15.4.2  Collecting data

Precise data collection requirements for the economics components of Cochrane reviews will need to be specified for each individual review, depending on the specific economics question or objective and on the measures of incremental resource use, costs or cost-effectiveness included as target outcomes. In general terms, two types of data will need to be collected: details of the characteristics of included health economics studies and details of their results. The potential to extract data as suggested below from published reports may be constrained by the quality of reporting of the health economics studies (where information is missing, a further option is to contact study authors to request additional details).

 

Useful data to be collected regarding the characteristics of each economic study are likely to include: year of study; details of interventions and comparators; study design and source(s) of resource use, unit costs and (if applicable) effectiveness data (see also Sections 15.1.2 and 15.2.3); decision-making jurisdiction, geographical and organizational setting; analytic viewpoint; and time horizon for both costs and effects (see Section 15.2.1).

 

For results, estimates of specific items of resource use associated with interventions and comparators and estimates of their unit costs should be extracted separately, if reported, as well as estimates of costs of the resource use (i.e. number of units of resource X unit cost). The type and quantity of each resource used should be extracted in natural units (e.g. length of hospital stay in days, duration of operation in minutes, number of outpatient attendances at six-month follow-up, number of days of work). It is also important to collect information on the price year and currency used to calculate estimates of costs and incremental costs. Measures of incremental resource use and costs should be collected at the individual patient level (i.e. resource use per patient, cost per patient), wherever possible. Both a point estimate and a measure of uncertainty (e.g. standard error or confidence interval) should be extracted for measures of incremental resource use, costs and cost-effectiveness, if reported. Additionally, it is useful to collect details of any sensitivity analyses undertaken, and any information regarding the impact of varying assumptions on the magnitude and direction of results.

 

CRD Report 6 (Craig 2007) includes a template for producing structured abstracts of full economic evaluations for inclusion in NHS EED (see also Section 15.3.2), together with notes to guide data collection and critical appraisal. These materials can provide a useful template for the design of data collection forms for use in the economics components of Cochrane reviews.

 

If a full economic evaluation already has a corresponding NHS EED structured abstract, this may obviate the need for researchers to undertake further data collection from the study. In parallel, given that critical appraisal and data collection from economic evaluation studies with no completed NHS EED abstract will need to be undertaken for the Cochrane review, authors are encouraged to consider registering with NHS EED to produce an abstract, in order to avoid duplication of effort. Please contact CCEMG for further information, or to initiate a request that a structured abstract is produced by NHS EED (see also Section 15.3.2).