Skip to main content

Table 4 Meta-analysis of interaction terms coefficient estimate bias (mean [IQR])

From: Statistical approaches to identify subgroups in meta-analysis of individual participant data: a simulation study

  MA-IT BIAS [IQR]
Binary outcome Continuous outcome Survival outcome
Heterogeneity magnitude Ecological Bias Large subgroup effect Medium subgroup effect No sugroup effect Large subgroup effect Medium subgroup effect No sugroup effect Large subgroup effect Medium subgroup effect No sugroup effect
No heterogeneity No ecological bias 0.044 [0.573] 0.015 [0.578] 0.002 [0.496] -0.006 [0.277] -0.011 [0.252] 0.002 [0.25] -0.032 [0.461] -0.014 [0.46] -0.014 [0.408]
No heterogeneity Favouring treatment ecological bias 0.028 [0.543] 0.03 [0.563] -0.003 [0.568] 0.005 [0.273] 0.007 [0.253] 0 [0.241] -0.036 [0.471] -0.01 [0.467] -0.018 [0.424]
No heterogeneity Favouring placebo ecological bias 0.06 [0.54] 0.056 [0.541] -0.019 [0.576] 0 [0.25] 0.007 [0.244] -0.006 [0.254] -0.038 [0.553] 0.003 [0.513] -0.005 [0.374]
Medium heterogeneity No ecological bias 0.018 [0.577] 0.039 [0.514] 0.008 [0.54] -0.003 [0.26] -0.008 [0.249] -0.006 [0.237] -0.058 [0.461] -0.0390 [0.467] -0.016 [0.386]
Medium heterogeneity Favouring treatment ecological bias 0.017 [0.532] 0.015 [0.538] 0.007 [0.567] -0.01 [0.259] 0.005 [0.259] 0.003 [0.25] -0.058 [0.48] -0.0390 [0.497] -0.025 [0.402]
Medium heterogeneity Favouring placebo ecological bias 0.046 [0.53] 0.006 [0.497] -0.019 [0.533] 0.002 [0.25] 0.003 [0.26] -0.01 [0.268] -0.077 [0.566] -0.0381 [0.511] -0.005 [0.37]
Large heterogeneity No ecological bias 0.049 [0.556] -0.004 [0.519] 0.003 [0.573] -0.001 [0.258] 0.002 [0.281] 0.003 [0.257] -0.053 [0.451] -0.039 [0.472] -0.016 [0.38]
Large heterogeneity Favouring treatment ecological bias 0.048 [0.538] 0.016 [0.525] -0.008 [0.588] -0.001 [0.255] -0.008 [0.271] 0.006 [0.256] -0.063 [0.497] -0.0431 [0.492] -0.024 [0.403]
Large heterogeneity Favouring placebo ecological bias 0.063 [0.563] 0.023 [0.508] -0.012 [0.56] 0.004 [0.25] -0.003 [0.257] -0.003 [0.248] -0.077 [0.541] -0.0350 [0.532] -0.006 [0.376]