations. Each site creates a limited dataset that contains only the participants’ FIPS codes (at the census block group level) and WIHS identification (ID) numbers. The WIHS links the FIPS codes to census-linked datasets, such as the American Community Survey, to create group-level variables that describe neighbourhood characteristics. The WIHS creates individualized contextual datasets for investigators which only contain the WIHS ID and the value of the specific group-level variables requested by the investigator (e.g. percentage of individuals under the federal poverty line). Residences of 89% of HIV-seropositive and 92% of HIV-seronegative active participants have been geocoded.
Host genetics
Several WIHS studies have examined the genetic characteristics of cohort participants and their association with key study outcomes. These include candidate gene analyses and whole-exome studies of selected participant subsets, as well as WIHS-wide genotyping (Haemoglobin S, Haemoglobin C, CCR5Δ32, APOE). Viable cells are available in the WIHS specimen repository for epigenetics research. Genome-wide data have also complemented clinical, imaging, biochemical, biological and sociodemographic studies. WIHS benefits from considerable genetic diversity. A total of 74% of all 4982 WIHS participants (63% of 2363 active WIHS participants) consented to and have undergone genome-wide association study (GWAS) testing; genome-wide data from 4789 participants are available for analysis. Investigators have genotyped DNA isolated from peripheral blood mononuclear cells, which are maintained in the WIHS DNA Repository. About 2.5 million common single nucleotide polymorphisms (SNPs) and 2.5 million rare SNPs have been identified. The WIHS GWAS analytical support group performs quality control of WIHS genomic data and provides analytical and bioinformatics support to investigators.
Liver disease
Participants underwent hepatitis C virus (HCV) antibody testing at or shortly after enrolment and, among HCV-seropositive participants, HCV RNA testing was performed to determine clearance or persistence of infection. HCV RNA quantification is also performed at least 12 weeks after HCV treatment to determine achievement of cure. Periodic HCV antibody and HCV RNA testing has detected incident HCV infections. WIHS has documented HCV infection in 29% of the cohort, and over 400 women have cleared HCV infection (either spontaneously or through treatment).
The cohort represents a unique opportunity to evaluate the effects of HCV mono- and co-infection and HCV treatment, substance use and obesity on liver disease among women. WIHS measures serum biomarkers [aspartate aminotransferase-to-platelet ratio index (APRI) and Fibrosis-4 (FIB4) score] to estimate the severity of liver fibrosis. In addition, WIHS participants undergo transient elastography [via FibroScan® 502 Touch device with CAPTM software (Echosens, Paris)] to evaluate extent of liver fibrosis and steatosis according to standardized protocols, which are performed by trained and certified WIHS staff. These data allow longitudinal analysis of the progression of liver fibrosis and steatosis.
Antiretroviral pharmacology
WIHS supports one of the largest and most diverse datasets focused on antiretroviral pharmacology, with intensive pharmacokinetic data for a number of antiretroviral drugs and a database that includes 2000 measurements of antiretroviral drugs in scalp hair and individual plasma samples.
Specimen repository
In addition to data that are accessible to researchers, the study maintains a repository of biological specimens, including serum, plasma, urine, peripheral blood mononuclear cells, oral fungal cultures, hair, cervico-vaginal lavage and cervical swabs that are collected at each visit for all participants. Also available are plasma anti-Müllerian hormone levels, a biomarker of ovarian reserve that estimates the time of onset of menopause. Table 4 outlines the type and quantity of each specimen collected and the number of specimens available in the repository.
Table 4.
