What is the Negative Binomial Model?

Many research questions require healthcare economists to measure the effect of various patient, physician or market-level characteristics on specific health events.  Oftentimes, these events are discrete in nature.  For instance, doctor’s visits, ER visits, and hospitalizations are all discrete events. To properly estimate the effect of certain characteristics on a discrete event, count models are needed.  The…

Risk Adjustment: Overfitting the Model

When creating risk adjustment models to predict health care spending, many researchers aim to maximize the goodness-of-fit of the model.  Maximize the goodness of fit, however, can produce the problem of overfitting. Overfitting occurs when a model describes random error or noise instead of the underlying relationship. Overfitting generally occurs when a model is excessively…

How to Measure Physician Efficiency

Many payers are moving towards value-based purchasing programs that reward efficient physicians with additional payments and punish inefficient physicians with decreased payments.  Medicare’s Quality and Resource Use Reports (QRUR) are a step in this direction. However, summarizing overall physician quality is a difficult prospect.  First, the types of cases each physician treats is not homogeneous,…

Attrition Bias

If you are evaluating the treatment effect of a policy or medical intervention, does it matter if some of your subjects leave the sample? In many cases, the answer is ‘yes’. The Problem As outlined in Grasdal (2001), the effect of the treatment is simply: Δ = E(Y|X, T=1) − E(Y|X, T=0) However, in some…