Opinions expressed by Digital Journal contributors are their own.
Measurement-based care — the idea of using patient data that is collected during treatment to guide clinical care — isn’t new. In fact, its benefits have been observed for years, most notably in its ability to better assess patient symptoms and improve treatment outcomes.
However, actually leveraging that data has often been easier aid than done. One of the biggest challenges with trying to use standardized instruments and assessments is that practitioners need to be familiar with their scoring, as well as which interventions should be applied based on the results you get.
Many providers are either unfamiliar with these systems, or they fall into a type of templated medicine, where they rely on a small number of interventions or therapeutic practices. Rather than looking into the hundreds of methods at their disposal, they tend to narrow in on just one dozen or so, in part because that’s easier for our minds to manage.
Jon Read, co-founder of Confidant Health, a virtual care provider focused on helping individuals with mental health and substance use issues, recently addressed these obstacles. As his own work experience has revealed, the introduction of AI has been a watershed moment in enabling stronger measurement-based healthcare, which can have dramatic outcomes for patients.
Understanding the value of measurement-based healthcare
So, what is behind the emphasis on measurement-based healthcare in the first place?
“Measurement-based care is looking to quantify the impact an intervention has on a patient,” Read explains. “So, if I’m trying to improve your symptoms of depression, I’m going to look at that through the lens of its impact on your quality of life — factors like how it affects your sleep, eating, concentration and overall functioning. I would want to use the PHQ-9 to identify if my interventions are having a positive impact, as it provides a standardized way to measure that. The idea is that this allows us to gain insights and then adjust our interventions accordingly.”
By directly measuring whether an intervention method is achieving the desired outcomes, clinicians can subsequently either continue with the current treatment plan or make adjustments as necessary. And when AI is used to offer guidance and support to healthcare practitioners, measurement-based healthcare can become even more effective.
Using AI to gain more detailed insights
So how does AI make a difference? Read suggests that using AI and machine learning as part of a measurement-based healthcare model can actually result in more personalized healthcare by analyzing patient statements and leveraging a database of synthetic data and knowledge of various intervention methods.
“For example, a clinician providing a PHQ-9 might ask a patient if they’ve had trouble sleeping over the last few weeks. The patient says they’ve mostly just had trouble sleeping on the weekends because they get out of work late on Fridays and then go to happy hour to get drinks. In a typical data structure, all that context about binge drinking on Fridays gets lost because the PHQ-9 only records answers like ‘some days’ or ‘most days.’ With an AI analysis, on the other hand, that context is identified so we can tailor specific interventions or therapeutic exercises based on their full response.”
In this instance, AI ensures that additional information contributing to the patient’s answer to the PHQ-9 question isn’t lost. Instead, the full context of the answer is evaluated, and a personalized intervention is recommended — one that isn’t necessarily one of the primary interventions a clinician may have come to fall back on.
Improving the quality of care
Read views AI’s impact on measurement-based healthcare as having tremendous potential for improving quality of care for patients.
“First, we have the fact that the AI can pick up on these conversational nuances that might get missed by a typical evaluation,” he says. “This naturally leads to better care with more personalized interventions. But AI also provides incredible time savings to healthcare providers. It’s very rare for providers to have the time and bandwidth to plan deeply for their sessions with their patients. Just by freeing up some of that time that might otherwise go to billing, scheduling or managing insurance, AI can ensure they provide more personalized care.”
This can also have a direct impact on patient trust, as studies have found patient trust in healthcare providers is often influenced by a doctor’s treatment assurance, ability to provide patient-centered care and familiarity with the patient. As AI enables clinicians to perform better in these and other areas, patient trust and compliance may also improve, further improving treatment outcomes.
Combining AI and measurement-based healthcare for good
The potential benefits of measurement-based healthcare are already well-known, but many clinicians and healthcare facilities have lacked the ability to fully implement this style of care. With the growing capabilities of AI models and machine learning, however, the ability to successfully implement measurement-based care is becoming more of a reality.
Even more valuable, AI makes it easier for clinicians to identify and utilize data that will help them help their patients, increasing the quality of care for all. With AI working to support healthcare practitioners in their efforts to deliver measurement-based care, patients can get the personalized assistance they need to thrive.
