Table 2.
Results of meta-regression analyses assessing the association between HIV prevalence and HSV-2 prevalence among female sex workers globally.
| Factors | Studies | Samples | Univariable analyses | Multivariable analysis-model 1 | Multivariable analysis-model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total n | Total n | OR (95% CI) | p value | F p valuea | Adj. R2 (%) | AOR (95% CI) | p value | F p valueb | AORc (95% CI) | p value | F p valueb | |
| HSV-2 prevalence | ||||||||||||
| < 25% | 8 | 2410 | 1.0 | < 0.01 | 10.8 | 1.0 | < 0.01 | – | – | – | ||
| 25–49% | 23 | 8334 | 1.4 (0.4–5.0) | 0.60 | 2.8 (1.2–6.3) | 0.01 | – | – | – | |||
| 50– 74% | 92 | 35,830 | 4.0 (1.3–12.6) | 0.02 | 5.2 (2.4–11.3) | < 0.01 | – | – | – | |||
| 75–100% | 71 | 21,072 | 7.2 (2.3–22.7) | < 0.01 | 13.4 (6.1–29.9) | < 0.01 | – | – | – | |||
| HSV-2 prevalencec | ||||||||||||
| 191 | 66,239 | 1.03 (1.02–1.04) | < 0.01 | < 0.01 | 10.8 | – | – | – | 1.04 (1.03–1.05) | < 0.01 | < 0.01 | |
| WHO region | ||||||||||||
| AMRO | 41 | 12,037 | 1.0 | < 0.01 | 48.5 | 1.0 | < 0.01 | 1.0 | < 0.01 | |||
| AFRO | 41 | 12,998 | 31.7 (19.0–53.0) | < 0.01 | 37.1 (23.2–59.4) | < 0.01 | 36.2 (23.6–55.7) | < 0.01 | ||||
| EURO | 3 | 718 | 1.3 (0.3–5.2) | 0.70 | 3.5 (1.0–12.1) | 0.05 | 5.5 (1.7–17.5) | < 0.01 | ||||
| SEARO | 71 | 24,047 | 8.5 (5.4–13.4) | < 0.01 | 11.2 (7.0–17.8) | < 0.01 | 12.5 (8.1–19.1) | < 0.01 | ||||
| WPRO | 38 | 17,846 | 3.8 (2.3–6.4) | < 0.01 | 5.8 (3.4–9.9) | < 0.01 | 6.2 (3.8–10.0) | < 0.01 | ||||
| Publication year | ||||||||||||
| < 2000 | 15 | 5049 | 1.0 | 0.83 | 0.0 | – | – | – | – | – | – | |
| 2000–2004 | 7 | 2368 | 1.2 (0.3–5.5) | 0.80 | – | – | – | – | – | – | ||
| 2005–2009 | 56 | 13,855 | 1.7 (0.6–4.3) | 0.30 | – | – | – | – | – | – | ||
| 2010–2014 | 99 | 40,760 | 1.4 (0.6–3.6) | 0.44 | – | – | – | – | – | – | ||
| 2015– 2019 | 17 | 5614 | 1.2 (0.4–3.7) | 0.81 | – | – | – | – | – | – | ||
| Data collection yeard | ||||||||||||
| < 1995 | 18 | 6478 | 1.0 | 0.11 | 1.8 | 1.0 | 0.15 | 1.0 | 0.17 | |||
| 1995–1999 | 14 | 2462 | 1.0 (0.3–3.1) | 0.97 | 0.7 (0.3–1.5) | 0.31 | 0.7 (0.3–1.3) | 0.26 | ||||
| 2000–2004 | 61 | 15,736 | 1.0 (0.4–2.5) | 0.93 | 1.0 (0.6–1.7) | 0.98 | 1.0 (0.6–1.6) | 0.85 | ||||
| 2005–2009 | 88 | 37,770 | 2.0 (0.9–4.6) | 0.10 | 1.4 (0.8–2.5) | 0.30 | 1.3 (0.8–2.3) | 0.34 | ||||
| 2010–2014 | 13 | 5200 | 1.4 (0.4–4.6) | 0.56 | 0.6 (0.3–1.3) | 0.19 | 0.6 (0.3–1.3) | 0.22 | ||||
| Sample size | ||||||||||||
| < 200 | 52 | 5507 | 1.0 | 0.76 | 0.0 | – | – | – | – | – | – | |
| ≥ 200 | 142 | 62,139 | 0.92 (0.54–1.56) | 0.76 | – | – | – | – | – | – | ||
| Proportion of FSWs reporting consistent condom use | ||||||||||||
| 75–100% | 77 | 31,462 | 1.0 | 0.09 | 2.1 | 1.0 | 0.09 | 1.0 | 0.08 | |||
| 50–74% | 19 | 8129 | 0.4 (0.2–1.0) | 0.04 | 1.1 (0.6–1.9) | 0.79 | 1.2 (0.7–2.0) | 0.55 | ||||
| 25–49% | 31 | 6367 | 1.4 (0.7–2.7) | 0.38 | 1.8 (1.0–3.2) | 0.05 | 1.9 (1.1–3.2) | 0.02 | ||||
| < 25% | 9 | 3715 | 1.1 (0.3–3.3) | 0.91 | 0.7 (0.3–1.5) | 0.31 | 0.8 (0.4–1.6) | 0.51 | ||||
| Unclear | 58 | 17,973 | 0.7 (0.4–1.2) | 0.19 | 1.3 (0.8–2.1) | 0.30 | 1.2 (0.8–1.8) | 0.47 | ||||
Adj, Adjusted; AFRO, African Region; AMRO, Region of the Americas; AOR, adjusted odds ratio; CI, confidence interval; EURO, European Region; FSWs, female sex workers; HSV-2, herpes simplex virus type 2; OR, odds ratio; SEARO, South-East Asia Region; WHO, World Health Organization; WPRO, Western Pacific Region.
Adjusted R2 is 65.3% in the multivariable model 1, and 70.6% in the multivariable model 2.
aFactors with p value ≤ 0.2 were eligible for inclusion in the multivariable analysis.
bFactors with p value ≤ 0.05 and those with 0.05 < p value ≤ 0.1 in the multivariable model were considered as showing, respectively, “strong” and “some” evidence for an association with HIV prevalence.
cAnalysis of the association with HSV-2 prevalence as a linear term excluded three measures with HSV-2 prevalence ≤ 20% in light of the observed threshold effect.
dMissing values for year of data collection were imputed using data for year of publication adjusted by the median difference between year of publication and median year of data collection for studies with complete information.