THE CHALLENGES IN EVALUATING THE COST-EFFECTIVENESS OF COMPLEX INTERVENTIONS
Author(s)
Faria R1, Weatherly H2, Kiss N3, Manca A1, Parker G2, Beresford B2, Pilkington G4, Laver Fawcett AJ5, Kanaan M2, Rabiee P2, Mann R2, Aspinal F2
1University of York, Heslington, York, UK, 2University of York, York, UK, 3Medical University of Vienna, Vienna, Austria, 4Gerald Pilkington Associates, New Malden, UK, 5York St John University, York, UK
Standard cost-effectiveness methods and critical appraisal toolkits may not be adequate for complex interventions. We systematically reviewed and quality assessed cost-effectiveness studies of a complex intervention, propose a series of new questions to inform their critical appraisal and discuss how future research should be targeted to improve the methods. Reablement was used as an example of a complex intervention. Reablement is a multidisciplinary and multifactorial intervention to support people to relearn activities of daily living. The systematic review identified 12 cost-effectiveness studies on reablement, out of 3,311 unique records. The 12 included studies were data extracted and quality evaluated using a standard checklist. No study provided enough information to inform the decision on whether reablement is cost-effective and should be reimbursed by the payer. The issues included: (i) the use of a perspective not relevant for the decision-maker, (ii) lack of consideration for inter-sectoral effects, (iii) short time horizon, (iv) poor descriptive detail on the interventions, (v) limited comparators, (vi) poor quality evidence on effectiveness, (vii) limited evaluation of uncertainty and (viii) no consideration of the opportunity cost. These issues informed the development of a new checklist, which was subsequently applied. Critical appraisals of cost-effectiveness studies should consider the aforementioned issues to conclude on their quality and potential to inform decision-making. More research is needed how to quantify the opportunity costs of complex interventions, particularly when multiple sectors are affected.
Conference/Value in Health Info
2015-11, ISPOR Europe 2015, Milan, Italy
Value in Health, Vol. 18, No. 7 (November 2015)
Code
PRM256
Topic
Methodological & Statistical Research
Topic Subcategory
Confounding, Selection Bias Correction, Causal Inference
Disease
Multiple Diseases