Skip to main content
. 2003 Aug;38(4):1103–1120. doi: 10.1111/1475-6773.00165

Table 3.

Comorbidity Scores and Their Performance in a New Jersey Medicare/ PAAD Population in Predicting One-Year Mortality

New Jersey /PAAD

Score Model Degrees of Freedom c 95% Confidence Interval
Age + gender 2 0.658 (0.654–0.662)
CDS-1 Age + gender + CDS-1 3 0.703 (0.699–0.707)
Romano (1 score) Age + gender + Romano 3 0.771 (0.767–0.775)
Romano + CDS-1 Age + gender + Romano + CDS-1 4 0.777 (0.773–0.780)
Romano +# of RX* Age + gender + Romano + numb. of Rx 4 0.775 (0.771–0.779)
Romano (17 cat.) Age + gender + Romano17 19 0.781 (0.777–0.784)
Elixhauser (30 cat.) Age + gender + Elixhauser30 32 0.781 (0.778–0.785)
Elixhauser plus †† Age + gender + Elixhauser30+ AMI + stroke 34 0.783 (0.779–0.786)
*

Prescription medications that have different chemical structures but may be of the same therapeutic group.

All Romano coded conditions were included as 17 binary categories in the model: 1=condition present) or 0=condition absent.

††

Thirty Elixhauser indicators based on ICD-9-CM diagnoses plus two indicators for acute myocardial infarction (AMI) and stroke according to Romano's ICD-9-CM coding.