The rapid increase in overdose deaths has been relentless, and public health officials have scrambled for some kind of response to the problem. Among the best-known solutions are the prescription drug monitoring systems that virtually every state has set up to reduce drug abuse. These systems collect, monitor and analyze electronic prescriptions submitted by pharmacies and doctors, and can flag individuals who might be misusing or abusing painkiller medications.
Overall, these systems have been very effective. Missouri’s Medicaid program, MO HealthNet, for example, saw Vicodin use drop more than 30 percent between January 2012 and January 2015. Other states have reported dramatic drops in the number of patients who “doctor-shop” for more prescription painkillers as well.
But prescription monitoring systems have their limits. For one, they can’t track illicit drug use. For another, to be effective, the systems depend on electronic prescriptions for monitoring, but more than 40 percent of prescriptions are still written on paper. Furthermore, the information the systems provide about overutilization can be late in reaching the doctors who prescribe painkillers.
That’s where analytics come in. The Massachusetts Department of Public Health (DPH) has begun using analytics -- basically software algorithms -- to sift through sets of data to spot patterns and to devise an early warning system about hotspots in the state where possible overdoses and deaths might occur.
Using data in this fashion might enable health officials to contain an outbreak of opioid overdoses before the problem gets worse, says Tom Land, DPH’s director of the Office of Data Management and Outcome Assessment. By interrelating different types of information, such as painkiller prescriptions, data on deaths and other types of information, the state could make more informed decisions on where to use resources quickly and more effectively.
The tools to analyze data have been around for a while, but bringing everything together to thwart opioid abuse in a systematic way hasn’t happened for primarily two reasons. First, it’s challenging to connect and integrate data so that it can be useful when analyzed, according to Land.
Second, states have to spend money in order to have the kind of sophisticated computers that can access, in real-time, data to stop prescription abuse more effectively. Opioid abuse, for instance, is estimated to be over 10 times higher in Medicaid beneficiaries than among private insurance subscribers. Yet states have been slow to modernize their Medicaid systems, says Ellen Bouchery, a program analyst with Mathematica Policy Research. Analytics can not only help identify individuals who are at risk of developing an opioid disorder, but it can also aid in finding the right services for that person at that time. Unfortunately, she says, “it’s going to take a lot of money to get the right technology in place to do that.”
