Including adverse drug events in economic evaluations of anti-tumour necrosis factor-α drugs for adult rheumatoid arthritis: a systematic review of economic decision analytic models.

BACKGROUND: Anti-tumour necrosis factor-α drugs (anti-TNFs) have revolutionised the treatment of rheumatoid arthritis (RA). More effective than standard non-biological disease-modifying anti-rheumatic drugs (nbDMARDs), anti-TNFs are also substantially more expensive. Consequently, a number of model-based economic evaluations have been conducted to establish the relative cost-effectiveness of anti-TNFs. However, anti-TNFs are associated with an increased risk of adverse drug events (ADEs) such as serious infections relative to nbDMARDs. Such ADEs will likely impact on both the costs and consequences of anti-TNFs, for example, through hospitalisations and forced withdrawal from treatment.
OBJECTIVE: The aim of this review was to identify and critically appraise if, and how, ADEs have been incorporated into model-based cost-effectiveness analyses of anti-TNFs for adult patients with RA.
METHODS: A systematic literature review was performed. Electronic databases (Ovid MEDLINE; Ovid EMBASE; Web of Science; NHS Economic Evaluations Database) were searched for literature published between January 1990 and October 2013 using electronic search strategies. The reference lists of retrieved studies were also hand searched. In addition, the National Institute for Health and Care Excellence technology appraisals were searched to identify economic models used to inform UK healthcare decision making. Only full economic evaluations that had used an economic model to evaluate biological DMARDs (bDMARDs) (including anti-TNFs) for adult patients with RA and had incorporated the direct costs and/or consequences of ADEs were critically appraised. To be included, studies also had to be available as a full text in English. Data extracted included general study characteristics and information concerning the methods used to incorporate ADEs and any associated assumptions made. The extracted data were synthesised using a tabular and narrative format.
RESULTS: A total of 43 model-based economic evaluations of bDMARDs for adult RA were identified from 2,483 initially identified studies (2,473 published; ten technology appraisals). Of these, nine studies had incorporated the incidence and costs of ADEs and were critically reviewed. One study also explicitly estimated the potential consequences for patient utility. There was a general lack of detail specifically reporting on how ADEs were included in the economic models. Furthermore, there was substantial heterogeneity amongst the nine studies concerning the (i) application of risk-related terminology; (ii) method of incorporating the incidence, costs and consequences of ADEs; and (iii) ADE-related assumptions.
CONCLUSIONS: Model-based economic evaluations have played an integral role in healthcare reimbursement and funding decisions relating to anti-TNFs for adult patients with RA. However, current economic models have not routinely or systematically considered the direct costs or consequences of ADEs, which may bias the estimates of the relative cost-effectiveness of anti-TNFs. Omitting information on relevant costs and consequences of interventions for RA will affect the validity of the associated recommendations for informed decision making. To improve current practice it is recommended that (i) greater efforts be made to provide appropriate long-term safety data on the use of anti-TNFs in adult RA; (ii) empirical research be undertaken to identify and quantify the impact of, and possible methods for, including ADEs in economic models to inform future good practice guidelines; and (iii) economic modelling guidelines and reference cases be updated to explicitly identify ADEs as an important treatment outcome and address how they might be incorporated into economic models. Improved consideration of the possible implications of ADEs in economic models will ensure that healthcare decision makers are provided with reliable and accurate information with which to make efficient reimbursement and financing decisions.