How Purdue Pharma Hopes To Use The Apple ResearchKit

By Ed Miseta, Chief Editor, Clinical Leader

With the continuing effort of many in the industry to bring mobile and wearable technologies to patients, Larry Pickett, CIO for Purdue Pharma, is quick to note that no one has all the answers. Although many in the industry will always seem anxious to jump on the latest technology wave, no one can accurately predict where it will eventually end up. “When the iPad first came out, there were very intelligent technology people who said consumers would never adopt it,” he says. “They could not have been more wrong.”

Few seem to doubt that mHealth technologies will pave the way to more efficient trials and a better patient experience. But as with any new technology, someone still has to be the pioneer to prove its viability in a clinical trial. Purdue Pharma is attempting to do just that by evaluating the use of the iPhone, Apple Watch, ResearchKit, and various healthcare apps for patients suffering from pain.

ResearchKit is an open source framework that was released by Apple last year. Using sensor technology specifically associated with the watch, developers can use the API to create apps that will allow sponsors to track, collect, and record data based on a multitude of measures.

There are three customizable modules for companies interested in conducting clinical trials: The survey module collects patient information, the consent module captures the patient’s consent and shares information about the trial, and the active task module tracks patient activity levels. The latter features voice capture, which may allow physicians to detect pain levels based on the speaking patterns of the patient. For certain conditions, the speed at which information is entered into a device can also be indicative of patient pain levels.

“We are particularly interested in the Apple Watch as a sensor device not because of what it can do today, but the potential it holds for the future,” notes Pickett. “When I look at the watch today, it may not have features that would compel me to buy and wear one. But we know it will be improved over the years and continue to gain additional capabilities. Today it might measure heart rate and activity levels. But there is a huge potential for the watch to eventually allow us to better connect with the patient experience. We may soon be able to use it to remind patients to take their medication or communicate pain levels to a physician. It’s that potential that has us evaluating it as a game changer in the clinical space.”

BETTER PATIENT-REPORTED OUTCOMES
Purdue is attempting to use Apple technology to give patients the opportunity to communicate their pain levels over a long period of time. Pickett believes this is a critical feature, as it will enable a physician, looking at a portal, to know how patients are feeling and what their pain trends are before they ever step foot in the office. In the past, a physician might know only how patients felt the morning of their visits, but not see those pain levels over a lengthier period of time. Pickett notes there may even be a role for gamification (the mixing of entertainment with self-monitoring) to help patients set goals for their health and wellness while providing feedback.

“I see many benefits coming from this technology,” notes Pickett. “First, there is information being supplied remotely from patients to physicians over a long period of time. There is also the opportunity to correlate that information with factors such as sleep, activity levels, or heart rate to determine connections between them. There is also a good opportunity for patient engagement. There is anecdotal evidence noting that if you can engage patients in the management of their own health, they are more likely to take ownership of it. I think we will see more of that in the future.”

While the opportunities are promising, Pickett notes there are some disadvantages to the Apple Watch. One is that the battery life is very short and typically has to be charged every night. Cost is also a concern. The watch has to be paired with an iPhone for certain capabilities such as GPS. In an effort to scale up the use of the technology in trials, the use of both devices could become prohibitive. Many patients would not be able to afford an Apple Watch or iPhone, which would require the sponsor to absorb the cost of supplying them. “We will likely fund a study that will provide the devices so as to learn more about them in a limited way, but I do have concerns about that cost going forward,” Pickett says.

Because of both of these issues, Purdue is looking at other devices, hoping there might be a companion product that patients could wear. For example, someone wearing an Apple Watch to monitor pain might also wear a $30 device to monitor their sleep. Pickett notes some of these devices, such as the Misfit, have a six-month battery life, are small and light, and are more comfortable to wear to bed.

ARCHITECTURE IS IMPORTANT
One vital consideration to incorporating mobile technologies into a trial is deciding on the appropriate information to capture. Having that knowledge also will help with understanding how the technology can best impact the lives of patients. According to Pickett, that is why Purdue is performing this investigative work. “We have to make an impact on the lives of patients, improve their outcomes, and reduce their pain,” he notes. “The key to doing that properly is studying the tools that are available, deciding what information is the most important, and providing patients with the tools [phones, watches, and apps] to make it happen. Then we can take what we learn and keep improving the process going forward.”

