Communications of the ACM
The Analytics of Rehabilitation
Credit: California College of the Arts
With the increase of deaths related to substance, opioid, and alcohol abuse, the need for outcomes' analysis in relation to rehabilitation centers has arisen. Virginia Miori, Kathleen Campbell Garwood, and Catherine Cardamone used data mining to study treatment success rates for the Futures of Palm Beach rehabilitation center. They describe their work in "Forecasting Treatment Outcomes for the Futures Drug and Alcohol Rehabilitation Center," published in the journal Advances in Business and Management Forecasting, Vol. 12.
Miori, associate professor of decision and system sciences at Saint Joseph's University, led the project, which has included two studies so far. "While our first study established the benchmark for data analysis, we were able to fully analyze results with the second," she says. "The healthcare industry lags behind others in the area of analytics, and I saw an opportunity with Futures."
She engaged Campbell Garwood, assistant professor of decision and system sciences, in the collaboration because of her work in data mining. "I like to keep my data analysis based around subjects that can help others," says Campbell Garwood. "This was an ideal position for my interests."
The team used surveys taken by the rehabilitation center's clients and alumni, which included questions regarding any ongoing drug abuse, life experience, and their time at Futures. These interviews were conducted in intervals upon their termination from the rehabilitation program throughout the course of 2016. The demographics for this test were based on gender, age, marital status, primary drug of choice, insurance coverage, and type of discharge from the facility.
The data scientists found that one primary indicator of a client's success was their length of stay at Futures. When a patient stays for 28-to-30 days, their likelihood for success rises exponentially compared to someone who leaves early. While it may seem obvious that their duration of stay would affect a client's success, Miori explains that some may have to leave before they are ready.
"Length of stay often depends on insurance coverage," Miori says. "Insurance companies only allow a limited number of days at a center, and every week the center has to make inquiries as to whether patients can still be covered. So while some leave on their own, others don't have another option."
The researchers' results indicate a need for more studies based on what is successful at rehabilitation centers and the characteristics that make for an accommodating staff.
"I commend Futures for their studies, because they have tried to find data behind what works to raise their reputation and to inform insurance companies of their success," says Miori. "I would recommend other treatment centers take a look at what works and what doesn't when it comes to the rehabilitation process."
No entries found