The Use of Decision-Analytic Models in Atopic Eczema: A Systematic Review and Critical Appraisal

Abstract

Objective: The objective of this systematic review was to identify and assess the quality of published economic decision-analytic models within atopic eczema against best practice guidelines, with the intention of informing future decision-analytic models within this condition.

Methods: A systematic search of the following online databases was performed: MEDLINE, EMBASE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Cochrane Central Register of Controlled Trials, Database of Abstracts of Reviews of Effects, Cochrane Database of Systematic Reviews, NHS Economic Evaluation Database, EconLit, Scopus, Health Technology Assessment, Cost-Effectiveness Analysis Registry and Web of Science. Papers were eligible for inclusion if they described a decision-analytic model evaluating both the costs and benefits associated with an intervention or prevention for atopic eczema. Data were extracted using a standardised form by two independent reviewers, whilst quality was assessed using the model-specific Philips criteria.

Results: Twenty-four models were identified, evaluating either preventions (n = 12) or interventions (n = 12): 14 reported using a Markov modelling approach, four utilised decision trees and one a discrete event simulation, whilst five did not specify the approach. The majority, 22 studies, reported that the intervention was dominant or cost effective, given the assumptions and analytical perspective taken. Notably, the models tended to be short-term (16 used a time horizon of ≤1 year), often providing little justification for the limited time horizon chosen. The methodological and reporting quality of the studies was generally weak, with only seven studies fulfilling more than 50% of their applicable Philips criteria.

Conclusions: This is the first systematic review of decision models in eczema. Whilst the majority of models reported favourable outcomes in terms of the cost effectiveness of the new intervention, the usefulness of these findings for decision-making is questionable. In particular, there is considerable scope for increasing the range of interventions evaluated, for improving modelling structures and reporting quality.