The York Research Database

We assess the impact of the English version of prospective payment, termed Payment by Results (PbR), on hospital quality, as measured by in-hospital mortality and 28-day emergency readmission. To do this, we exploit the phased introduction of PbR across hospitals and across three treatments (hip replacement, hernia repair and stroke) which were exposed to PbR at different times. We estimate regression models to analyse factors associated with patient survival and readmission for all those admitted for hip replacement (n=499,555), hernia repair (n=414,959) or following stroke (n=487,040) between 2002/3 and 2007/8. Factors include patient and hospital characteristics and the proportion of hospital income derived from PbR. We find that the probability of survival improved over time while changes in crude readmission rates varied by condition. Patient characteristics are important at explaining survival and readmission, and hospital size and specialisation also appear significant, though not consistently so across conditions or time. The probability of surviving stroke is lower for those admitted over the weekend. Given the high mortality rate for stroke, it is critical to account for the probability of surviving the initial admission when evaluating readmissions. PbR does not appear to have influenced the probability of survival or readmission.