Summary of Women’s Interagency HIV Study specimen repository
| Type of specimen | Volume of specimen | No. of specimens stored |
|
|---|---|---|---|
| HIV+ N = 3677 | HIV- N = 1305 | ||
| Serum | 0.5, 1.0 or 1.8 ml aliquots | 251584 | 94539 |
| Plasma | 0.5, 1.0 or 1.8 ml aliquots | 332191 | 125460 |
| Viable PBMC cells | > 5 million cells per aliquot | 126106 | 52449 |
| PBMC pellets | 0.5 or 2 million cells per aliquot | 284118 | 103761 |
| Urine (clean void) | 1.0 or 5.0 ml aliquots | 37992 | 13305 |
| Urine supernatant | 1.0 ml aliquots | 37404 | 15587 |
| Whole CVL | 1.0 or 1.5 ml aliquots | 266442 | 104093 |
| CVL supernatant | 0.5 or 1.0 ml aliquots | 16017 | 4847 |
| CVL pellet (suspended) | > 2.5 million cells per aliquot | 5187 | 6586 |
| Cervical swab | 1 swab | 38075 | 15609 |
| Oral fungal culture (mouth scrapings) | 1 swab | 2165 | 650 |
| Vaginal fungal culture | 1 swab | 1763 | 629 |
PBMC, peripheral blood mononuclear cells; CVL, cervico-vaginal lavage.
Key findings
The WIHS has yielded more than 800 publications to date, see [http://statepi.jhsph.edu/wihs/wordpress/wp-content/uploads/2015/12/wihs_archives.pdf]. Below we outline a few of the study’s major findings.
The WIHS has documented the dramatic changes in mortality and its contributing factors among WWH since the early years of the epidemic. After the advent of HAART, the age- and sex-specific standardized mortality ratio of 24.7 in 1996 fell to a mean of about 10.3 in 2001-04, and causes of mortality shifted from a predominance of AIDS to non-AIDS causes, such as overdose, trauma, cancer, liver disease and heart disease by 2003.8 Although survival increased with improvements in ART, PWH continued to die earlier; in both the WIHS and the Multicenter AIDS Cohort Study (MACS) (a prospective cohort of men with and at risk for HIV infection), PWH died a median 7.6 years earlier than their HIV-seronegative counterparts. Whereas the proportion of non-AIDS causes of death was similar between women and men with HIV after HAART’s introduction (47% vs 50%), among PWH who died of non-AIDS causes, women have had substantially lower life expectancies than men (median age at death 55.9 vs 66.0 years).9 An earlier study of WIHS and MACS participants revealed higher death rates due to accident or injury among women (2.96 deaths per 1000 person-years among both HIV+ and HIV-) than men (0.79 for HIV+ and 0.63 for HIV-), with differing risk factors for death among men (higher education, symptoms of depression and more sex partners) and women (lower CD4+ T cell count, unemployment, increased alcohol use and injection drug use).10 WIHS mortality analyses also demonstrated that among WWH, Black women died at nearly double the rate of White women, even after adjusting for confounders known to be associated with death due to HIV infection.11 WIHS participants also contributed to an early landmark study of patients in care between 1996 and 2005, which demonstrated the survival benefits of early initiation of ART at CD4+ T cell counts above the previously recommended threshold.12
Cardiovascular disease has emerged as a significant cause of morbidity among PWH. A cross-sectional analysis of carotid ultrasound examinations among MACS and WIHS participants established low CD4+ T cell count (< 200 cells/ml) as a major risk factor for atherosclerosis among PWH.13 A prospective evaluation using carotid ultrasonography of men and women in the MACS and WIHS demonstrated an association of HIV infection with progression of subclinical carotid atherosclerosis, as evidenced by a 1.6-fold greater risk of new plaque formation after adjusting for cardiac and metabolic risk factors.14
WIHS has enhanced our understanding of the natural history of HPV infection among women with HIV. WIHS studies have documented the increased prevalence of human papillomavirus (HPV) infection, including oncogenic HPV types,15 abnormal Pap tests16 and cervical precancerous lesions.17 Despite the higher prevalence of HPV among WWH, the risk of invasive cervical cancer is only modestly increased among WWH who undergo regular screening,18 a finding that has informed cervical cancer screening guidelines for this population.
