Methodological Issues Surrounding the Use of Baseline Health-Related Quality of Life Data to Inform Trial-Based Economic Evaluations of Interventions Within Emergency and Critical Care Settings: A Systematic Literature Review

Background Trial-based cost-utility analyses require health-related quality of life data that generate utility values in order to express health outcomes in terms of quality-adjusted life years (QALYs). Assessments of baseline health-related quality of life are problematic where trial participants are incapacitated or critically ill at the time of randomisation. This review aims to identify and critique methods for handling non-availability of baseline health-related quality of life data in trial-based cost-utility analyses within emergency and critical illness settings. Methods A systematic literature review was conducted, following PRISMA guidelines, to identify trial-based cost-utility analyses of interventions within emergency and critical care settings. Databases searched included the National Institute for Health Research (NIHR) Journals Library (1991–July 2016), Cochrane Library (all years); National Health Service (NHS) Economic Evaluation Database (all years) and Ovid MEDLINE/Embase (without time restriction). Strategies employed to handle non-availability of baseline health-related quality of life data in final QALY estimations were identified and critiqued. Results A total of 4224 published reports were screened, 19 of which met the study inclusion criteria (mean trial size 1670): 14 (74 %) from the UK, four (21%) from other European countries and one (5%) from India. Twelve studies (63%) were based in emergency departments and seven (37%) in intensive care units. Only one study was able to elicit patient-reported health-related quality of life at baseline. To overcome the lack of baseline data when estimating QALYs, eight studies (42%) assigned a fixed utility weight corresponding to either death, an unconscious health state or a country-specific norm to patients at baseline, four (21%) ignored baseline utilities, three (16%) applied values from another study, one (5%) generated utility values via retrospective recall and one (5%) elicited utilities from experts. A preliminary exploration of these methods shows that incremental QALY estimation is unlikely to be biased if balanced trial allocation is achieved and subsequent collection of health-related quality of life data occurs at the earliest possible opportunity following commencement of treatment, followed by an adequate number of follow-up assessments. Conclusion Trial-based cost-utility analyses within emergency and critical illness settings have applied different methods for QALY estimation, employing disparate assumptions about the health-related quality of life of patients at baseline. Where baseline measurement is not practical, measurement at the earliest opportunity following commencement of treatment should minimise bias in QALY estimation.