Island Jewish Medical Center in New Hyde Park, Maimonides Medical Center in Brooklyn and Mount Sinai Hospital in Manhattan) had RAMRs that were signicantly lower than the statewide rate. It should be noted that hospitals are more likely to have results that show a statistically significant difference from the statewide rate when three years of data are used than when one year of data is used because the three-year volumes are higher.
Table 2 also presents the 3-year in-hospital/30-day RAMRs for non-emergency cases based on the model in Appendix 4. Non-emergency cases comprise 87.90 percent of cases for the period 2006-2008. e statewide in-hospital/30-day mortality rate for the 144,853 non-emergency cases during the 3-year period was 0.63 percent. Observed mortality rates for this group of patients ranged from 0.00 percent to 1.25 percent and the RAMRs ranged from 0.00 to 1.72 percent. Six hospitals (Crouse Hospital in Syracuse, Monteore Medical Center – Weiler Division in the Bronx, St. Peter’s Hospital in Albany, St. Vincent’s Catholic Medical Center in Manhattan, Strong Memorial Hospital in Rochester and University Hospital of Brooklyn) had RAMRs that were signicantly higher than the statewide average. Four hospitals (Maimonides Medical Center in Brooklyn, Mount Sinai Hospital in Manhattan, NYU Hospitals Center in Manhattan and St. Francis Hospital in Roslyn) had RAMRs that were signicantly below the statewide rate for non-emergency cases.
e last three columns in Table 2 present data on emergency cases based on the model in Appendix 5. Emergency
cases represented 12.10 percent of cases for the period 2006-2008. e statewide in-hospital/30-day mortality rate for the 19,939 emergency PCI cases during the 3-year period was 3.09 percent. Observed mortality rates for this group ranged from 0.00 percent to 11.11 percent and the RAMRs ranged from 0.00 percent to 8.40 percent. ree hospitals (Brookdale Hospital medical Center in Brooklyn, Erie County medical Center in Bualo and Glens Falls Hospital in Glens Falls) had RAMRs that were signicantly above the statewide average. Two hospitals (Mount Sinai Hospital in Manhattan and North Shore University Hospital in Manhasset had RAMRs that were signicantly below the statewide average for emergency cases.
e observed in-hospital mortality rate for all 164,792 cases included in Table 2 was 0.54 percent. e in-hospital
mortality rate was 0.28 percent for the 144,853 non-emergency cases and 2.43 percent for the 19,939 emergency cases. As stated above, cases with shock are excluded from these analyses. erefore, volume and mortality rates for the all cases and emergency cases analyses are not directly comparable to some previously published by the Department of Health.
Note on Hospitals Not Performing PCI During Entire 2006-2008 Period Several hospitals began performing PCI during the 2006 - 2008 time period on which this report is based. ese hospitals and the month of the first PCI are listed below. ese hospitals were allowed to perform PCI only on STEMI (heart attack) patients in 2006-2008: Jamaica Hospital Medical Center – May 2006; Faxton – St. Luke’s Healthcare, St. Luke’s Division – February 2008; Brookdale Hospital Medical Center – October 2008; Orange Regional Medical Center
Definitions of key terms are as follows:
e observed mortality rate (OMR) is the observed number of deaths divided by the total number of cases.
e expected mortality rate (EMR) is the sum of the predicted probabilities of death for all patients divided by the total
number of patients.
e risk-adjusted mortality rate (RAMR) is the best estimate, based on the statistical model, of what the provider’s
mortality rate would have been if the provider had a mix of patients similar to the statewide mix. It is obtained by first dividing the OMR by the EMR, and then multiplying that quotient by the statewide mortality rate (0.95 percent in-hospital/30-day mortality for all PCI patients discharged in 2008).
Confidence intervals indicate which hospitals had significantly more or fewer deaths than expected given the risk factors of their patients. Hospitals with significantly higher rates than expected after adjusting for risk are those with confidence intervals entirely above the statewide rate. Hospitals with significantly lower rates than expected, given the severity of illness of their patients before the PCI, have confidence intervals entirely below the statewide rate.