The handling of missing data in trial-based economic evaluations: should data be multiply imputed prior to longitudinal linear mixed-model analyses?

Ângela Jornada Ben (), Johanna M. Dongen, Mohamed El Alili, Martijn W. Heymans, Jos W. R. Twisk, Janet L. MacNeil-Vroomen, Maartje Wit, Susan E. M. Dijk, Teddy Oosterhuis and Judith E. Bosmans
Additional contact information
Ângela Jornada Ben: Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute
Johanna M. Dongen: Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute
Mohamed El Alili: Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute
Martijn W. Heymans: Amsterdam Public Health Research Institute
Jos W. R. Twisk: Amsterdam Public Health Research Institute
Janet L. MacNeil-Vroomen: Amsterdam Public Health Research Institute, Amsterdam UMC, University of Amsterdam
Maartje Wit: Amsterdam UMC, Vrije Universiteit, Amsterdam Public Health Research Institute
Susan E. M. Dijk: Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute
Teddy Oosterhuis: Netherlands Society of Occupational Medicine (NVAB)
Judith E. Bosmans: Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute

The European Journal of Health Economics, 2023, vol. 24, issue 6, No 9, 965 pages

Abstract: Abstract Introduction For the analysis of clinical effects, multiple imputation (MI) of missing data were shown to be unnecessary when using longitudinal linear mixed-models (LLM). It remains unclear whether this also applies to trial-based economic evaluations. Therefore, this study aimed to assess whether MI is required prior to LLM when analyzing longitudinal cost and effect data. Methods Two-thousand complete datasets were simulated containing five time points. Incomplete datasets were generated with 10, 25, and 50% missing data in follow-up costs and effects, assuming a Missing At Random (MAR) mechanism. Six different strategies were compared using empirical bias (EB), root-mean-squared error (RMSE), and coverage rate (CR). These strategies were: LLM alone (LLM) and MI with LLM (MI-LLM), and, as reference strategies, mean imputation with LLM (M-LLM), seemingly unrelated regression alone (SUR-CCA), MI with SUR (MI-SUR), and mean imputation with SUR (M-SUR). Results For costs and effects, LLM, MI-LLM, and MI-SUR performed better than M-LLM, SUR-CCA, and M-SUR, with smaller EBs and RMSEs as well as CRs closers to nominal levels. However, even though LLM, MI-LLM and MI-SUR performed equally well for effects, MI-LLM and MI-SUR were found to perform better than LLM for costs at 10 and 25% missing data. At 50% missing data, all strategies resulted in relatively high EBs and RMSEs for costs. Conclusion LLM should be combined with MI when analyzing trial-based economic evaluation data. MI-SUR is more efficient and can also be used, but then an average intervention effect over time cannot be estimated.

Keywords: Cost–benefit analysis; Longitudinal studies; Epidemiologic methods; Computer simulation (search for similar items in EconPapers)
JEL-codes: C15 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10198-022-01525-y Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:eujhec:v:24:y:2023:i:6:d:10.1007_s10198-022-01525-y

Ordering information: This journal article can be ordered from
http://www.springer. ... cs/journal/10198/PS2

DOI: 10.1007/s10198-022-01525-y

Access Statistics for this article

The European Journal of Health Economics is currently edited by J.-M.G.v.d. Schulenburg

More articles in The European Journal of Health Economics from Springer, Deutsche Gesellschaft für Gesundheitsökonomie (DGGÖ) Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().