Advertisement

European Journal of Epidemiology

, Volume 21, Issue 1, pp 15–24 | Cite as

Analysis of the Benefits of a Mediterranean Diet in the GISSI-Prevenzione Study: A Case Study in Imputation of Missing Values from Repeated Measurements

  • Federica BarziEmail author
  • Mark Woodward
  • Rosa Maria Marfisi
  • Gianni Tognoni
  • Roberto Marchioli
  • on behalf of GISSI-Prevenzione Investigators
Cardiovascular Diseases

Abstract

The problem of missing values has increasingly being recognized in epidemiology. New methods allow for the analysis of missing data that can provide valid estimates of epidemiological quantities of interest. The GISSI-Prevenzione study was aimed to reliably assess the long-term relationship between the consumption of foods typical of the Mediterranean diet and the risk of mortality amongst 11,323 Italians with prior myocardial infarction. Food intake frequencies were recorded repeatedly over the 4.5 years of follow-up and missing values affected each food variable at increasing rates over the course of the study. Comparisons were made between the results obtained from the analysis of the complete data and those obtained after imputing the missing data with simple imputation methods and with various implementations of the multiple imputation (MI) method. MI appeared to best address the issue of missing data on the food intake frequencies, preserving the observed distributions and relationships between variables whilst producing plausible estimates of variability. Given its theoretical properties and flexibility to different types of data, MI is more likely to provide valid estimates, compared to complete data analysis and imputation by simple methods, and is thus worthy of wider consideration amongst epidemiological researchers.