Pickett is convinced that Purdue will have no problems with either the watch or the phone; he says those technologies are pretty straightforward. But getting the information to the portal or the EHR (electronic health record) does require additional expertise. Therefore, it is necessary for them to partner with technology companies to enable those capabilities. Purdue would still be very much involved in mapping things out.

“Without getting into an overly technical discussion, our architecture integrates between the phone and the watch using various servers and technologies,” he says. “It is not that complicated, and it does not take long to do. For example, I know of one developer who got the ResearchKit when it came out and churned out an application in just four weeks. If you know Swift or Objective C (two programming languages) it is not that difficult. But you really need to put some thought into the architecture you put in place, especially in terms of where you collect and store the data and how it will move between those locations. What we are most interested in is how the information will get into the physician portal and then ultimately into an EHR.”

One innovative idea is for patients to actually have access to their EHR. They would be able to see it and actually monitor and edit some of the data. At this point it is not clear how that might be done, but Pickett notes it is something researchers are discussing. It’s certainly an innovative idea that would empower patients by giving them ownership over their data.

"We are particularly interested in the Apple Watch as a sensor device not because of what it can do today, but the potential it holds for the future."

Larry Pickett
CIO, Purdue Pharma

 

TWO POTENTIAL APPLICATIONS
Pickett notes there are two projects being examined at Purdue, but stresses at this time nothing has been finalized. One involves the use of an iPhone app, while the other looks to incorporate the Apple Watch.

With the Apple Watch, there is a small amount of space to work with. Therefore, anything involving the watch should be kept simple. Indicating pain level is one option being examined. By pressing a plus or minus button, patients can increase or decrease their current pain level and then record it. The information will go from the watch to the iPhone to the EHR portal. Additional measures, such as the patient’s heart rate, could also be recorded.

With the iPhone app, patients would learn how the study works, register their information, complete surveys, and perform informed consent. The iPhone also allows for the sharing of a short video or slideshow to provide additional information about the device, the study, and how information will be transferred.

Questions asked in the survey might include age, sex, where they live, and any other information that would qualify them for the study. If a patient is eligible, they could then enroll and start recording their pain levels via the phone. One of the options being evaluated is showing patients a rotating 3D image of the human body. By simply tapping on the phone, individuals are able to indicate where pain exists. By tapping multiple times, they are able to increase the severity level of the pain.

“We also feel it might be helpful to try and correlate a person’s pain with their activity level, as measured by their number of daily steps they take,” says Pickett. “The pain level is a patient-reported outcome, and what the patient is feeling is certainly subjective, but what they have to say is valuable and important. Over a period of time, be it weeks, months, or years, there are good insights that could be derived from this data.”

The use of this app does raise a few concerns, which Purdue is evaluating. One is adverse events. If patients feel they have experienced an adverse event, Purdue would like a quick and easy method for them to report it. Pickett notes they are evaluating the option of having patients click on a button and be automatically connected with either a physician or nurse practitioner.

A second concern is data privacy, which Pickett describes as absolutely critical to any study. For that reason, he likes the idea of patients having control over their data, including the ability to delete the information if they decide to do so, although he stresses that this is his personal opinion, and that it may, in fact, not be compatible or permissible in a clinical trial.

Medication reminders are also being evaluated. Pickett states they are an important in any pain study. “If a patient is feeling really good, it’s possible they will forget to take their medicine. Oftentimes they will not remember to do so until the pain suddenly gets severe in a short amount of time. Sending those reminders once or twice a day can keep the patients taking their medicines on time and prevent those situations from occurring.”

NEXT STEPS
At this time, Pickett is feeling very good about his initial assessment of ResearchKit. He notes Purdue has (fairly quickly) developed a proof of concept application for both the iPhone and the watch. The company is also discussing with a partner the best method to transmit the data into a portal or EHR. The next version of both the watch and ResearchKit will soon be hitting the market, and he is excited about the prospect of additional sensors being incorporated into devices and how his company might be able to use them.

“We think this is just the beginning,” he adds. “We still need to figure out how to apply data science and predictive analytics to combine the data we will be collecting with other data sets that may be available. There is a lot of data integration that will have to happen, and we are in need of new techniques for doing that. They need to be visual, they need to be easy to search and retrieve, and they need to tell us things we do not already know. We will have to do this in a smart way in order to make it pay off for both pharma and all of our patients.”