Comparing methods for handling missing cost and quality of life data in the Early Endovenous Ablation in Venous Ulceration trial

Table 1 Overview of approaches employed to handle missing data

	RMM and RMFE	CCA	MILR and MIPMM	BPA
Number of patients included at 3 years	450	44	450	450
Total number of non-missing observations included at 3 years^a	1929 EQ-5D, 6861 period costs	44 total costs, 44 QALY	450 EQ-5D, 450 period costs	377 total costs, 44 QALY
Format of data as input	Longitudinal	Aggregate	Longitudinal	Aggregate
Statistical model of the missing data	Implicit imputation of missing EQ-5D and period costs	None	Explicit imputation of missing EQ-5D and period costs	Logit model of probability of missingness
How are total costs and QALY over the desired time horizon predicted at individual level?	Not necessary	Not done	Passively in each imputed dataset	Missing total cost and QALY are parameters to estimate
How are mean total incremental costs and QALY over the desired time horizon estimated	Weighted sum of EQ5D and period cost coefficients estimated in the statistical model	Bivariate normal regression	Bivariate normal regression for each imputed dataset, synthesised using Rubin’s rules	Bivariate normal regression
Estimation of standard errors and CEAC	Bootstrap	Parametrically	Parametrically	Parametrically

^aIf aggregate data are used, there will be one observation per patient. If longitudinal data are used, the inputs to the model may consist of several observations per patient
RMM repeated measure mixed model, RMFE repeated measure fixed effect, CCA complete-case-analysis, MIPMM multiple imputation using predictive mean matching, MILR multiple imputation using linear regression, BPA Bayesian parametric approach

ISSN: 1478-7547