The Charlson Comorbidity Index (CCI) is calculated by assigning a weighted score to each of 17 medical conditions a patient has, then adding those scores together. Each condition carries a weight of 1, 2, 3, or 6 points based on how strongly it predicts one-year mortality. The total score represents a patient’s overall burden of chronic disease.
The 17 Conditions and Their Weights
Every condition in the CCI falls into one of four point categories. The higher the weight, the greater that condition’s independent impact on mortality risk.
1 point each:
- History of heart attack (myocardial infarction)
- Congestive heart failure
- Peripheral vascular disease
- Cerebrovascular disease (stroke or TIA, without paralysis)
- Dementia
- Chronic pulmonary disease (including asthma)
- Connective tissue disease (lupus, rheumatoid arthritis, polymyositis)
- Peptic ulcer disease
- Mild liver disease (chronic hepatitis or cirrhosis without portal hypertension)
- Diabetes without organ damage
2 points each:
- Hemiplegia or paraplegia
- Moderate or severe kidney disease (on dialysis, post-transplant, or with significantly elevated creatinine)
- Diabetes with end-organ damage (retinopathy, neuropathy, or nephropathy)
- Non-metastatic solid tumor (treated within the last 5 years, excluding non-melanoma skin cancers)
- Leukemia or lymphoma
3 points:
- Moderate or severe liver disease (cirrhosis with portal hypertension, with or without variceal bleeding)
6 points:
- AIDS (HIV with an opportunistic infection or AIDS-related complex)
Note that HIV infection alone, without an AIDS-defining condition, is scored separately at 3 points in some updated coding schemes. The original 1987 index assigned 6 points only to AIDS, reflecting the era’s mortality rates.
How Severity Levels Work
Several conditions appear at two different severity levels, and only the higher-scoring version counts. A patient with mild liver disease (1 point) and moderate-to-severe liver disease (3 points) does not receive 4 points total. They receive 3. The same logic applies to diabetes: uncomplicated diabetes scores 1 point, but if there is end-organ damage like retinopathy or nephropathy, only the 2-point version is counted.
Each condition also has specific qualifying criteria. Congestive heart failure, for example, must be symptomatic and have responded to treatment. Peripheral vascular disease includes intermittent claudication, arterial bypass for insufficiency, gangrene, or an untreated aneurysm 6 centimeters or larger. Chronic pulmonary disease requires symptomatic shortness of breath from a chronic respiratory condition, not just a diagnosis on paper. These definitions matter because a condition only earns points if it meets the threshold of clinical significance the index was designed around.
Adding the Age Adjustment
The age-adjusted version of the CCI (sometimes called ACCI) adds 1 point for every decade of life past age 40, up to a maximum of 4 additional points. A 52-year-old adds 1 point. A 65-year-old adds 2. An 80-year-old adds 4. This adjustment accounts for the independent effect of aging on mortality, separate from any specific disease.
Whether the age adjustment is used depends on context. Many clinical studies report both the standard CCI and the age-adjusted version. If you are looking at a CCI score in a medical record or research paper, check whether age points were included, because the same patient can have meaningfully different scores depending on the version.
What the Total Score Means
The total CCI score translates roughly into one-year mortality risk. In the original validation study by Mary Charlson and colleagues, the breakdown looked like this:
- Score of 0: approximately 12% one-year mortality
- Score of 1 to 2: approximately 26%
- Score of 3 to 4: approximately 52%
- Score of 5 or higher: approximately 85%
These numbers came from a hospitalized patient population, so they reflect baseline risk for people already sick enough to be admitted. The index is not designed to predict your personal life expectancy. It is used by researchers and clinicians to compare groups of patients, adjust for differences in chronic disease burden across studies, and guide treatment decisions where comorbidity load matters.
How CCI Is Calculated Automatically
In practice, most CCI scores today are not calculated by hand. Hospitals and research teams extract scores automatically from electronic health records using diagnosis codes. The standard approach maps each of the 17 CCI conditions to families of ICD-10 codes. Diabetes, for instance, maps to the E08 through E13 code families. AIDS maps to B20 combined with codes for specific opportunistic infections like Pneumocystis pneumonia, Kaposi’s sarcoma, or toxoplasmosis of the brain.
This automated process has its own complexity. Different countries use slightly different versions of ICD-10, which means the code mappings vary. The Canadian ICD-10 version developed by Quan and colleagues uses diabetes codes E10 through E14, while the U.S. version includes E08 and E09 instead of E12 and E14. If you are comparing CCI scores across international datasets, these differences can introduce inconsistencies.
Some institutions go further, building algorithms that search the free text of clinical notes for synonyms, abbreviations, and symptoms associated with each condition, while excluding negation terms (phrases like “no history of” or “ruled out”). A Mayo Clinic team refined this kind of text-mining approach to reach sensitivity and specificity above 95% for extracting CCI comorbidities, scanning five years of medical records before a patient’s admission date.
CCI Compared to Other Comorbidity Measures
The CCI is the most widely used comorbidity index, but it is not the only one. The Elixhauser Comorbidity Index is a common alternative that tracks 31 conditions instead of 17 and does not collapse them into a single weighted score (though weighted versions exist). In head-to-head comparisons, the two perform similarly for predicting mortality. A study of total ankle replacement outcomes found both indices predicted 180-day mortality with an area under the curve of 0.88, meaning excellent and essentially identical predictive accuracy.
Where the indices differ is scope. The Elixhauser captures conditions like depression, obesity, and alcohol abuse that the CCI does not include. For outcomes beyond mortality, such as hospital readmission or length of stay, one index may outperform the other depending on the patient population. The CCI’s advantage is simplicity: a single number that is easy to calculate, widely understood, and validated across decades of research in dozens of clinical settings.