Assessment of severity of liver disease is important in counselling women with HIV-hepatitis C virus (HCV) co-infection. Studies of HIV-HCV co-infected WIHS participants revealed that although Black women are less likely to spontaneously clear HCV infection than Hispanic or Caucasian women,19 they are also less likely than Caucasian and Hispanic women to die of liver disease, a finding that had not previously been reported in women.20 Non-invasive assessment of liver disease severity has become increasingly important in evaluation of HCV infection. The WIHS has demonstrated the utility of non-invasive serum markers, such as APRI, FIB-4 and enhanced liver fibrosis (ELF), for assessing liver fibrosis severity and predicting mortality among women with HCV.21–23
HIV’s impact on neurocognitive function among men has been well described, but neurocognitive function among WWH has received relatively little research attention. Early studies among WIHS WWH revealed deficits in episodic verbal memory which correlated with neuroimaging evidence of hippocampal dysfunction, suggesting an effect of HIV on the neurological systems that govern verbal memory.24 Neurocognitive testing in another WIHS study identified an interaction between HIV infection and recent cocaine or heroin use on verbal learning and memory, with recent illicit drug use affecting only WWH.25 Despite the worse performance of WWH on tests of verbal learning, delayed recall and recognition and psychomotor speed and attention, the effect of HIV on cognition was small; other factors such as reading level, age and years of education had a greater effect than HIV infection on cognitive performance.6
Intensive drug pharmacokinetic studies have revealed factors associated with elevated anti-retroviral drug levels among Black and Hispanic women, groups that are often poorly represented in pre-marketing studies.26 The WIHS has evaluated antiretroviral drug concentrations in hair, as a non-invasive measure of long-term drug exposure. The WIHS has demonstrated a relationship between concentrations of antiretroviral drugs in hair and HIV suppression,27,28 and has also shown a relationship between genetic markers associated with decreased metabolism of some antiretrovirals and increased drug concentrations in hair.29
Strengths and limitations
The cohort’s strengths have made it a major platform for women’s HIV research: a large sample size, collection of rich clinical, behavioural and laboratory data at 6-month intervals and a large diverse biospecimen repository that spans more than 20 years. The control group is well matched to WWH with respect to demographic, behavioural and other risk characteristics. The cohort is especially unique in its racial/ethnic diversity and inclusion of large numbers of women of colour and women of lower socioeconomic status, who mirror the HIV epidemic among US women. The interval cohort design of the WIHS, with data collection at specified time points, has several advantages over the clinical cohort design; interval data collection renders the data more uniform and complete than is typically afforded by clinical cohorts. WIHS recruited women from both clinical and non-clinical settings. Although study staff facilitate participants’ care by referring them for clinical, social and other services, some participants struggle to maintain access to care. Inclusion of participants from outside clinical settings allows more accurate reflection of WWH, including those who do not receive consistent HIV care.
A major challenge of the WIHS is intrinsic to observational cohorts: potential confounding restricts ability to make causal inferences. Randomized controlled trials permit causal inferences between exposures and outcomes, but their generalizability is often limited. The success of the WIHS is due in large part to the dedication of its participants, many of whom have remained in the study for decades, but this very characteristic likely limits the study’s generalizability to the entire population of WWH in the USA. In addition, participants in clinical trials and other studies sometimes experience improved outcomes compared with non-participants—due to better care, behavioural changes and/or possibly other unknown factors.30,31 Nevertheless, enrolment of new participants over four enrolment periods during the greater than 20-year span of the study has helped to ensure the cohort’s continued reflection of US women who are affected by HIV.
Can I get hold of the data? Where can I find out more?
The WIHS welcomes collaborations with investigators. Information on proposing new studies and analyses, data and specimens is available at [wihshealth.org]. The WIHS also provides de-identified data in its Public Use Dataset, which is updated annually and may be obtained by submitting a data request form.
Profile in a nutshell
The Women’s Interagency HIV Study (WIHS) investigates the impact and progression of HIV infection among women in the USA.
The cohort consists of 4982 women at nine sites in the USA, who are either living with HIV (3677) or are HIV-seronegative (1305) but have characteristics that increase their risk of acquiring HIV.
