Table 1.
Covariate | No. of Surveys | Pooled ρ Estimate a | 95% CI | % of Variance Explained b | β c , d | 95% CI |
---|---|---|---|---|---|---|
Country | 58.2 | |||||
Ethiopia | 162 | 0.100 | 0.077, 0.128 | 1.000 | Referent | |
Mozambique | 44 | 0.032 | 0.023, 0.045 | 0.471 | 0.348, 0.638 | |
Nigeria | 55 | 0.010 | 0.008, 0.012 | 0.789 | 0.273, 2.281 | |
No. of children examined per cluster | 44.7 | |||||
15–29 | 36 | 0.097 | 0.071, 0.132 | 1.000 | Referent | |
30–49 | 172 | 0.071 | 0.050, 0.100 | 0.882 | 0.658, 1.181 | |
50–79 | 34 | 0.016 | 0.010, 0.024 | 0.518 | 0.332, 0.809 | |
≥80 | 19 | 0.005 | 0.003, 0.008 | 0.212 | 0.096, 0.470 | |
TFe prevalence, % | 36.2 | |||||
<5.0 | 56 | 0.015 | 0.011, 0.020 | 1.000 | Referent | |
5.0–9.9 | 56 | 0.033 | 0.023, 0.048 | 1.638 | 1.230, 2.181 | |
10.0–29.9 | 87 | 0.080 | 0.057, 0.114 | 2.392 | 1.816, 3.150 | |
≥30.0 | 62 | 0.111 | 0.077, 0.162 | 2.493 | 1.810, 3.432 | |
No. of recordersf | 48.7 | |||||
<5 | 53 | 0.009 | 0.007, 0.012 | 1.000 | Referent | |
5–9 | 43 | 0.057 | 0.039, 0.083 | 2.335 | 0.834, 6.534 | |
10–19 | 149 | 0.082 | 0.061, 0.109 | 2.740 | 0.945, 7.944 | |
≥20 | 16 | 0.106 | 0.063, 0.178 | 2.727 | 0.891, 8.343 | |
No. of children examined per household | 30.1 | |||||
1.0–1.9 | 105 | 0.081 | 0.066, 0.099 | 1.000 | Referent | |
2.0–2.9 | 130 | 0.051 | 0.038, 0.067 | 0.786 | 0.627, 0.986 | |
3.0–3.9 | 19 | 0.006 | 0.003, 0.010 | 0.721 | 0.387, 1.339 | |
≥4.0 | 7 | 0.010 | 0.004, 0.023 | 1.710 | 0.677, 4.316 | |
Quartile of distance between clustersg, km | 13.8 | |||||
1 | 65 | 0.023 | 0.017, 0.030 | 1.000 | Referent | |
2 | 66 | 0.051 | 0.034, 0.076 | 0.974 | 0.734, 1.289 | |
3 | 65 | 0.084 | 0.056, 0.127 | 1.059 | 0.785, 1.430 | |
4 | 65 | 0.065 | 0.043, 0.098 | 0.963 | 0.700, 1.324 |
Abbreviations: CI, confidence interval; GPS, Global Positioning System; ICC, intracluster correlation coefficient; TF, trachomatous inflammation—follicular.
aPooled estimate of the ICC.
bProportion of the variability between survey ICC estimates explained by each covariate on the natural logarithmic scale (P < 0.0001 for all variables).
cExponentiated meta-regression coefficient.
dFull meta-regression model adjusting for country, number of children examined per cluster, and prevalence of TF in children aged 1–9 years (P < 0.0001). 69.2% of the variance in the ICC was explained by the full model.
eTF in children aged 1–9 years, the primary clinical sign associated with ocular Chlamydia trachomatis infection used to guide intervention programs under current World Health Organization guidelines (23).
fEstimated as the number of unique recorder identification codes used in the survey.
gEstimated as the square root of the variance of the distance of survey clusters from the geometric center of the GPS coordinates of all survey clusters, converted to kilometers and accounting for latitude.