Table 2.
Determinants of loss to follow-up from PrEP care (n = 647 participants, 4831 measures) 1.
| Univariate | Multivariate | |||
|---|---|---|---|---|
| Variable | n (%) 2 | Mean Follow-Up Time in Months (SD) | HR [95% CI], p-Value | aHR [95% CI], p-Value |
| Age group (in years) | ||||
| 18–24 | 759 (35.9) | 7.9 (7.7) | 1.64 [1.31–2.05], <0.001 | 1.50 [1.17–1.94], 0.002 |
| >24 | 1253 (59.4) | 12 (9.1) | 1 (ref) | 1 (ref) |
| Employment status | ||||
| Employed | 1137 (53.9) | 11.4 (9.3) | 1 (ref) | 1 (ref) |
| Unemployed | 863 (40.9) | 9.0 (8.0) | 1.44 [1.5–1.8], 0.001 | 1.33 [1.03–1.71], 0.027 |
| Chosen PrEP regimen | ||||
| Event-driven | 1356 (64.3) | 11.3 (8.5) | 0.82 [0.64–1.06], 0.125 | |
| Daily | 370 (17.5) | 10.3 (8.6) | 1 (ref) | |
| PrEP adherence | ||||
| No PrEP | 561 (26.6) | 10.8 (8.9) | 1.89 [1.44–2.49], <0.001 | |
| Poor | 114 (5.4) | 11.0 (7.9) | 2.22 [1.32–3.73], 0.003 | |
| Suboptimal | 205 (9.7) | 11.6 (8.3) | 1.53 [1.05–2.23], 0.026 | |
| Optimal | 751 (35.6) | 14.2 (7.6) | 1 (ref) | |
| Contacted PE outside of scheduled visits | ||||
| Yes | 601 (28.4) | 11.4 (9.7) | 0.73 [0.56–0.94], 0.014 | 0.74 [0.57–0.97], 0.032 |
| No | 1349 (63.9) | 9.9 (8.4) | 1 (ref) | 1 (ref) |
| Self-defined sexual orientation | ||||
| Heterosexual | 43 (2.0) | 9.0 (5.5) | 1.9 [0.96–3.75], 0.066 | 1.84 [0.92–3.69], 0.086 |
| Homosexual/gay | 730 (34.6) | 9.8 (8.8) | 1.18 [0.93–1.49], 0.170 | 1.20 [0.94–1.53], 0.149 |
| Bisexual | 1210 (57.4) | 10.8 (9.1) | 1 (ref) | 1 (ref) |
| Self-defined gender identity | ||||
| Both a man and a woman; more a woman; neither a woman or a man | 784 (37.2) | 9.9 (8.2) | 1.29 [1.04–1.61], 0.023 | |
| Man or boy | 1216 (57.6) | 10.5 (9.2) | 1 (ref) | |
| Depression score (PHQ-9) | ||||
| Moderate to high | 954 (45.2) | 9.4 (8.5) | 1.44 [1.04–1.98], 0.029 | 1.63 [1.12–2.38], 0.010 |
| Minimal | 1156 (54.8) | 10.7 (8.9) | 1 (ref) | 1 (ref) |
| Had a casual male sexual partner in the previous three months | ||||
| Yes | 1058 (50.1) | 9.3 (7.8) | 1.33 [1.04–1.69], 0.022 | |
| No | 1058 (49.7) | 10.9 (9.6) | 1 (ref) | |
| Had a casual female sexual partner in the previous three months | ||||
| Yes | 345 (16.4) | 8.7 (8.9) | 1.40 [1.03–1.90] 0.030 | 1.40 [1.00–1.95], 0.047 |
| No | 1762 (83.5) | 10.4 (8.7) | 1 (ref) | 1 (ref) |
| Number of male sexual partners in the previous three months | ||||
| ≤2 | 1497 (70.9) | 10.3 (9.2) | 1 (ref) | 1 (ref) |
| >2 | 609 (28.9) | 9.9 (7.7) | 1.26 [0.99–1.61], 0.058 | 1.37 [1.03–1.83], 0.031 |
| HIV risk perception with stable male partner | ||||
| No risk | 760 (36.0) | 10.2 (9.1) | 1.41 [1.10–1.81], 0.007 | 1.61 [1.23–2.10], <0.001 |
| At risk | 790 (35.4) | 1 (ref) | 1 (ref) | |
| No stable partner | 597 (28.3) | 9.1 (7.7) | 1.04 [0.78–1.39], 0.766 | 1.04 [0.73–1.46], 0.843 |
| HIV risk perception with casual male partners | ||||
| No risk | 321 (15.2) | 8.6 (7.6) | 1.08 [0.78–1.49], 0.639 | |
| At risk | 728 (34.5) | 9.7(7.8) | 1 (ref) | |
| No casual partners | 1055 (50.0) | 10.9 (9.5) | 0.79 [0.60–1.03], 0.083 | |
| Combined prevention use during most recent anal intercourse | ||||
| No PrEP & no condom use | 801 (37.9) | 8.35 (9.2) | 2.25 [1.71–2.96], <0.001 | 2.56 [1.92–3.42], <0.001 |
| PrEP use only | 367 (17.4) | 13.1 (7.2) | 1.32 [0.95–1.84], 0.100 | 1.50 [1.05–2.13], 0.025 |
| Condom use only | 267 (12.7) | 5.6 (6.8) | 2.21 [1.52–3.23], <0.001 | 2.19 [1.47–3.25], <0.001 |
| PrEP & condom use | 635 (30.1) | 13.7 (8.1) | 1 (ref) | 1 (ref) |
1 Time-varying Cox’s proportional hazards model adjusted for country-fixed effects and recruitment type, AIC = 3507. 2 Visits from non-LTFU participants accounted for 2721 measures; visits from LTFU participants accounted for 2110 measures; the difference between the total sample and the total number of measures for each variable corresponds to missing values. SD: standard deviation; HR: hazard ratio; aHR: adjusted hazard ratio; CI: confidence interval; ref: reference; CohMSM: cohort of men who have sex with men; PrEP: pre-exposure prophylaxis; PE: peer educator; PHQ-9: Patient Health Questionnaire-9.