A preliminary stage to be undertaken before addressing risk of bias is to classify the included health economics studies by study design. Methods underpinning critical appraisal of the methodological quality of health economics studies will vary slightly depending on study design.
Classification should consist of two stages:
classification of the design of the health economics study; and
classification of the design of the study generating the effectiveness data on which the health economics study is based, if applicable.
Each health economics study may be classified (stage 1) as a type of full economic evaluation, a type of partial economic evaluation, or a type of effectiveness study (e.g. a randomized trial) reporting more limited information on the resource use or costs associated with an intervention (see Section 15.1.2). Classifying the design of the study that generates the effectiveness data on which the health economics study is based (stage 2) is only applicable in the case of health economics studies classified as a full economic evaluation or as a cost-outcome description at the first stage of classification. The study generating the effectiveness data may be a single study design (e.g. a randomized trial, a non-randomized trial, an observational study) or a synthesis of several studies (e.g. a meta-analysis of randomized trials) (see also Section 15.1.2).
It is likely to be useful to consult with a health economist when undertaking classification of health economics studies. This is because health economics studies reported to use one type of study design (e.g. a cost-benefit analysis) may, on closer inspection, turn out to use another (e.g. a cost-effectiveness analysis). This means that particular care is required when classifying economic studies encountered during a review (Zarnke 1997).
Depending on the scope of the critical review of health economics studies and the types of studies that will be considered for inclusion (see Section 15.2.3), health economics studies may be excluded at this stage, based on classification by study design. Once again, reasons for excluding health economics studies at this stage should be reported in ‘Characteristics of excluded studies’ tables.