A total of 2623 participants were enrolled at baseline in 1993; three subsequent waves have expanded the study’s enrolment. Since the beginning of the study, 1268 participants have died, 130 have withdrawn, 806 have been dis-enrolled for administrative reasons, and 415 have been lost to follow-up. Current median age of the 2363 actively followed participants is 49 years [HIV-seronegative; interquartile range (IQR) 41, 55) and 51 years (HIV+; 44, 56).
Participants attend bi-annual study visits. Data collection includes: clinical, medication and behavioral history; general physical, gynaecological and neurocognitive examination; and laboratory data including haematological, metabolic, immunological and virological testing, as well as cervical cytology, genetic testing and transient liver elastography. The specimen repository contains blood, urine, cervico-vaginal lavage and cervical swab specimens.
The WIHS welcomes collaborations; information on collaboration can be found at [wihshealth.org].
Funding
This work was supported by the National Institutes of Health (U01 AI103401 to M.C.K., U01 AI103408 to G.W. and I.O., U01 AI035004 to K.A., U01 AI031834 to H.M., U01 AI034993 to M.C., U01 AI034994 to S.G.K., U01 AI103397 to M.A.F., U01 AI103390 to A.A.A., U01 AI034989 to R.M.G. and P.C.T., U01 AI042590 to S.G).
Acknowledgements
The authors thank Eryka Wentz for assistance with specimen availability, Christine Alden for manuscript review, Gayle Springer for assistance with data preparation and the WIHS participants for their time and willingness to help advance knowledge of HIV.
Conflict of interest: A.A.A. has received research funding from Gilead and is on a Merck Advisory Board. P.C.T. is conducting research sponsored by Merck and Theratechnologies.
References
- 1. World Health Organization. Global Health Observatory (GHO) Data: Number of Women Living With HIV. Geneva: World Health Organization, 2017. [Google Scholar]
- 2. Barkan SE, Melnick SL, Preston-Martin S. et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998;9:117–25. [PubMed] [Google Scholar]
- 3. Bacon MC, von Wyl V, Alden C. et al. The Women's Interagency HIV Study: an observational cohort brings clinical sciences to the bench. Clin Diagn Lab Immunol 2005;12:1013–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in HIV-1-infected Adults and Adolescents. Washington, DC: Department of Health and Human Services (US; ), 2008. [Google Scholar]
- 5. Bradley H, Hall HI, Wolitski RJ. et al. Vital signs: HIV diagnosis, care, and treatment among persons living with HIV—United States, 2011. MMWR Morb Mortal Wkly Rep 2014;63:1113–17. [PMC free article] [PubMed] [Google Scholar]
- 6. Maki PM, Rubin LH, Valcour V. et al. Cognitive function in women with HIV: findings from the Women's Interagency HIV Study. Neurology 2015;84:231–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. World Health Organization. Social Determinants of Health. Geneva: World Health Organization, 2010. [Google Scholar]
- 8. French AL, Gawel SH, Hershow R. et al. Trends in mortality and causes of death among women with HIV in the US: A ten-year study. J Acquir Immune Defic Syndr 2009;51:399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wada N, Jacobson LP, Cohen M, French A, Phair J, Muñoz A.. Cause-specific life expectancies after 35 years of age for human immunodeficiency syndrome-infected and human immunodeficiency syndrome-negative individuals followed simultaneously in long-term cohort studies, 1984–2008. Am J Epidemiol 2013;177:116–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hessol NA, Kalinowski A, Benning L. et al. Mortality among participants in the Multicenter AIDS Cohort Study and the Women's Interagency HIV Study. Clin Infect Dis 2007;44:287–94. [DOI] [PubMed] [Google Scholar]
