In particular, each patient is assigned a Glasgow Score which is a number between 0-15 which indicates the severity of the heart attack (lower numbers imply higher severity with 3 being the worst and 0 indicating “unknown” or “missing”). While the bulk of observations are coded with the weakest severity (GLASGOW = 15), there exists a substantial minority for which there is variation in the data.

We are also able to observe whether the incident is believed to be a cardiac arrest or simply a cardiac incident (CARD_ARR = 1 or 0).

In addition to these measures of severity, the data includes relatively detailed information about each individual in the dataset, including their insurance status, age, and sex. As well, we observe some information about the types of procedures administered by the emergency response paramedics, including whether the patient receives defibrillation treatment prior to arrival at the hospital. However, since the decision to defibrillate a patient is conditioned on patient characteristics which are unobserved to the econometrician, this variable serves mainly as a control in our analysis. payday one loans

Finally, we are able to observe some concrete measures associated with patient outcomes. In our main analysis, we will focus on the most extreme of these measures, DEAD: whether or not the patient dies from the incident, either in the emergency room or in the hospital afterwards.