Keywords

Imputation Longitudinal study Mediterranean diet Missing values Mortality Odds ratio 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Schafer, JL 1997Analysis of Incomplete Multivariate DataChapman & HallLondonGoogle Scholar
  2. 2.
    Landrum, MB, Becker, MP 2001A multiple imputation strategy for incomplete longitudinal dataStat Med2027412760CrossRefPubMedGoogle Scholar
  3. 3.
    Twisk, J, Vente, W 2002Attrition in longitudinal studies: How to dealt with missing dataJ Clin Epidemiol55329337CrossRefPubMedGoogle Scholar
  4. 4.
    Rubin, DB 1987Multiple Imputations for Nonresponse in SurveysWiley & SonsNew YorkGoogle Scholar
  5. 5.
    Allison, PD. 2001Missing Data SeriesSage Publications Inc.Thousand Oaks, CA(Series: quantitative applications in the social sciences. Vol 136. A Sage University paper)Google Scholar
  6. 6.
    Kmetic, A, Ioseph, L, Berger, C, Tenenhouse, A 2002Multiple imputation to account for missing data in survey: Estimating the prevalence of osteoporosisEpidemiology13437444CrossRefPubMedGoogle Scholar
  7. 7.
    Wu, H, Wu, L 2001A multiple imputation method for missing covariates in non-linear mixed-effect models with application to HIV dynamicsStat Med2017551769PubMedGoogle Scholar
  8. 8.
    Greenland, S, Finkle, WD 1995A critical look at methods for handling missing covariates in epidemiologic regression analysisAm J Epidemiol14212551264PubMedGoogle Scholar
  9. 9.
    Taylor, JMG, Cooper, KL, Wei, JT, Sarma, AV, Raghunathan, TE, Heeriga, SG 2002Use of multiple imputation to correct for non-response bias in a survey of urologic symptoms among African-American menAm J Epidemiol156774782CrossRefPubMedGoogle Scholar
  10. 10.
    Arnold, AM, Kronmal, RA 2003Multiple imputation of baseline data in the cardiovascular health studyAm J Epidemiol1577484CrossRefPubMedGoogle Scholar
  11. 11.
    Perez, A, Dennis, RJ, Gil, JFA, Rondon, MA, Lopez, A 2002Use of the mean, hot deck and multiple imputation techniques to predict outcome in intensive care unit patients in ColombiaStat Med2138853896PubMedGoogle Scholar
  12. 12.
    Barnard, J, Meng, XL 1999Application of multiple imputation in medical studies: From AIDS to NHANESStat Methods Med Res81736CrossRefPubMedGoogle Scholar
  13. 13.
    Molenberghs, G, Williams, PL, Lipsitz, S 2002Prediction of survival and opportunistic infections in HIV-infected patients: A comparison of imputation methods of incomplete CD4 countsStat Med2113871408CrossRefPubMedGoogle Scholar
  14. 14.
    Barzi, F, Woodward, M 2004Imputation of missing values in practice: Results from imputation of serum cholesterol in 28 cohort studiesAm J Epidemiol1603445CrossRefPubMedGoogle Scholar
  15. 15.
    Rubin, DB 1996Multiple imputation after 18+ yearsJ Am Stat Assoc91473489Google Scholar
  16. 16.
    Barzi, F, Woodward, M, Marfisi, RM, Tavazzi, L, Valagussa, F, Marchioli, R 2003Mediterranean diet and all-causes mortality after myocardial infarction: Results from the GISSI-Prevenzione trialEur J Clin Nutr57604611CrossRefPubMedGoogle Scholar
  17. 17.
    Little, RJA, Rubin, DB 1987Statistical Analysis with Missing DataJohn Wiley and SonsNew YorkGoogle Scholar
  18. 18.
    Rubin, DB 1976Inference and missing dataBiometrika63581592Google Scholar
  19. 19.
    Marchioli, R, Avanzini, F, Barzi, F,  et al. 2001Assessment of absolute risk of death after myocardial infarction by use of multiple-risk-factor assessment equations: GISSI-Prevenzione mortality risk chartEur Heart J2220472049CrossRefGoogle Scholar
  20. 20.
    SAS. SAS/STAT Software (2001) Changes and Enhancements Release 8.2. Cary, NC: SAS Institute Inc., www.sas.com/newversionGoogle Scholar
  21. 21.
    Schafer JL and Yucel R. (1999) PAN: Multiple imputation for multivariate panel data, software for windows 95/98/NT, http://www.stat.psu.edu/~jls/ misoftwa.html
  22. 22.
    Schafer JL. (1999) Software for Multiple Imputation. University Park, PA: The Pennsylvania State University Department of Statistics, http://www.stat.psu.edu/~jls/misotwa.html
  23. 23.
    Yuan YC. (2000) Multiple Imputation for Missing Data: Concepts and New Development. Rockville, MD: SAS Institute Inc. (Paper 267–25), http://support.sas.com/rnd/app/papers/multipleimputation.pdf
  24. 24.
    Buuren, S, Boshuizen, HC, Knook, DL 1999Multiple imputation of missing blood pressure covariates in survival analysisStat Med18681694PubMedGoogle Scholar
  25. 25.
    Zhou, X-H, Ecket, GJ, Tierney, WM 2001Multiple imputation in public health researchStat Med2015411549PubMedGoogle Scholar
  26. 26.
    Gmel, G 2001Imputation of missing values in the cases of multiple item instrument measuring alcohol consumptionStat Med2023692381CrossRefPubMedGoogle Scholar
  27. 27.
    Van Buuren S. (2003) TNO Prevention and Health, Department of Statistics. Multiple Imputations Online, Leiden, http://www.multiple-imputation.com
  28. 28.
    Longford, NT 2000Handling missing data in diaries of alcohol consumptionJ R Stat Soc (A)163381402Google Scholar
  29. 29.
    Little,  1993Pattern-mixture models for multivariate incomplete dataJ Am Stat Assoc88125134Google Scholar
  30. 30.
    Horton, NJ 2001Multiple imputation in practice: Comparison of software packages for regression models with missing variablesAm Stat55244254CrossRefGoogle Scholar

Copyright information

© Springer 2006

Authors and Affiliations

  • Federica Barzi
    • 1
    Email author
  • Mark Woodward
    • 1
  • Rosa Maria Marfisi
    • 2
  • Gianni Tognoni
    • 2
  • Roberto Marchioli
    • 2
  • on behalf of GISSI-Prevenzione Investigators
  1. 1.The George Institute for International HealthUniversity of SydneySydneyAustralia
  2. 2.Laboratory of Clinical Epidemiology of Cardiovascular Disease, Department of Clinical Pharmacology and EpidemiologyConsorzio Mario Negri SudSanta Maria ImbaroItaly

Personalised recommendations