Table 2. Acceptability of and preferences for HIV self-testing according to pre-testing and post-testing phases among 597 adolescents living in Kisangani.
| Variable | Pre-testing phaseμ | Post-testing phase# | Difference£ | P-value* |
|---|---|---|---|---|
| n (%) | n (%) | % [95% CI] | ||
| Reasons to accept HIVST | ||||
| ○ Easy to use | 419 (67.8) | 570 (95.5) | −27.7 [−31.4, −24.3] | < 0.0001 |
| ○ Fast results | 347 (56.1) | 383 (64.2) | −8.1 [−10.6, −6.2] | 0.008 |
| ○ Confidential | 602 (97.4) | 588 (98.5) | −1.1 [−2.3, −0.5] | 0.254 |
| ○ Others** | 64 (10.4) | 67 (11.2) | −0.8 [−1.9, −0.3] | 0.867 |
| Acceptability of recommending self-testing to other | 515 (82.5) | 575 (96.3) | −13.8 [−16.8, −11.3] | 0.004 |
| Acceptability of distributing the self-test to other | 492 (78.8] | 530 (88.8) | −10.0 [−12.7, −7.8] | 0.006 |
| Substitution of VCT for HIV self-testing | 344 (55.8) | 516 (86.4) | −30.6 [−34.4, −27.0] | < 0.0001 |
| Willingness to buy HIV self-test | 261 (42.2) | 273 (45.7) | −3.6 [−5.4, −2.4] | 0.196 |
| Preference of pre-test counseling | ||||
| ○ Not useful | 101 (16.4) | 274 (45.9) | −29.5 [−33.3, −26.0] | < 0.0001 |
| ○ Rather useful | 208 (33.8) | 141 (23.6) | 10.2 [8.0, 12.9] | 0.005 |
| ○ Useful | 170 (27.6) | 120 (20.1) | 7.5 [5.6, 9.9] | 0.011 |
| ○ Essential | 137 (22.2) | 62 (10.4) | 11.8 [9.5, 14.6] | 0.005 |
| Preference of post-test counseling | ||||
| ○ Not useful | 33 (5.4) | 9 (1.5) | 3.9 [2.6, 5.8] | 0.078 |
| ○ Rather useful | 48 (7.8) | 21 (3.5) | 4.3 [2.9, 6.2] | 0.063 |
| ○ Useful | 248 (40.3) | 161 (27.0) | 13.3 [10.8, 16.3] | 0.004 |
| ○ Essential | 286 (46.5) | 406 (68.0) | −21.5 [−25.0, −18.4] | 0.001 |
| Preference for different ways of doing counseling | ||||
| ○ Face-to-face | 482 (78.3) | 471 (78.9) | −0.6 [−1.6, 0.2] | 0.897 |
| ○ Via telephone | 120 (19.4) | 111 (18.6) | 0.8 [0.3, 1.9] | 0.886 |
| ○ Via internet | 14 (2.3) | 15 (2.5) | −0.2 [−0.68, 0.28] | 0.991 |
μ Missing values were excluded from the Table 2 in pre-testing phase;
# All of 597 participants assessed in post-testing phase evaluation without missing values;
£ Difference assessed with Wald asymptotic test using only data collected from all 597 participants in the post-testing phase paired to those from the pre-testing phase;
* P-value calculated using Mac Nemar’s test of paired data;
** Other reasons included such as no stigma and discrimination, and curiosity to use HIV self-testing.
CI: Confidence interval; NS: Not significant; MV: Missing values; VCT: Voluntary counseling and testing.