Table 3. Certainty of evidence of effect of peer-support for people living with HIV.
|
Population: People living with HIV Countries: China, Kenya, Mozambique, Nigeria, Spain, Uganda, USA, Vietnam Intervention: Peer-support Comparison: Usual care or education/counselling | |||||
| Outcome, follow-up time | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | No. of participants (Studies) | Quality of evidence (GRADE) | |
| Assumed risk with control | Assumed risk with peer-support | ||||
| Retention in care | |||||
| Retention in care (6 mo) | 64.6% | 67.1% | RR = 1.05 (0.92, 1.20) | 1916 (4 RCTs) | ⨁⨁⨁◯ MODERATE 1,2 |
| Retention in care (12 mo) | 74.8% | 80.0% | RR = 1.07 (1.02, 1.12) | 1556 (3 RCTs) | ⨁⨁⨁⨁ HIGH |
| ART initiation | |||||
| ART initiation (12 mo) | 24.0% | 18.8% | RR = 0.99 (0.74, 1.32) | 180 (2 RCTs) | ⨁⨁◯◯ LOW 1,3 |
| ART adherence | |||||
| ART adherence (3 mo) | 78.2% | 81.8% | RR = 1.06 (1.01,1.10) | 1282 (4 RCTs) | ⨁⨁⨁⨁ HIGH |
| ART adherence (6 mo) | 72.3% | 74.8% | RR = 1.03 (0.97, 1.08) | 1823 (6 RCTs) | ⨁⨁⨁◯ MODERATE 3,4 |
| ART adherence (12 mo) | 87.3% | 92.1% | RR = 1.05 (0.98, 1.12) | 1140 (4 RCTs) | ⨁⨁⨁◯ MODERATE 1 |
| Pill count adherence <95% (12 mo) | 2.4% | 1.4% | RR = 0.57 (0.23, 1.37) | 1336 (1 RCT) | ⨁⨁◯◯ LOW 5 |
| Pill count adherence <100% (12 mo) | 23.3% | 25.5% | RR = 1.09 (0.87, 1.37) | 1336 (1 RCT) | ⨁⨁⨁◯ MODERATE 3 |
| Any missed doses (12 mo) | 19.2% | 17.6% | RR = 0.91 (0.71, 1.19) | 1336 (1 RCT) | ⨁⨁⨁◯ MODERATE 3 |
| CD4 cell count | |||||
| CD4 cell count (6–24 mo) | Estimates shown in Table 2. No studies found a stat.sign. difference between the groups | 2733 (5 RCTs) | ⨁⨁◯◯ LOW 3,6 |
||
| Viral load/suppression/failure | |||||
| Viral suppression (3 mo) | RR = 0.98 (0.85, 1.14) to RR = 1.31 (1.03, 1.67). See Table 2. | 283 (2 RCTs) | ⨁⨁◯◯ LOW 1,3 |
||
| Viral suppression (6 mo) | RR = 1.02 (0.94, 1.11) to OR = 6.24 (1.28–30.5). See Fig 6 and Table 2. | 704 (7 RCTs) | ⨁⨁⨁⨁ HIGH |
||
| Viral suppression (12 mo) | RR = 0.80 (0.64, 0.99) to RR = 1.38 (1.03, 1.85). See Table 2. | 494 (2 RCTs) | ⨁⨁◯◯ LOW 1,3 |
||
| Virologic failure (6 mo) | 7.0% | 6.0% | RR = 0.93 (0.60, 1.45) | 1275 (2 RCTs) | ⨁⨁◯◯ LOW 3,6 |
| Virologic failure (12 mo) | 6.8% | 6.4% | RR = 0.79 (0.53, 1.19) | 1468 (3 RCTs) | ⨁⨁◯◯ LOW 3,6 |
| Virologic failure (18 mo) | 2.2% | 3.5% | RR = 1.23 (0.34, 4.45) | 1162 (2 RCTs) | ⨁◯◯◯ VERY LOW 5,6 |
| Virologic failure (24 mo) | 4.3% | 3.8% | RR = 0.54 (0.31, 0.94) | 1172 (2 RCTs) | ⨁⨁⨁◯ MODERATE 6 |
| Quality of life | |||||
| QoL–mental (12 mo) | Estimates shown in Fig 8. | SMD = -0.06 (-0.27, 0.15) | 251 (2 RCTs) | ⨁⨁◯◯ LOW 1,3,4 |
|
| QoL–physical (12 mo) | Estimates shown in Fig 8. | SMD = -0.04 (-0.35, 0.28) | 251 (2 RCTs) | ⨁⨁◯◯ LOW 1,3,4 |
|
| Mental health | |||||
| Depressive symptoms (6–12 mo) | Estimates shown in Table 2. No studies found a stat.sign. difference between the groups | 338 (2 RCTs) | ⨁◯◯◯ VERY LOW 5,6 |
||
1. Downgraded by 1 level because of inconsistency.
2. 1 other RCT found a stat.sign. difference in favour of the intervention (see Table 2)
3. Downgraded by 1 level because of imprecision.
4. 1 other RCT found no stat.sign. difference between the groups (see Table 2)
5. Downgraded by 2 levels because of imprecision.
6. Downgraded by 1 level because of risk of bias.
Legend. CI: Confidence interval; RCT: Randomised controlled study; SD: Standard deviation.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).