Multiple imputation to deal with missing EQ-5D-3L data: Should we impute individual domains or the actual index?
- 1.1k Downloads
Missing data are a well-known and widely documented problem in cost-effectiveness analyses alongside clinical trials using individual patient-level data. Current methodological research recommends multiple imputation (MI) to deal with missing health outcome data, but there is little guidance on whether MI for multi-attribute questionnaires, such as the EQ-5D-3L, should be carried out at domain or at summary score level. In this paper, we evaluated the impact of imputing individual domains versus imputing index values to deal with missing EQ-5D-3L data using a simulation study and developed recommendations for future practice.
We simulated missing data in a patient-level dataset with complete EQ-5D-3L data at one point in time from a large multinational clinical trial (n = 1,814). Different proportions of missing data were generated using a missing at random (MAR) mechanism and three different scenarios were studied. The performance of using each method was evaluated using root mean squared error and mean absolute error of the actual versus predicted EQ-5D-3L indices.
In large sample sizes (n > 500) and a missing data pattern that follows mainly unit non-response, imputing domains or the index produced similar results. However, domain imputation became more accurate than index imputation with pattern of missingness following an item non-response. For smaller sample sizes (n < 100), index imputation was more accurate. When MI models were misspecified, both domain and index imputations were inaccurate for any proportion of missing data.
The decision between imputing the domains or the EQ-5D-3L index scores depends on the observed missing data pattern and the sample size available for analysis. Analysts conducting this type of exercises should also evaluate the sensitivity of the analysis to the MAR assumption and whether the imputation model is correctly specified.
KeywordsEQ-5D-3L Missing data Multiple imputation Missing data pattern Quality of life
We are indebted to the ISAT Collaborative Group for providing the data for this methodological work. ISAT was supported by grants from: The Medical Research Council, UK; Programme Hospitalier de Recherche Clinique 1998 of the French Ministry of Health (AOM 98150) sponsored by Assistance Publique-Hôpitaux de Paris (AP-HP); the Canadian Institutes of Health Research; and the Stroke Association of the UK. An early version of this paper was presented in the 83rd Health Economists’ Study Group (HESG) at the University of Warwick and we are grateful to Lazaros Andronis for discussing the manuscript and providing feedback and useful suggestions. This report is independent research arising from a NIHR Research Methods Fellowship, Claire Simons MET-12-15, supported by the National Institute for Health Research. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.
Conflict of interest
Oliver Rivero-Arias discloses that he is a member of the EuroQol Research Foundation.
All human studies have been approved by the appropriate ethics committee and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All persons gave informed consent prior to their inclusion in the ISAT study.
The work reported in this article was not funded by a specific grant.
- 7.Faria, R., Gomes, M., Epstein, D., & White, I. R. (2014). A guide to handling missing data in cost-effectiveness analysis conducted within randomised controlled trials. PharmacoEconomics. doi: 10.1007/s40273-014-0193-3.
- 8.Little, R. J. & D. B. Rubin. (2002). Statistical analysis with missing data. 2nd ed. Wiley Series in Probability and Statistics. Hoboken, NJ: Wiley.Google Scholar
- 15.National Institute for Health and Care Excellence. (2013). Guide to the methods of technology appraisal. London: National Institute for Health and Care Excellence.Google Scholar
- 18.Szende, A., M. Oppe, & N. Devlin.(2007). EQ-5D value sets: Inventory, comparative review and user guide. A. Szende, M. Oppe, and N. Devlin. (Eds.) Dordrecht: Springer.Google Scholar
- 19.StataCorp. Stata Statistical Software. (2011). Stata Press: College Station. TX: StataCorp LP.Google Scholar
- 20.Molyneux, A., Kerr, R., Stratton, I., Sandercock, P., Clarke, M., Shrimpton, J., et al. (2002). International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: A randomised trial. Lancet, 360(9342), 1267–1274.CrossRefPubMedGoogle Scholar
- 23.EuroQol Research Foundation (2014). Available from http://www.euroqol.org/. [Accessed 14 September 2014].
- 25.Fairbank, J., Frost, H., Wilson-MacDonald, J., Yu, L. M., Barker, K., & Collins, R. (2005). Spine stabilisation trial. Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: The MRC spine stabilisation trial. BMJ, 330(7502), 1233.CrossRefPubMedCentralPubMedGoogle Scholar
- 28.Kendrick, T., Simons, L., Mynors-Wallis, L., Gray, A., Lathlean, J., Pickering, R., et al. (2006). Cost-effectiveness of referral for generic care or problem-solving treatment from community mental health nurses, compared with usual general practitioner care for common mental disorders: Randomised controlled trial. British Journal of Psychiatry, 189, 50–59.CrossRefPubMedGoogle Scholar
- 32.Benjamini, Y., & Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B (Methodological), 57(1), 289–300.Google Scholar
- 33.Efron, B. (1979). The 1977 rietz lecture—bootstrap methods—another look at the Jackknife. The annals of Statistics, 7(1), 1–26.Google Scholar
- 34.Kind, P., Hardman, G., & Macran, S. (1999). UK population norms for EQ-5D. UK: Centre for Health Economics, University of York.Google Scholar
- 36.Konig, H. H., Born, A., Gunther, O., Matschinger, H., Heinrich, S., Riedel-Heller, S. G., et al. (2010). Validity and responsiveness of the EQ-5D in assessing and valuing health status in patients with anxiety disorders. Health and Quality of Life Outcomes, 8, 47.CrossRefPubMedCentralPubMedGoogle Scholar
- 37.Long, J. S. (1997). Regression models for categorical and limited dependent variables. London: Sage.Google Scholar
- 38.Ramsey, J. B. (1969). Tests for specification errors in classical linear least-squares regression analysis. Journal of the Royal Statistical Society. Series B-Statistical Methodology, 31(2), 350–371.Google Scholar