Addressing missing data in CEA
An outstanding methodological concern in health economic evaluation is that there may be missing data, for example, because patients are lost to follow-up or fail to respond to quality-of-life or resource use questionnaires.
The main problem that arises with missing data is that individuals with missing information tend to be systematically different from those with complete data. However, most published studies fail to address this issue and report inferences based solely on the complete cases. Inappropriate methods may lead to biased results, and ultimately can affect the decision of whether an intervention should be prioritised.
While standard multiple imputation methods have been proposed for handling missing data in cost-effectiveness analysis, these may be insufficient in many settings. For example, they assume that individual observations are independent (which may be implausible in multicentre studies or meta-analysis of individual-participant data) or that the imputation model is correctly specified. In addition, the methods proposed assumed that data are missing at random, i.e. the probability of missingness is only conditional on the observed data. However, the probability of missing costs or outcomes may depend on unobserved values, i.e. data may be missing not at random.
As the true missing data mechanism is unknown with the data at hand, it is important to examine whether cost-effectiveness inferences are robust to alternative assumptions concerning the reasons for the missing data. This is an area of rapid development in biostatistics, but this form of structural uncertainty has not been addressed in health economics.
This research will develop appropriate methods for handling missing data across a wide range of circumstances in health economic evaluation.