Table 3.
PrEP adherence, defined as high and medium adherence, any and average use measured by intracellular tenofovir‐diphosphate levels at Months 1, 2, 3, 6 and 12, by study randomized group
|
High adherence by TFV‐DP ≥700 fmol/punch a |
Medium adherence by TFV‐DP ≥350 to 699 fmol/punch b |
TFV‐DP Detected (> 16.6 fmol/punch) |
TFV‐DP fmol/punch Median (IQR) if detected | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Incentive N (%) |
Control N (%) |
Relative risk (95% CI) | p‐value |
Incentive N (%) |
Control N (%) |
Relative risk (95% CI) | p‐value |
Incentive N (%) |
Control N (%) |
Relative risk (95% CI) | p‐value | Incentive | Control | |
| Month 1 | 53/87 (60.9%) | 53/85 (62.4%) | 0.98 (0.77, 1.24) | 0.85 | 75/87 (86.2%) | 74/85 (87.1%) | 0.99 (0.88, 1.11) | 0.87 | 86/87 (98.9%) | 85/85 (100%) | 0.99 (0.97, 1.01) | 0.32 | 646 (375, 934) | 610 (421, 811) |
| Month 2 | 44/81 (54.3%) | 34/75 (45.3%) | 1.20 (0.87, 1.65) | 0.27 | 65/81 (80.3%) | 62/75 (82.7%) | 0.97 (0.84, 1.13) | 0.70 | 81/81 (100%) | 75/75 (100%) | – | – | 792 (413, 1099) | 675 (384, 930) |
| Month 3 | ||||||||||||||
| ITT | 45/81 (55.5%) | 35/85 (41.2%) | 1.35 (0.98, 1.86) | 0.07 | 68/81 (84.0%) | 61/85 (71.8%) | 1.17 (0.99, 1.38) | 0.06 | 80/81 (98.8%) | 82/85 (96.5%) | 1.02 (0.98, 1.07) | 0.33 | 760 (449, 1148) | 609 (325, 969) |
| Month 3 Secondary analysis (Missing = TND) | 45/101 (44.6%) | 35/98 (35.7%) | 1.25 (0.89, 1.76) | 0.21 | 68/101 (67.3%) | 61/98 (62.2%) | 1.08 (0.88, 1.33) | 0.45 | 80/101 (79.2%) | 82/98 (83.7%) | 0.95 (0.83, 1.08) | 0.42 | ||
| Month 6 | 13/69 (18.8%) | 7/69 (10.1%) | 1.86 (0.79, 4.37) | 0.16 | 35/69 (50.7%) | 25/69 (36.2%) | 1.40 (0.95, 2.07) | 0.09 | 60/69 (87.0%) | 54/69 (78.3%) | 1.11 (0.95, 1.30) | 0.18 | 409 (79, 621) | 325 (95, 548) |
| Month 12 | 5/67 (7.5%) | 3/62 (4.8%) | 1.54 (0.39, 6.19) | 0.54 | 13/67 (19.4%) | 11/62 (17.7%) | 1.09 (0.53, 2.26) | 0.81 | 39/67 (58.2%) | 33/62 (53.2%) | 1.09 (0.80, 1.50) | 0.57 | 204 (80, 438) | 199 (43, 410) |
The median and IQR were calculated among detectable results. All analyses were intention‐to‐treat (ITT) and included all randomized participants other than one participant subsequently found to be HIV positive at enrolment. The primary endpoint was high adherence by tenofovir diphosphate (TFV‐DP) concentration in dried blood spots (DBS), at three months.
High adherence was defined for M1 as TFV‐DP ≥ 500 fmol/punch; and for months two through 12 as TFV‐DP ≥ 700 fmol/punch.
Medium adherence was defined for M1 as TFV‐DP ≥ 250 fmol/punch; and for months two through 12 as TFV‐DP ≥ 350 fmol/punch TND denotes target not detected (<16.6 fmol/punch). For the secondary analysis, those who had no DBS measurement, whether due to a missed visit or DBS not being collected at the visit, were categorized as undetectable. The relative risk (RR), confidence interval (CI) and p‐value were generated via modified Poisson regression with robust standard errors comparing the outcome by randomized group.