An introduction to multilevel regression models.

P C Austin, V Goel, C van Walraven
Author Information
  1. P C Austin: Institute for Clinical Evaluative Sciences, G-160, 2075 Bayview Avenue, North York, ON, M4N 3M5. peter.austin@ices.on.ca

Abstract

Data in health research are frequently structured hierarchically. For example, data may consist of patients nested within physicians, who in turn may be nested in hospitals or geographic regions. Fitting regression models that ignore the hierarchical structure of the data can lead to false inferences being drawn from the data. Implementing a statistical analysis that takes into account the hierarchical structure of the data requires special methodologies. In this paper, we introduce the concept of hierarchically structured data, and present an introduction to hierarchical regression models. We then compare the performance of a traditional regression model with that of a hierarchical regression model on a dataset relating test utilization at the annual health exam with patient and physician characteristics. In comparing the resultant models, we see that false inferences can be drawn by ignoring the structure of the data.

References

  1. Soc Sci Med. 1998 Jan;46(1):97-117 [PMID: 9464672]
  2. J Health Serv Res Policy. 1996 Jul;1(3):154-64 [PMID: 10180862]
  3. J Eval Clin Pract. 2000 May;6(2):215-24 [PMID: 10970015]
  4. Ann Intern Med. 1997 Oct 15;127(8 Pt 2):764-8 [PMID: 9382395]
  5. Lancet. 1998 Feb 21;351(9102):555-8 [PMID: 9492774]
  6. Stat Med. 1998 Jan 15;17(1):41-57 [PMID: 9463848]
  7. Med Care. 1995 Jun;33(6):625-42 [PMID: 7760578]

MeSH Term

Bias
Clinical Laboratory Techniques
Data Interpretation, Statistical
Female
Health Services Research
Humans
Linear Models
Male
Normal Distribution
Ontario
Poisson Distribution
Predictive Value of Tests
Regression Analysis

Word Cloud

Created with Highcharts 10.0.0dataregressionmodelshierarchicalstructurehealthstructuredhierarchicallymaynestedcanfalseinferencesdrawnintroductionmodelDataresearchfrequentlyexampleconsistpatientswithinphysiciansturnhospitalsgeographicregionsFittingignoreleadImplementingstatisticalanalysistakesaccountrequiresspecialmethodologiespaperintroduceconceptpresentcompareperformancetraditionaldatasetrelatingtestutilizationannualexampatientphysiciancharacteristicscomparingresultantseeignoringmultilevel

Similar Articles

Cited By