AI Opens Door To Improving Patient Experience

Pharmaceutical Executive, Pharmaceutical Executive-08-01-2022, Volume 42, Issue 8

The technology is not just about digital enablement anymore.

In a previous column, one of my colleagues wrote about how artificial intelligence (AI)-based technology is set to play a big role in the pharma industry’s promotional activities. Based on recent conversations with several industry executives, these programs have a direct use in the industry.

During the COVID-19 pandemic, the pharma industry embraced digital channels to engage with patients. This made it possible to implement AI programs to be used not just for data collection, but to directly impact the patient experience. Using advanced algorithms, these programs help healthcare providers source and catalog many of the experiences they have with patients, even when the patients are in their own homes.

Amy Brown, founder and CEO of Authenticx, recently spoke with Pharm Exec about her work providing near real-time patient and provider feedback to life sciences companies. Her organization’s program uses AI to sort through customer conversations so that they can be cataloged and analyzed based on pharma-relevant criteria. For example, the program can find all the calls where patients discuss a certain side effect.

She explains that the problem many companies face is that many of these patient conversations are outsourced to third-party partners. One of her pharmaceutical clients may have eight to 12 outsourced vendors handling these conversations, which can make it difficult for the companies to have direct access to what their patients or clients are saying.

When AI is used, however, all of the data from these various sources is brought together to tell a “comprehensive story,” as Amy puts it.

“What we’ve found is that the pharmaceutical industry deeply values and feels a sense of responsibility for their patient or provider experience, yet they don’t have direct ownership of those conversations in many cases because they outsourced it,” she says. “We’re helping them bring that data together so that they can understand what’s being delivered in terms of a patient experience.”

By analyzing these conversations, pharmaceutical companies can learn more than just their patients’ symptoms. Patients often reveal details about what it’s like to have an illness, disease, or diagnosis and the struggles that come from living with that. Being able to collect this data allows life sciences companies to learn what social, economic, or environment situations are also impacting their patients.

Dr. Rich Christie, MD, PhD, chief medical officer at AiCure, spoke with Pharm Exec about how his company’s AI helps pharma companies running clinical trials to confirm that the patients are properly following their programs. For example, the software can use a patient’s phone to confirm something as simple as whether or not they’re taking their pills. The program can recognize the patient’s face and, by using the camera, can be used to confirm that the pills were actually swallowed.

“That seems like a small thing,” he says, “but in the world of pharma, where you’re trying to develop medication and run these massive clinical trials, just understanding who actually took their medication vs. who didn’t really helps you understand what you’re doing and who’s generating good, quality data.”

As he explains, the AI is used to fully understand the patient experience. The program can build models, based on patient data, that can predict which patients are taking their medication and who will continue to take it.

According to Christie, it’s not uncommon for patients to struggle to maintain and follow a treatment plan, for a variety of reasons; patients may forget, or they may struggle to follow a complicated program that involves taking different pills on different days. For some patients, the program can simply be used to help remind them to properly follow the plan. For others, the AI can be used to recognize certain patterns in the patient’s behavior and symptoms and then adjust the treatment regimen to fit their specific needs.

“Let’s say a patient has a tremor from Parkinson’s disease,” he says. “You notice that they take their medication and their tremor is okay, but then it starts to get worse at 8 [p.m.]. Maybe you send a flag back to the physician so they can change how and when [the patient] takes their medication so that it’s working optimally.”

Christie says that this data can even be used to predict when patients who have episodic diseases, such as multiple sclerosis, are about to have a flare-up. These patients could potentially be connected to medical care to better manage those situations.

Mike Hollan is Pharm Exec’s Editor. He can be reached at mhollan@mjhlifesciences.com.