Study: Detailed EHR Analyses Could Help Identify New Rx Benefits
November 24, 2014 in News
Detailed analysis of electronic health records could be an effective way to identify new prescription drug benefits, according to a study published in the Journal of the American Medical Informatics Association, Health Data Management reports (Goth, Health Data Management, 11/24).
Details of Study
For the study, researchers linked electronic health records from Vanderbilt University Medical Center and Mayo Clinic with the facilities’ tumor registries. The study included:
- 32,415 adult patients diagnosed with cancer at VUMC from 1995 to 2010; and
- 79,258 cancer patients at Mayo (Xu et al., JAMIA, 7/22).
Researchers analyzed the patients’ five-year survival rates when exposed and not exposed to metformin, a drug used “as a first-line therapy” for patients with Type 2 diabetes, according to Health Data Management.
Among VUMC patients, the study found that using metformin as a Type 2 diabetes therapy resulted in a:
- 22% decrease in mortality compared with patients who received other oral hypoglycemic drugs;
- 23% decrease in all-cause mortality compared with non-diabetic, metformin-free patients; and
- 39% decrease in mortality compared with diabetics who received insulin but did not receive metformin.
Researchers said the associations between metformin use and the outcomes were replicated when analyzing the Mayo Clinic EHRs.
In addition, the study found that metformin use was associated with a decrease in mortality from several site-specific cancers, including:
- Breast cancer;
- Colorectal cancer;
- Lung cancer; and
- Prostate cancer.
Reaction, Next Steps
Joshua Denny, associate professor of biomedical informatics and medicine at Vanderbilt and co-author of the study, said, “Our EHR allowed us to delve into details of treatment and response — cancer staging, control of cancer, the various timelines involved and cancer subtypes.”
Denny said the researchers are “now building on this study, pursuing opportunities for using our EHR to look at all drug exposures across a given disease — starting with cancer.” In addition, they are “trying to find other signals that may look like metformin in terms of affecting patient outcomes.”
VUMC officials said their goal is to eventually use a similar strategy of conducting detailed analyses of EHRs for larger groups that are more geographically dispersed (Health Data Management, 11/24).