Sentence examples for between study estimates was from inspiring English sources
Statistical heterogeneity between study estimates was significant, probably due to variation in measured periods of prevalence and symptom measurement methods.
The degree of heterogeneity between the study estimates was calculated using the I statistic [ 14], a measure describing the percentage of total variation across the studies that can be explained by heterogeneity rather than chance.
On the other hand, the biases of between study variance estimates were between 3.6% and 9.8%, though their 95% CP were within (92.93%, 97.07%).
On the other hand, the biases of between study variance estimates were between 5% and 12.8%, though their 95% CP were all around 95%. Table 4 reported the simulation results.
Our simulation results show that the between-study variance estimates were less frequently truncated at zero in the BRMA than the URMA (Table 4).
These gave findings consistent with those identified previously (Appendix 2 – see Additional file 2); the BRMA was still beneficial over URMA for estimating the pooled endpoints, and where the between-study correlation estimate was often +1 or -1 there was again an upward bias in the BRMA between-study variance estimates.
We conducted meta-analysis even if significant heterogeneity was seen between study estimates, but explored possible reasons in sensitivity analyses, including study quality.
Statistical heterogeneity between study estimates for a particular site of pain was significant (with the exception of chest pain).
The observed heterogeneity between study-specific effect estimates was either high (TNF-α and IL-1β) or moderate (IL-6 and CRP).
The results again show that the between-study correlation estimate is often +1 or -1 and the BRMA is still preferable to URMA, with improved mean-square error, coverage and, especially, bias of estimates (Table 5).
Between-study heterogeneity of effect estimates was examined by creating a multiplicative interaction term between study site (fixed effect covariate) and each risk factor and performing a likelihood ratio test comparing models with and without the risk factor-study site interaction terms.
