Table 3 Effect of risk-adjustment approach on standardized mortality/morbidity ratio chart. a) Number of clinics with observed-expected ratio > 1. b) Cumulative absolute distances of observed-expected ratio of all clinics to the standard value 1
From: Comparison of three different methods for risk adjustment in neonatal medicine
| Â | Raw values | Indirect standardization | Logistic regression | Multilevel |
|---|---|---|---|---|
Outcome | a) SMR values above 1 | |||
Mortality | 3 | 3 | 3 | 3 |
In hosp. Mortality | 3 | 3 | 3 | 3 |
Late onset sepsis | 4 | 2 | 2 | 2 |
NECa | 3 | 4 | 3 | 3 |
sIVHb | 5 | 5 | 6 | 6 |
BPDc | 4 | 4 | 4 | 4 |
sROPd | 2 | 2 | 1 | 2 |
| Â | b) Cumulative absolute distances to 1 | |||
Mortality | 2.8 | 2.7 | 3.0 | 3.0 |
In hosp. Mortality | 3.0 | 3.2 | 3.1 | 3.0 |
Late onset sepsis | 2.7 | 3.0 | 3.0 | 3.1 |
NEC | 3.4 | 3.4 | 3.6 | 3.6 |
sIVH | 2.2 | 2.3 | 2.4 | 2.4 |
BPD | 4.1 | 3.2 | 3.8 | 3.8 |
sROP | 5.7 | 5.4 | 5.4 | 5.5 |
Total | 23.9 | 23.2 | 24.3 | 24.4 |