Our patient-level variables are drawn from a database of every ambulance ride in Pennsylvania which could be linked to a hospital discharge during 1995 (approximately 170,000 observations). This dataset is gathered by the Pennsylvania Department of Health and has only recently been made available to a limited number of researchers; we are not aware of prior work on this database (or a similar ambulance-level database) by health care economists. how to get out of payday loans

The information provided in this patient-level data is unusually rich. First, there are several indicators associated with the responsiveness of the 911 system. We analyze three different measures of the timeliness of ambulance response: the amount of time it takes to get to the scene of an emergency (TIME_TO_SCENE), the amount of time spent at the scene (TIME_AT_SCENE), and the amount of time elapsed from when the ambulance leaves the scene to the time when the ambulance arrives at the hospital (TIME_TO__HOSP).

In the next sections, we will examine how the response time measures vary with other features of the medical care system. To better motivate that type of analysis, we restrict our analysis of the Pennsylvania data to the case of cardiac incidents. One of the main advantages of analyzing the case of cardiac incidents is that, in contrast to many datasets, there are in fact a number of quite precise indicators of the level of severity of each patient.