Leveraging a Technology Accelerator to Drive Addiction Treatment Success

THE DIFFICULTIES OF MEASURING BEHAVIORAL HEALTH

One of the major hurdles that behavioral health providers face is measurement of treatment efficacy and success Hayes explained.

Broadly speaking, the nature of psychology is latent or unseen. Psychological change often occurs on a spectrum and requires different measurement tools than those used in medicine.

“In addiction, we know that the nucleus accumbens in the midbrain is sort of the epicenter, or one of the epicenters, of where addiction is stored, if you will, in our brain,” Hayes said. “But we know that in spite of individuals healing their brains, that they might actually return to addiction over time, right? So, this is unlike medicine in the fact that, if I had a broken bone and the bone healed, generally speaking, the area of the break would be even stronger and probably would not break [again].”

Measuring treatment efficacy and success for those with substance use disorders is not straightforward. But capturing data to gain insight into long-term patient success is critical for improving outcomes across the behavioral healthcare landscape.

Cumberland Heights is addressing this challenge by leveraging a combination of data science and an adaptive, cloud-based EMR ‘technology accelerant’ platform from Sunwave Health. With it, Cumberland Heights is capturing both how patients perceive their change trajectories during treatment and how those changes are reflected in the data.

LEVERAGING PATIENT, TREATMENT, AND OUTCOMES DATA TO ASSESS RISK

Once these data are captured, Cumberland Heights staff is tasked with making the gathered insights actionable to improve patient journeys. In addiction treatment, this involves assessing and managing risk.

There are various risks that clinicians must be aware of while treating patients with substance use disorders. Some of the biggest potential risks are patients leaving against medical advice or without completing treatment, relapse, or suicide. Identifying and addressing these requires patient data that is organized and accessible, which Cumberland Heights achieves using the ‘technology accelerant’ platform.

The data is organized in three domains: who the patients are, the treatments they are receiving, and the outcomes Cumberland Heights is observing. The first deals with patient demographics, experiences, and medical history. The second is concerned with the type of treatments patients are receiving, including treatment dosage.The third looks at patient outcomes reflected in the data.

These domains generate a wealth of data, which is pulled from patient records and analyzed to generate risk scores in real time while the patient is receiving treatment. Much of the data is pulled from standardized risk assessments that patients complete upon admission, such as a suicide risk assessment.

Following these assessments, patients may be flagged for follow-up or other interventions based on the risk scores and indicators assigned to them based on their assessment results.

But Cumberland Heights does not only rely on one-time patient assessments to determine risk.

Hayes stated that these data are updated in real-time to reflect any changes in a patient’s record or risk scores. These updates trigger the structured query language (SQL) system to alert staff members that are responsible for that patient of the change. This allows staff to follow up in a timely manner as they do not have to manually log in and search for a patient to see any relevant changes.

In terms of predicting who will relapse or not complete treatment, Cumberland Heights measures multiple factors, including anxiety, depression, impulsivity, readiness to change, recovery capital, craving, trauma, and history of not completing treatment in the past.

Hayes noted that these variables are indicative of which Cumberland Heights patients are more at risk than others. But he also clarified that these may not apply outside the facility’s network, as the only data they pull to generate risk scores and train their predictive algorithms are from patients within the Cumberland Heights Foundation network.

MEASURING PREDICTORS OF SUCCESS POST-TREATMENT

If a patient completes treatment, they’ve managed to take a major step toward recovery, but measuring long-term patient success post-treatment is complex.

Post-discharge, Cumberland Heights has six follow-up windows with patients: 30 days, 60 days, 90 days, six months, nine months, and one year. According to Hayes, patients undergo multiple post-discharge assessments that are based on national outcome measure scales, which evaluate if patients are interacting with law enforcement, attending recovery meetings, or making unplanned trips to the hospital, among other outcomes such as employment status.

Hayes further explained that traditionally, success in addiction treatment had been measured by organizations based on the size of their alumni bases and their patient satisfaction scores. However, Cumberland Heights is working to expand this by measuring the historical predictors of recovery, remission, and relapse as patients move through the post-treatment journey.

These predictors help Cumberland Heights gain a better picture of what the unique profile of a successful patient looks like so that the organization can help provide the tools for success to those undergoing treatment. Hayes likened these tools to those used to build a house.

“There's lots of reasons why people don't have access to really good treatments, because what it takes to be in recovery is having a job, being plugged into community, having medical resources, having a family or support network of individuals that are going to hold you accountable and literally pay for you to make sure you're going to your doctor's appointments and all that,” he stated. “And there are so many people that quite literally don't have the time or cannot afford the luxury of being able to sort of build that house and having the tools associated with building a house.”

These tools are part of what Hayes called recovery capital, which refers to the skills and positive indicators associated with a patient’s success and recovery. While using these self-reported patient data can provide unique insights into patient success, Hayes also cautioned that they could be a limitation in this work.

Since implementing the platform and leveraging these data points to stratify risk and predict outcomes, Cumberland Heights has seen a small but significant reduction in the percentage of patients leaving treatment against medical advice. Hayes also shared that Cumberland Heights is working with other organizations to expand their work in this area.