Multilevel Modeling



Mortality is the most frequently modeled outcome in injury research. It is easy to recognize, ­relatively free from measurement error, and fundamentally interesting. Injury researchers in public health or clinical medicine have become familiar with logistic regression as a standard way to model a binary outcome like mortality (or alternatively survival). Many other outcomes encountered in injury research can also be considered binary, such as the occurrence of a serious complication or an extended length of stay in hospital.


Trauma Center Multilevel Model Standardize Mortality Ratio Multilevel Logistic Regression Logit Scale 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



Supported in part by Grant R21HD061318 (PI Clark) from the National Institutes of Health, Grants R01HS015656 (PI Clark) and R01HS017718 (PI Shafi) from the Agency for Healthcare Research and Quality, a grant from the Maine Medical Center Research Strategic Plan (PI Clark), and a research award from the Canadian Institutes of Health Research (PI Moore).


  1. Adewale, A. J., Hayduk, L., Estabrooks, C. A., et al. (2007). Understanding hierarchical linear models: Applications in nursing research. Nursing Research, 56, S40–6.PubMedCrossRefGoogle Scholar
  2. Antretter, E., Dunkel, D., Osvath, P., et al. (2006). Multilevel modeling was a convenient alternative to common regression designs in longitudinal suicide research. Journal of Clinical Epidemiology, 59, 576–86.PubMedCrossRefGoogle Scholar
  3. Austin, P. C., Tu, J. V., & Alter, D. A. (2003). Comparing hierarchical modeling with traditional logistic regression analysis among patients hospitalized with acute myocardial infarction: Should we be analyzing cardiovascular outcomes data differently? American Heart Journal, 145, 27–35.PubMedCrossRefGoogle Scholar
  4. Borrell, C., Rodriguez, M., Ferrando, J., et al. (2002). Role of individual and contextual effects in injury mortality: New evidence from small area analysis. Injury Prevention, 8, 297–302.PubMedCrossRefGoogle Scholar
  5. Browne, W. J., & Draper, D. (2006). A comparison of Bayesian and likelihood-based methods for fitting multilevel models. Bayesian Analysis, 1, 473–514.CrossRefGoogle Scholar
  6. Centre for multilevel modelling. Accessed 1 Mar 2011.
  7. Champion, H. R., Copes, W. S., Sacco, W. J., et al. (1990). The Major Trauma Outcome Study: Establishing national norms for trauma care. Journal of Trauma, 30, 1356–65.PubMedCrossRefGoogle Scholar
  8. Chan, C. K., Feinstein, A. R., Jekel, J. F., & Wells, C. K. (1988). The value and hazards of standardization in clinical epidemiologic research. Journal of Clinical Epidemiology, 41, 1125–34.PubMedCrossRefGoogle Scholar
  9. Clark, D. E., Hannan, E. L., & Wu, C. (2010a). Predicting risk-adjusted mortality for trauma patients: Logistic versus multilevel logistic models. Journal of the American College of Surgeons, 211, 224–31.PubMedCentralPubMedCrossRefGoogle Scholar
  10. Clark, D. E., Hannan, E. L., & Raudenbush, S. W. (2010b). Using a hierarchical model to estimate risk-adjusted ­mortality for hospitals not included in the reference sample. Health Services Research, 45, 577–87.PubMedCrossRefGoogle Scholar
  11. Clark, D. E., Lucas, F. L., & Ryan, L. M. (2007). Predicting hospital mortality, length of stay, and transfer to long-term care for injured patients. Journal of Trauma, 62, 592–600.PubMedCrossRefGoogle Scholar
  12. Cohen, M. E., Dimick, J. B., Bilimoria, K. Y., et al. (2009). Risk adjustment in the American College of Surgeons National Surgical Quality Improvement Program: A comparison of logistic versus hierarchical modeling. Journal of the American College of Surgeons, 209, 687–93.PubMedCrossRefGoogle Scholar
