Health Care Management Science

, Volume 10, Issue 3, pp 231–238 | Cite as

A Bayesian approach to assess heart disease mortality among persons with diabetes in the presence of missing data

  • Betsy L. Cadwell
  • James P. Boyle
  • Edward F. Tierney
  • Theodore J. Thompson


Some states’ death certificate form includes a diabetes yes/no check box that enables policy makers to investigate the change in heart disease mortality rates by diabetes status. Because the check boxes are sometimes unmarked, a method accounting for missing data is needed when estimating heart disease mortality rates by diabetes status. Using North Dakota’s data (1992–2003), we generate the posterior distribution of diabetes status to estimate diabetes status among those with heart disease and an unmarked check box using Monte Carlo methods. Combining this estimate with the number of death certificates with known diabetes status provides a numerator for heart disease mortality rates. Denominators for rates were estimated from the North Dakota Behavioral Risk Factor Surveillance System. Accounting for missing data, age-adjusted heart disease mortality rates (per 1,000) among women with diabetes were 8.6 during 1992–1998 and 6.7 during 1999–2003. Among men with diabetes, rates were 13.0 during 1992–1998 and 10.0 during 1999–2003. The Bayesian approach accounted for the uncertainty due to missing diabetes status as well as the uncertainty in estimating the populations with diabetes.


Missing data Bayesian methods Random walk Metropolis-Hastings Diabetes mortality Death certificate data Diabetes check box 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Anderson RN, Smith BL (2005) Deaths: Leading causes for 2002. National Vital Statistics Report 53 (17) 1–89. (Available at http://www/cdc/gov/nchs/data/nvsr/nvsr53/nvsr53_17.pdf, accessed 9 June 2006.)
  2. 2.
    Narayan KM, Boyle JP, Thompson TJ, Sorensen SE, Williamson DF (2003) Lifetime risk for diabetes mellitus in the United States. J Am Med Assoc 290:1884–1890CrossRefGoogle Scholar
  3. 3.
    Geiss LS, Herman WH, Smith PJ (1995) Mortality in non-insulin-dependent diabetes. In: Harris MI, Cowie CC, Stern MP, Boyko EJ, Reiber GE, Bennet PH (eds) Diabetes in America, 2nd ed. US Government Printing Office, Washington, DCGoogle Scholar
  4. 4.
    Engelgau MM, Geiss LS, Saaddine JB, Boyle JP, Benjamin SM, Gregg EW, Tierney EF, Rios-Burrows N, Mokdad AH, Ford ES, Imperatore G, Narayan KM (2004) The evolving diabetes burden in the United States. Ann Intern Med 140:945–950Google Scholar
  5. 5.
    Honeycutt AA, Boyle JP, Broglio KR, Thompson TJ, Hoerger TJ, Geiss LS, Narayan KM (2003) A dynamic Markov model for forecasting diabetes prevalence in the United States through 2050. Health Care Manage Sci 6:155–164CrossRefGoogle Scholar
  6. 6.
    Tierney EF, Geiss LS, Engelgau MM, Thompson TJ, Schauber D, Shireley LA, Bukelic PJ, McDonough SL (2001) Population-based estimates of mortality associated with diabetes: use of a death certificate check box in North Dakota. Am J Public Health 91:84–92Google Scholar
  7. 7.
    Tierney EF, Cadwell BL, Engelgau MM, Shireley L, Parsons SL, Moum K, Geiss LS (2004) Declining mortality rate among persons with diabetes in North Dakota, 1997–2002. Diabetes Care 27:2723–2725CrossRefGoogle Scholar
  8. 8.
    Zhou X, Eckert GJ, Tierney WM (2001) Multiple imputation in public health research. Stat Med 20:1541–1549CrossRefGoogle Scholar
  9. 9.
    Schafer JL, Graham JW (2002) Missing data: Our view of the state of the art. Psychol Methods 7(2):147–177CrossRefGoogle Scholar
  10. 10.
    Abraham WT, Russell DW (2004) Missing data: a review of current methods and applications in epidemiological research. Curr Opin Psychiatry 17:315–321CrossRefGoogle Scholar
  11. 11.
    Heitjan DF (1991) Multiple imputation for the fatal accident reporting system. Appl Stat 40(1):13–29CrossRefGoogle Scholar
  12. 12.
    Clogg CC, Rubin DB, Schenker N, Schultz B, Weidman L (1991) Multiple imputation of industry and occupation codes in census public-use samples using Bayesian logistic regression. J Am Stat Assoc 86(413):68–78CrossRefGoogle Scholar
  13. 13.
    Gelman A, Carlin JB, Stern HS, Rubin DB (2004) Bayesian data analysis, 2nd ed. Chapman & Hall, Boca RatonGoogle Scholar
  14. 14.
    Spiegelhalter DJ, Best NG, Carlin BP, van der Linde A (2002) Bayesian measure of model complexity and fit (with discussion). J R Stat Soc B44:377–387Google Scholar
  15. 15.
    SUDAAN (2004) Language Manual. Release 9.0 Research Triangle Institute, Research Triangle ParkGoogle Scholar
  16. 16.
    Anderson RN, Rosenberg HM (1998) Age standardization of death rates: implementation of the year 2000 standard. National Center for Health Statistic, National Vital Statistics Reports 47(3)Google Scholar
  17. 17.
    Schafer JL (1997) Analysis of incomplete multivariate data. Chapman & Hall, Boca RatonGoogle Scholar
  18. 18.
    Spiegelhalter D, Thomas A, Best N, Lunn D (2003) WinBUGS user manual version 1.4. MRC Biostatistics Unit, CambridgeGoogle Scholar
  19. 19.
    Thomas RJ, Palumbo PJ, Melton LJ, Roger VL, Ransom J, O’Brien PC, Leibson CL (2003) Trends in the mortality burden associated with diabetes mellitus. Arch Intern Med 163:445–451CrossRefGoogle Scholar
  20. 20.
    Fox CS, Coady S, Sorlie PD, Levy D, Meigs JB, Agostino RBD, Wilson PW, Savage PJ (2004) Trends in cardiovascular complications of diabetes. J Am Med Assoc 292:2495–2499CrossRefGoogle Scholar
  21. 21.
    Gu K, Cowie CC, Harris MI (1999) Diabetes and decline in heart disease mortality in US adults. J Am Med Assoc 281:1291–1297CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  • Betsy L. Cadwell
    • 1
  • James P. Boyle
    • 1
  • Edward F. Tierney
    • 1
  • Theodore J. Thompson
    • 1
  1. 1.Division of Diabetes Translation, National Center for Chronic Disease Prevention and Health PromotionCenters for Disease Control and PreventionAtlantaUSA

Personalised recommendations