According to a new study, an algorithm to help prevent suicide, and works like a cholesterol number works to prevent heart disease, holds hope for bringing down veteran suicide rates. From the New York Times (again!):
In a study published Thursday in The American Journal of Public Health, researchers reported that a computer algorithm using hundreds of variables among millions of V.A. patients was able to correctly predict small subgroups with suicide rates up to 80 times higher than V.A. patients as a whole. The study also found that current practices that rely on doctors and other medical staff to flag high-risk patients miss the vast majority of such veterans. [emphasis added]
According to iHealthBeat, 381 variables among the millions of patients were searched for patterns. And voila:
Using Veterans Health Administration (VHA) health system electronic medical record data, Veterans Affairs (VA) and National Institute of Mental Health (NIMH) scientists were able to identify very small groups of individuals within the VHA’s patient population with very high, predicted suicide risk — most of whom had not been identified for suicide risk by clinicians. Such methods can help the VHA to target suicide prevention efforts for patients at high risk, and may have more wide-ranging benefits. [emphasis added]
As mentioned in the National Institutes of Health press release on the study, this deployment of big data in the health care world holds promise beyond any one group of patients. This should not come as a surprise to Northeast Ohioans who follow news in open and big data. Cleveland-based Explorys was bought recently by IBM because of the predictive potential of its massive trove of big data health data sets. And on the public health side, the Health Improvement Partnership-Cuyahoga just launched its data-driven approach to combat poor health outcomes for urban dwellers, particularly minorities.
It’s possible that tomorrow the NYT will again publish news about how some such cache of potentially life-saving data got hacked and the hacker got a bounty. However, for today, let’s reinforce how making data open and interoperable leads to discoveries we could not otherwise stumble upon.