An introduction to multilevel regression models

Data in health research are frequently structured hierarchically. For examp le, data may consist of patients nested within physicians, who in turn may be nested in hospitals or geographic regions. Fitting regression models tha t ignore the hierarchical structure of the data can lead to false inference s being drawn from the data. Implementing a statistical analysis that takes into account the hierarchical structure of the data requires special metho dologies. In this paper, we introduce the concept of hierarchically structured data, and present an introduction to hierarchical regression models. We then comp are the performance of a traditional regression model with that of a hierar chical regression model on a dataset relating test utilization at the annua l 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.