Impact of missing data on bias and precision when estimating change in patient-reported outcomes from a clinical registry

Table 3 Mixed-Effect Regression Model Parameter Estimates for the SF-12v2 PCS and MCS Scores

	Model Effect	CCA		ML		MI
	Model Effect	Estimate (SE)	CI Width	Estimate (SE)	CI Width	Estimate (SE)	CI Width
PCS	β ₀	37.85 (1.10)	4.32	35.91 (0.86)	3.36	37.72 (0.92)	3.65
	β _T	12.42 (0.30)	1.18	12.67 (0.27)	1.04	12.57 (0.26)	1.03
	β _TS	−1.03 (0.39)	1.53	−1.35 (0.35)	1.36	−1.34 (0.35)	1.39
MCS	β ₀	54.20 (1.32)	5.18	50.86 (1.07)	4.19	51.15 (1.07)	4.22
	β _T	−0.24 (0.29)	1.13	0.01 (0.27)	1.06	0.004 (0.25)	0.99
	β _TS	1.13 (0.38)	1.48	1.34 (0.35)	1.39	1.48 (0.33)	1.32

CCA Complete Case Analysis, ML Maximum Likelihood, MI Multiple Imputation, HBP High Blood Pressure, CI Width Width of the 95% confidence interval, SE Standard Error, Boldface font indicates statistically significant estimates at α = 0.05

ISSN: 1477-7525