Table 2. Multilevel univariate and multivariate meta-regression models of the various factors related to the prevalence of pterygium.
| Moderator | Number of studies | β | 95% CI | P value |
|---|---|---|---|---|
| Univariate meta regression | ||||
| Intercept | 47 | -2.008 | [-2.346]-[-1.670] | <0.001 |
| Age | 47 | 0.042 | 0.040–0.043 | <0.001 |
| Gender-Male$ | 40 | 0.115 | 0.083–0.147 | <0.001 |
| Latitude | 47 | -0.072 | [-0.114]-[-0.030] | <0.001 |
| Multivariate meta regression# | ||||
| Intercept | 40 | -2.069 | [-3.735]-[-0.403] | 0.015 |
| Age | 40 | 0.042 | 0.040–0.043 | <0.001 |
| Gender-Male | 40 | 0.105 | 0.073–0.137 | <0.001 |
| Latitude | 40 | -0.069 | [-0.117]-[-0.022] | 0.004 |
$ the estimate of gender effect was based on studies that provided pterygium prevalence for both males and females.
# the multivariate model was based on studies that provided pterygium prevalence for both males and females; coefficients represent log odds ratios (ORs).