  13. Cook, T. D., & DeMets, D. L. (2008). Introduction to statistical methods for clinical trials. Boca Raton, FL: Chapman and Hall.Google Scholar
  14. DeLong, E. R., Peterson, E. D., DeLong, D. M., et al. (1997). Comparing risk-adjustment methods for provider ­profiling. Statistics in Medicine, 16, 2645–64.PubMedCrossRefGoogle Scholar
  15. Diez-Roux, A. V. (2000). Multilevel analysis in public health research. Annual Review of Public Health, 21, 171–92.PubMedCrossRefGoogle Scholar
  16. Dimick, J. B., & Welch, H. G. (2008). The zero mortality paradox in surgery. Journal of the American College of Surgeons, 206, 13–6.PubMedCrossRefGoogle Scholar
  17. Flora, J. D., Jr. (1978). A method for comparing survival of burn patients to a standard survival curve. Journal of Trauma, 18, 701–5.PubMedCrossRefGoogle Scholar
  18. Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge: Cambridge University Press.Google Scholar
  19. Glance, L. G., Osler, T. M., Dick, A. W., et al. (2010). The survival measurement and reporting trial for trauma (SMARTT): Background and study design. Journal of Trauma, 68, 1491–7.PubMedCrossRefGoogle Scholar
  20. Goldstein, H. (2003). Multilevel statistical models. London: Hodder Arnold.Google Scholar
  21. Goldstein, H., Browne, W., & Rasbash, J. (2002). Multilevel modelling of medical data. Statistics in Medicine, 21, 3291–315.PubMedCrossRefGoogle Scholar
  22. Gorra, A. S., Clark, D. E., Mullins, R. J., & Delorenzo, M. A. (2008). Regional variation in hospital mortality and 30-day mortality for injured Medicare patients. World Journal of Surgery, 32, 954–9.PubMedCrossRefGoogle Scholar
  23. Hannan, E. L., Wu, C., DeLong, E. R., & Raudenbush, S. W. (2005). Predicting risk-adjusted mortality for CABG surgery: Logistic versus hierarchical logistic models. Med Care, 43, 726–35.PubMedCrossRefGoogle Scholar
  24. Hemmila, M. R., Nathens, A. B., Shafi, S., et al. (2010). The Trauma Quality Improvement Program: Pilot study and initial demonstration of feasibility. Journal of Trauma, 68, 253–62.PubMedCrossRefGoogle Scholar
  25. Jones, A. P., & Jorgensen, S. H. (2003). The use of multilevel models for the prediction of road accident outcomes. Accident Analysis and Prevention, 35, 59–69.PubMedCrossRefGoogle Scholar
  26. Jones, J. M., Redmond, A. D., & Templeton, J. (1995). Uses and abuses of statistical models for evaluating trauma care. Journal of Trauma, 38, 89–93.PubMedCrossRefGoogle Scholar
  27. Kim, D. G., Lee, Y., Washington, S., & Choi, K. (2007). Modeling crash outcome probabilities at rural intersections: Application of hierarchical binomial logistic models. Accident Analysis and Prevention, 39, 125–34.PubMedCrossRefGoogle Scholar
  28. Kirkham, J. J., & Bouamra, O. (2008). The use of statistical process control for monitoring institutional performance in trauma care. Journal of Trauma, 65, 1494–501.PubMedCrossRefGoogle Scholar
  29. Krumholz, H. M., Brindis, R. G., Brush, J. E., et al. (2006). Standards for statistical models used for public reporting of health outcomes. Circulation, 113, 456–62.PubMedCrossRefGoogle Scholar
  30. Lenguerrand, E., Martin, J. L., & Laumon, B. (2006). Modelling the hierarchical structure of road crash data – ­application to severity analysis. Accident Analysis and Prevention, 38, 43–53.PubMedCrossRefGoogle Scholar
  31. Lipsky, A. M., Gausche-Hill, M., Vienna, M., & Lewis, R. J. (2010). The importance of “shrinkage” in subgroup analyses. Annals of Emergency Medicine, 55(544–52), e3.PubMedGoogle Scholar
  32. MacNab, Y. C. (2003). A Bayesian hierarchical model for accident and injury surveillance. Accident Analysis and Prevention, 35, 91–102.PubMedCrossRefGoogle Scholar
  33. Milton, J. C., Shankar, V. N., & Mannering, F. L. (2008). Highway accident severities and the mixed logit model: An exploratory empirical analysis. Accident Analysis and Prevention, 40, 260–6.PubMedCrossRefGoogle Scholar