- 11. Murphy K, Hoover DR, Shi Q. et al. Association of self-reported race with AIDS death in continuous HAART users in a cohort of HIV-infected women in the United States. AIDS 2013;27:2413–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kitahata MM, Gange SJ, Abraham AG. et al . Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med 2009;360:1815–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kaplan RC, Kingsley LA, Gange SJ. et al . Low CD4+ T cell count as a major atherosclerosis risk factor in HIV-infected women and men. AIDS 2008;22:1615–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hanna DB, Post WS, Deal JA. et al. HIV infection is associated with progression of subclinical carotid atherosclerosis. Clin Infect Dis 2015;61:640–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Massad LS, Xie X, Burk R. et al. Long-term cumulative detection of human papillomavirus among HIV seropositive women. AIDS 2014;28:2601–08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Massad LS, Seaberg EC, Wright RL. et al. Squamous cervical lesions in women with human immunodeficiency virus: long-term follow-up. Obstet Gynecol 2008;111:1388–93. [DOI] [PubMed] [Google Scholar]
- 17. Massad LS, Xie X, D’Souza G. et al. Incidence of cervical precancers among HIV-seropositive women. Am J Obstet Gynecol 2015;212:606. e1–e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Massad LS, Seaberg EC, Watts DH. et al. Low incidence of invasive cervical cancer among HIV-infected US women in a prevention program. AIDS 2004;18:109–13. [DOI] [PubMed] [Google Scholar]
- 19. Sarkar M, Bacchetti P, Tien P. et al . Racial/ethnic differences in spontaneous HCV clearance in HIV infected and uninfected women. Dig Dis Sci 2013;58:1341–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Sarkar M, Bacchetti P, French AL. et al. Lower liver‐related death in African‐American women with human immunodeficiency virus/hepatitis C virus coinfection, compared to Caucasian and Hispanic women. Hepatology 2012;56:1699–705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Bambha K, Pierce C, Cox C. et al. Assessing mortality in women with hepatitis C virus and HIV using indirect markers of fibrosis. AIDS 2012;26:599–607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Swanson S, Ma Y, Scherzer R. et al. Association of HIV, hepatitis C virus, and liver fibrosis severity with the enhanced liver fibrosis score. J Infect Dis 2016;213:1079–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Peters MG, Bacchetti P, Boylan R. et al. Enhanced liver fibrosis marker as a noninvasive predictor of mortality in HIV/hepatitis C virus-coinfected women from a multicenter study of women with or at risk for HIV. AIDS 2016;30:723–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Maki P, Cohen M, Weber K, Little D. et al. Impairments in memory and hippocampal function in HIV-positive vs HIV-negative women: a preliminary study. Neurology 2009;72:1661–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Meyer VJ, Rubin LH, Martin E. et al. HIV and recent illicit drug use interact to affect verbal memory in women. J Acquir Immune Defic Syndr 2013;63:67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Baxi SM, Greenblatt RM, Bacchetti P. et al. Common clinical conditions - age, low BMI, ritonavir use, mild renal impairment - affect tenofovir pharmacokinetics in a large cohort of HIV-infected women. AIDS 2014;28:59–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Baxi SM, Greenblatt RM, Bacchetti P. et al. Nevirapine concentration in hair samples is a strong predictor of virologic suppression in a prospective cohort of HIV-infected patients. PLoS One 2015;10:e0129100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Gandhi M, Ameli N, Bacchetti P. et al. Atazanavir concentration in hair is the strongest predictor of outcomes on antiretroviral therapy. Clin Infect Dis 2011;52:1267–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Gandhi M, Greenblatt RM, Bacchetti P. et al. A single-nucleotide polymorphism in CYP2B6 leads to > 3-fold increases in efavirenz concentrations in plasma and hair among HIV-infected women. J Infect Dis 2012;206:1453–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Menezes P, Miller WC, Wohl DA, Adimora AA, Leone PA, Eron JJ Jr.. Does HAART efficacy translate to effectiveness? Evidence for a trial effect. PLoS One 2011;6:e21824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P.. The Hawthorne Effect: a randomised, controlled trial. BMC Med Res Methodol 2007;7:30. [DOI] [PMC free article] [PubMed] [Google Scholar]