Information Technology Reference
In-Depth Information
Note that there is a steady decrease in the number of patients as the Charlson Index increases in
value. Table 3 gives the total number and proportion of patients who died while inpatients compared to
the total number of patients in each of the Charlson categories.
There are some obvious problems with these results. The death rates for all conditions with weight
1 range from a low of 2.09 to a high of 7.76 while those with weight of 2 range from a low of 2.02 to a
high of 6.70, lower than the higher values for weight 1. Moreover, the mortality rate for diabetes with
complications is lower than the rate for diabetes without complications, indicating that the Charlson
Index favors the under-coding of diabetes complications. Some patients with diabetes complications are
under-coded as diabetes with no complications as indicated by the mortality rates.
Weight 3 seems fairly consistent with rates of 10.47 and 12.62 for metastatic cancer and severe liver
disease respectively, but the death rate for HIV with the highest weight of 6 is only 5.09. Moreover, some
very high risk diagnoses are omitted completely from this score. Consider, for example, pneumonia (ICD9
code 486, Table 4) with a mortality rate as high or higher than all conditions with weights 1 and 2.
As a result, patients with pneumonia and no other co-morbidities will receive a Charlson Index of 0
while having a fairly high risk of mortality. Therefore, a hospital with a community flu epidemic that
Table 4. Mortality for pneumonia
Table of Pneumonia by DIED
Pneumonia
DIED
Total
Frequency
Row Pct
Col Pct
0
1
0
7431129
98.21
94.97
135419
1.79
81.02
7566548
1
393728
92.54
5.03
31731
7.46
18.98
425459
Total
7824857
167150
7992007
Table 5. Mortality by TB
Table of TB by DIED
tb
DIED
Total
Frequency
Row Pct
Col Pct
0
1
0
7821310
97.91
99.95
166964
2.09
99.89
7988274
1
3547
95.02
0.05
186
4.98
0.11
3733
Total
7824857
167150
7992007
Frequency Missing = 3041
Search WWH ::
Custom Search