Healthcare & Clinical
Charlson Comorbidity Research: The 1987 Cohort and Formula
Examine the original age-comorbidity research, its historical cohort and the arithmetic of its theoretical 10-year model.
What the original index and follow-up measured
Charlson and colleagues developed a weighted comorbidity index using hospital patients and tested it in a historical cohort of 685 women treated for primary breast cancer. In that follow-up the outcome of interest was death attributed to comorbid disease; deaths attributed to cancer were censored in that analysis. This was not a universal all-cause survival model for every person.
The 1987 paper described an age-comorbidity score and calculated predicted 10-year survival against a theoretical low-risk reference of 0.983, using an exponent exp(0.9 × score). Its table compared modelled and observed estimates for score groups 0 through 5. The historical cohort, endpoint, baseline and limited number of participants in some groups constrain interpretation today.
| Historical score group | Observed 10-year study estimate (%) | Paper’s rounded prediction (%) |
|---|---|---|
| 0 | 99 | 99 |
| 2 | 87 | 90 |
| 3 | 79 | 77 |
| 4 | 47 | 53 |
| 5 | 34 | 21 |
Reproduce the historical equation for a fictional score
For scores 0 through 5, the formula below evaluates 100 × 0.983^EXP(0.9 × score). For example, a fictional score of 4 produces 53.39 using the unrounded expression, while the original table rounded its prediction to 53. This demonstrates historical model arithmetic only.
The calculation does not assign diagnoses, resolve overlapping comorbidity categories or provide an individual life-expectancy forecast. Different versions and coding rules exist; the 1994 combined-index study and the 1987 study should be interpreted in their own populations.
=IF(COUNT(B2)<>1,"Enter one numeric score",IF(OR(B2<0,B2>5,MOD(B2,1)<>0),"Enter a whole number from 0 to 5",ROUND(100*0.983^EXP(0.9*B2),2)))Try the example
Use the original values or change them to test your understanding.