Claims data found in most prostate cancer patients’ electronic health records (EHRs) have the potential to help estimate their noncancer-specific survival and aid in treatment decision making, according to a recent proof-of-principle study.
The study used data from 57,011 Medicare beneficiaries with localized prostate cancer to establish and test cancer survival and noncancer survival prediction models. The models were built using nearly 9000 distinct insurance claim codes describing comorbid diseases, procedures, surgeries, and diagnostic tests.
These billing codes represent a common “language” that are a part of most EHR systems, according to study researcher James D. Murphy, MD, of the University of California, San Diego.
“While EHR systems vary substantially in design and implementation, nearly all will include claim codes that describe procedures, diagnoses, hospitalizations, and other parts of health care,” Dr Murphy told Cancer Therapy Advisor. “Given the ubiquitous nature of claim codes in EHR systems, in theory, one could implement a risk score into an EHR system that could automatically calculate risk scores for individual patients.”
In patients with prostate cancer, risk scores, or estimation of life expectancy, are of particular interest because they may influence treatment decisions.
“Most men with early-stage prostate cancer will die with their disease, not from their disease,” Dr Murphy said. “In choosing a treatment option, patients and their physicians try to estimate whether they will live long enough to benefit from cancer treatment.”