Charlson Comorbidity Index Varies More among Older People in Four U.S. Health Systems: Taylor’s Law and Fluctuation Scaling

Joel E. E. Cohen, Rockefeller University and Columbia University
Jonathan N. Tobin, Clinical Directors Network

The enhanced Charlson Comorbidity Index (eCCI) combines a person’s concurrent chronic medical conditions (comorbidities) in 39 categories into an index. We show that the average eCCI and the variance of eCCI of people of each age increase with age in de-identified electronic health records of 238,156 adults in low-income communities served by four health systems in the United States (2 in Chicago, 2 in New York City). The variance of eCCI approximates a power function of the mean eCCI. This quantitative relationship, not previously recognized, approximates Taylor's power law of fluctuation scaling. The quintiles of eCCI diverge with increasing age, consistent with an increasing variance. Within almost every age group, the frequency distribution of eCCI per individual is well approximated by a negative binomial distribution. This frequency distribution of eCCI per individual of a given age can account approximately for the relation of variance to mean across ages. The increase with age in the mean and the variance of eCCI, and the faster increase of the variance compared to the mean, show that aging brings more comorbidity on average and greater differences in comorbidity among individuals of more advanced ages. Clinical practice should recognize and respond to the greater differences in comorbidity among older people.

Keywords: Health and Morbidity, Data and Methods, Mortality and Longevity, Older Adults and Intergenerational Relations

See paper.

  Presented in Session P2. Data & Methods; Older Adults & Intergenerational Relation; Health, Morbidity, Longevity and Mortality; Population Ageing