Veterans Aging Cohort Study–Charlson Comorbidity Index (VACS-CCI)

The VACS-CCI is a mortality prediction tool that combines elements of the Veterans Aging Cohort Study (VACS) Index with the Charlson Comorbidity Index (CCI). It was developed to improve prediction of all-cause mortality by blending physiologic markers from routine labs and BMI with age and coded comorbid conditions. In the recent development study, the model was built in a Veterans Health Administration population of about 6.5 million veterans and was designed to work in people with and without HIV.

Why the VACS-CCI was created

The traditional CCI captures comorbid diagnoses, but it does not directly measure current physiologic injury. The VACS Index 2.0 adds biomarkers that reflect liver function, renal function, anemia, inflammation, nutrition, and body habitus. The VACS-CCI was created because adding Charlson comorbidity groupings to the VACS framework improved prognostic accuracy compared with either approach alone.

What the index includes

In the prostate-cancer validation paper, the authors describe the VACS-derived inputs as age, BMI, hemoglobin, platelets, AST, ALT, creatinine, white blood count, albumin, and hepatitis C status, with FIB-4 and eGFR used as validated composite markers derived from those data. The model then adds Charlson comorbidity components extracted from diagnosis codes.

That structure is one reason the tool is appealing in health systems: most of its inputs are already present in the electronic record, so the score can be calculated with relatively little manual entry. The prostate-cancer paper explicitly describes it as EHR-compatible and says electronic health record–assisted calculation is feasible.

Performance in the general VA population

In the 2024 Frontiers study that introduced the VACS-CCI in a broad VA population, the investigators reported that VACS Index 2.0, CCI, and VACS-CCI all improved discrimination, but VACS-CCI was more consistently calibrated overall. Their conclusion was that combining VACS and Charlson information improved prognostic accuracy for all-cause mortality in veterans with and without HIV.

This is clinically important because good calibration matters when a score is used to estimate an actual mortality probability rather than simply rank patients from lower to higher risk. The authors emphasized that VACS-CCI offered a practical balance of discrimination and calibration across subgroups.

Role in non-metastatic prostate cancer

A 2024 validation study specifically recalibrated VACS-CCI for men with biopsy-confirmed nonmetastatic prostate cancer in the VA. In that cohort of 107,370 individuals, there were 24,977 deaths, and 86% were from causes other than prostate cancer. Compared with age plus CCI alone, VACS-CCI had better validated discrimination for non-prostate cancer mortality, improving the C statistic from 0.67 to 0.75. The authors concluded that VACS-CCI was ready for implementation within VA for this use.

The paper also gives a practical sense of risk stratification. Among men with a mean age of 65 years, those with VACS-CCI scores 30–34 had a mean estimated 10-year all-cause mortality of 15.5%, scores 35–39 had 28.8%, scores 40–44 had 41.3%, and scores 45–49 had 62.3%.

Strengths

The main strengths of VACS-CCI are that it is:

• More comprehensive than CCI alone, because it combines diagnoses with current physiologic status.
• Practical for EHR automation, since it uses routinely collected variables.
• Validated in a very large VA population and then recalibrated for a specific clinical setting, namely non-metastatic prostate cancer.

Limitations

The biggest limitation is generalizability. The prostate-cancer validation paper states that the score should be validated outside the VA prior to outside application.

Another issue is missing data. In the prostate-cancer cohort, 14% were excluded because of missing data in the primary analysis, though the authors also performed sensitivity analyses using assumptions for missing lab values.

Finally, VACS-CCI predicts all-cause or competing-cause mortality, not disease-specific outcomes by itself. In prostate cancer, it complements rather than replaces tools that estimate prostate cancer mortality.

Bottom line

The VACS-CCI is an EHR-friendly mortality index that combines Charlson comorbidities with VACS physiologic markers to improve prediction of long-term mortality. It appears especially useful when clinicians need a more realistic estimate of competing mortality risk than age or comorbidity count alone can provide, including in men with non-metastatic prostate cancer. Within VA populations, it has shown better performance than age plus CCI alone, and recent authors consider it ready for operational use in that setting.

References

1. McGinnis KA, Justice AC, Marconi VC, et al. Combining Charlson comorbidity and VACS indices improves prognostic accuracy for all-cause mortality for patients with and without HIV in the Veterans Health Administration. Frontiers in Medicine. 2024.
2. Justice AC, Leppert JT, and colleagues. Adaption and National Validation of a Tool for Predicting Mortality from Other Causes Among Men with Nonmetastatic Prostate Cancer. European Urology Oncology. 2024.
3. McGinnis KA, Justice AC, Tate JP, et al. Discrimination and Calibration of the Veterans Aging Cohort Study Index 2.0. Clinical Infectious Diseases. 2022. This supports the role of VACS Index 2.0 as the physiologic backbone later incorporated into VACS-CCI.
4. Justice AC, et al. Strengths and Limitations of the Veterans Aging Cohort Study Index as a Measure of Physiologic Injury in People With HIV. 2019 review. Useful background on why VACS biomarkers add information beyond diagnosis-based comorbidity alone.