  34. Moore, L., Hanley, J. A., Turgeon, A. F., et al. (2009a). A multiple imputation model for imputing missing physiologic data in the National Trauma Data Bank. Journal of the American College of Surgeons, 209, 572–9.PubMedCrossRefGoogle Scholar
  35. Moore, L., Lavoie, A., Turgeon, A. F., et al. (2009b). The trauma risk adjustment model: A new model for evaluating trauma care. Annals of Surgery, 249, 1040–6.PubMedCrossRefGoogle Scholar
  36. Moore, L., Hanley, J. A., Turgeon, A. F., & Lavoie, A. (2010a). Evaluating the performance of trauma centers: Hierarchical modeling should be used. Journal of Trauma, 69, 1132–7.PubMedCrossRefGoogle Scholar
  37. Moore, L., Hanley, J. A., Turgeon, A. F., et al. (2010b). A new method for evaluating trauma centre outcome performance: TRAM-adjusted mortality estimates. Annals of Surgery, 251, 952–8.PubMedCrossRefGoogle Scholar
  38. Mukamel, D. B., Glance, L. G., Dick, A. W., & Osler, T. M. (2010). Measuring quality for public reporting of health provider quality: Making it meaningful to patients. American Journal of Public Health, 100, 264–9.PubMedCrossRefGoogle Scholar
  39. Normand, S.-L. T., Glickman, M. E., & Gatsonis, C. A. (1997). Statistical methods for profiling providers of medical care: Issues and applications. Journal of the American Statistical Association, 92, 803–14.CrossRefGoogle Scholar
  40. Normand, S.-L. T., & Shahian, D. M. (2007). Statistical and clinical aspects of hospital outcomes profiling. Statistical Science, 22, 206–26.CrossRefGoogle Scholar
  41. O’Connell, A. A., & McCoach, D. B. (2004). Applications of hierarchical linear models for evaluations of health interventions: Demystifying the methods and interpretations of multilevel models. Evaluation and The Health Professions, 27, 119–51.PubMedCrossRefGoogle Scholar
  42. O’Hagan, A. (1994). Kendall’s advanced theory of statistics, volume 2B: Bayesian inference. New York: Halsted Press.Google Scholar
  43. Pollock, D. A. (1999). Summary of the discussion: Trauma registry data and TRISS evidence. Journal of Trauma, 47, S56–S8.PubMedCrossRefGoogle Scholar
  44. Rabe-Hesketh, S., & Skrondal, A. (2008). Multilevel and longitudinal modeling using Stata. College Station, TX: Stata Press.Google Scholar
  45. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models. London: Sage.Google Scholar
  46. Rothman, K. J. (1986). Modern epidemiology. Boston: Little, Brown and Co.Google Scholar
  47. Shahian, D. M., Normand, S.-L. T., Torchiana, D. F., et al. (2001). Cardiac surgery report cards: Comprehensive review and statistical critique. The Annals of Thoracic Surgery, 72, 2155–68.PubMedCrossRefGoogle Scholar
  48. Snijders, T. A. B., & Bosker, R. J. (1999). Multilevel analysis. London: Sage.Google Scholar
  49. Spiegelhalter, D. J. (2005). Funnel plots for comparing institutional performance. Statistics in Medicine, 24, 1185–202.PubMedCrossRefGoogle Scholar
  50. The BUGS project. Accessed 1 Mar 2011.
  51. Ulm, K. (1990). A simple method to calculate the confidence interval of a standardized mortality ratio (SMR). American Journal of Epidemiology, 131, 373–5.PubMedGoogle Scholar
  52. Yannis, G., Papadimitriou, E., & Antoniou, C. (2007). Multilevel modelling for the regional effect of enforcement on road accidents. Accident Analysis and Prevention, 39, 818–25.PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2012

Authors and Affiliations

  1. 1.Maine Medical CenterPortlandUSA
  2. 2.Harvard Injury Control Research CenterHarvard School of Public HealthBostonUSA
  3. 3.Département de Médecine Sociale et PréventiveUniversité LavalQuébec CityCanada
  4. 4.Centre Hospitalier Affilié Universitaire de QuébecQuebec CityCanada

Personalised recommendations