Table 1 Meta-regression results for the impact of risk of bias indicators on TCOC effect size in dollars per beneficiary per quarter (N = 82)
From: Assessing risk of bias in the meta-analysis of round 1 of the Health Care Innovation Awards
ROB variable | Mean or percent | Coefficient estimate ($) | Standard error | p value |
|---|---|---|---|---|
Intercept | –- | 43.8 | 48.5 | 0.37 |
PS-weighting method (PSW) | 10.1 |  − 328.9 | 613.6 | 0.59 |
Potential favorable bias (PFB) | 27.0 |  − 59.3 | 63.4 | 0.35 |
Risk of bias, direction unclear | 5.6 |  − 21.9 | 91.0 | 0.81 |
Percent unbalanced covariates | 9.6 |  − 0.89 | 1.71 | 0.61 |
Barriers to enrollment | 25.8 |  − 78.8 | 97.0 | 0.42 |
New innovation | 30.3 | 156.8 | 70.1 | 0.03 |