a systematic review and critical appraisal of economic evaluations in systemic lupus erythematosus

Background: New strategies to manage systemic lupus erythematosus (SLE), including the use of biomarkers to target novel or existing therapies, will require evidence of relative cost-effectiveness before being recommended in routine clinical practice. Decision-analytic model-based economic evaluations can synthesise all available evidence to estimate the cost-effectiveness of health technologies. Complexities in the diagnosis, management, and progression of disease pose challenges when estimating the cost-effectiveness of care for SLE. No systematic appraisal of economic evaluations in SLE has been published to date. Objectives: To identify and critically appraise all economic evaluations of treatments for SLE. Methods: A systematic review of published economic evaluations in SLE was performed. Studies were included if they had reported a full economic evaluation of any pharmacological therapy for SLE. Medline and Embase were searched electronically from inception until November 2016. The search strategy comprised disease-specific terms for SLE and published filters to identify economic evaluations. Abstracts were screened independently by two reviewers and read in full by one reviewer. Key features (study characteristics, data sources, methods of analysis, and results) were extracted from each economic evaluation. The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement was used to appraise whether each economic evaluation had reported eighteen items with respect to its methods and results in full, partially, or not at all. Results: The search strategy identified 2,001 abstracts and six published economic evaluations of treatments for SLE were included in the systematic review. These studies considered azathioprine (n=4), mycophenolate mofetil (n=3), cyclophosphamide (n=2), and belimumab (n=1) as relevant comparator therapies. The types of decision-analytic model included individual patient-level simulations (n=3), decision trees (n=2), and a cohort Markov model (n=1). Six elements of the CHEERS statement were reported incompletely across the sample: (1) target population, (2) choice of comparators, (3) measurement and valuation of preference-based outcomes, (4) estimation of resource use and costs, (5) choice and structure of the decision-analytic model, and (6) characterisation of heterogeneity. Conclusions: The choice of treatments that are available currently for SLE are limited and this is reflected in the quantity of economic evaluations published to date. The incomplete reporting of methods within these economic evaluations highlighted notable gaps within the literature. Deficiencies in the evidence base manifest as parameter and structural uncertainties within decision-analytic model-based economic evaluations which, ultimately, affect the estimated expected cost-effectiveness of care. Greater use of existing datasets for SLE, including those from randomised controlled trials and observational cohort studies, can reduce these uncertainties in subsequent economic evaluations of strategies to manage SLE. Acknowledgements: SPG is supported by two grants awarded to The University of Manchester for MASTERPLANS, funded by the Medical Research Council [grant number MR/M01665X/1] and Lupus UK. Disclosure of Interest: None declared DOI: 10.1136/annrheumdis-2018-eular.6460 Citation: Ann Rheum Dis, volume 77, supplement Suppl, year 2018, page A827Session: Public health, health services research and health economics