Skip to main content

Table 3 Parameter estimates (analogous to Table 2) for the regression analyses for the subgroups of “true placebos” and “ICS-placebos”

From: Decline of COPD exacerbations in clinical trials over two decades – a systematic review and meta-regression

Parameter Estimate Implied annual percentage change Pb
True placebos (21 studies):
 Rate intercept β0 0.440 (− 0.034, 0.912)   
 Rate slope β1 (publication year) −0.079 (− 0.132, − 0.026) − 7.553 (− 12.356, − 2.606) 0.006
 Rate random effect σλ 0.517 (0.356, 0.787)   
 Overdispersion mean μφ −5.139 (− 9.011, − 0.708)   
 Overdispersion random effect σφ 0.667 (0.031, 2.263)   
ICS-placebos (34 studies):
 Rate intercept β0 0.440 (−0.034, 0.912)   
 Rate slope β1 (publication year) −0.079 (− 0.132, − 0.026) −7.553 (− 12.356, − 2.606) <  0.001
 Rate random effect σλ 0.517 (0.356, 0.787)   
 Overdispersion mean μφ −5.139 (− 9.011, − 0.708)   
 Overdispersion random effect σφ 0.667 (0.031, 2.263)