With medication adherence becoming a focal point across the industry, myriad approaches have arisen to counter the $300 billion challenge, from the relatively simple like text messaging and direct mailers to multifaceted methods that involve increasingly creative uses of technology.
One approach involving the latter is the application of artificial intelligence, and New York-based AiCure is gaining traction with its model, which involves using facial recognition and motion-sensing technologies through mobile devices.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
The company, started in 2010, recently received recognition in the form of $3.4 million in grants from the National Institutes of Health to expand its model. Initially, AiCure carved itself a unique niche within medication adherence by tackling the problem in clinical trials, itself about a $15 billion problem for pharmaceutical companies with often unreliable patients. That, in turn, drags out the process and drives up the cost, according to AiCure CEO Adam Hanina.
“The problem is if you don’t know the patients are taking the medication, you don’t k now if the data you’re capturing are accurate,” Hanina said, adding that compliance within clinical trials is about 50 percent. Clinical trials strive for an 80 percent rate, a number consistent across the healthcare industry in general.
Now, AiCure is looking to expand into the broader space of medication adherence with targeted health systems and the pharmaceutical industry, while simultaneously embarking on a new pilot to monitor and intervene with patients seeking therapy for opioid addiction.
It’s perhaps one of the first companies to take on medication adherence and prescription drug abuse and opioid addiction at the same time, using artificial intelligence technology to monitor an otherwise elusive patient population. It is not the only company applying artificial intelligence, though, with startup Next IT similarly looking to tap into the space to address adherence for chronic illnesses.
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
“In population health management, we’re also looking at high-risk patients like stroke victims, substance abuse, depression,” he said. “The reason why adherence technology is often confusing is, if you can get people to behave correctly, in both scenarios, you’ve solved that problem. So you need to develop a program that brings behavioral change.”
And therein lays AiCure’s approach – by using facial recognition and ensuring that the right person is taking the right medications at the right time, a higher adherence rate can be achieved.
“If you can accurately monitor the behavior, you can hold them accountable and encourage them,” Hanina said. “A lot of technology is self reporting.”
The technology utilizes cameras on mobile devices, capturing the patient data from an application that uses automated algorithms to identify patients, the medication and the process of medication ingestion. The data are then transmitted in real time back to a clinician through a HIPAA-compliant secure network, who can then confirm that adherence is being achieved. It can also flag adverse events and monitor the surrounding environment, ensuring any would-be barriers are spotted and then addressed.
The technology is also scalable to work on any mobile device, Hanina said – a key part of the hoped-for expansion into high-risk population health as more healthcare goes mobile.
Hanina likened the monitoring technology to that of a personal trainer in a gym, who has direct access in viewing a consumer (in this case patient) as they work through a routine. While text messages or email prompts may work for some patients, it’s often not enough to spur a full-on behavioral change for high-risk patients.
There is, of course, a major difference between a personal interaction with a trainer – or a doctor, for that matter – and with a mobile tracking device. But Hanina emphasized that the data is used to help spur positive behavioral habits that lead to better health outcomes, and any concern over privacy is being taken into account. In addition, for patients who do require closer supervision, being able to take their medication at home and not travel to a clinic gives the patient far greater autonomy, Hanina said.
“What’s happening is the Hawthorne Effect – when you watch people, people change their behavior and that’s exactly what you want in a healthcare setting,” he said.
With the new pilot on patients with opioid dependency, the hope is to achieve that behavioral shift with an issue that has significantly worsened with the proliferation of prescription drug abuse such as Oxycontin.
“Fatal overdoses from prescription opiates have quadrupled in the last 15 years and now cause more deaths per year than heroin, cocaine and benzodiazepines combined,” the company said in announcing the initiative with University of Cincinnati. “Overall, the economic cost of opioid abuse is estimated to exceed $55 billion annually. Although adherence to therapy is associated with improved recovery, patients not taking their medication, taking it incorrectly, or giving it or selling it to others impedes patients and health systems from fully benefiting from treatment.”
The National Institute on Drug Abuse has provided $1 million to fund the effort, which will assess assess whether patients using the AiCure platform achieve better adherence and whether adoption of the system can improve treatment duration and reduce the risk of relapse.
The trial is being carried out with the Cincinnati Addiction Research Center at the university. A total of 130 participants will be enrolled over the course of 12 months. Preliminary results of the trial are expected to be published in August 2015.
Numerous other efforts are underway within the medication adherence space, including some companies exploring a microchip on the pill itself or other hardware devices But Hanina said that, while perhaps effective, such an effort would require a significant change in the manufacturing process, which can add to the overall cost.
“If you don’t think the technology is scalable because it requires a widget, it’s not going to be adopted in the mainstream,” he said. “There are different types of hardware devices but they’re not as scalable as mobile devices.”
