Sunday, June 28, 2015

Can Apple's ResearchKit change the face of medical research?

Every day, Medical News Today reports on a large number of studies. Many of these involve the recruitment of participants so researchers can obtain new and relevant data. The best of these studies will involve the participation of thousands of people, but getting hold of large numbers of willing volunteers can be difficult.

Man on a sofa using a smartphone.
With access to a smartphone, anyone anywhere in the world can now potentially participate in a medical study.
"We have sent out over 60,000 letters," says Kathryn Schmitz, PhD, of Penn Medicine, explaining the recruitment process for a recent study. "Those 60,000 letters have netted 305 women."
Some researchers will acquire a large number of subjects for their studies by using data obtained by previous investigations. This method is much quicker than signing up thousands of volunteers, but it does not give the researcher the control and flexibility that comes with recruiting new volunteers.
In developed countries, smartphones are ubiquitous. Walk down a street in any city and you will likely encounter at least one person with one of these gadgets glued to their hand or head. More than just a way of communicating with other people, for many, smartphones have become an integral part of navigating life due to the wide variety of apps available to users.
iPhones, for example, do not just contain the necessary bits and bobs needed for people to initiate conversations or access the Internet. They contain a myriad of sensors and processors that can track and measure a host of different things. The combination of widespread usage and the capacity to collect data makes smartphones an ideal tool for researchers to utilize.
Here, then, could be a solution to one of the most pressing problems experienced by medical researchers. In this Spotlight, we take a brief look at ResearchKit, the latest initiative launched by Apple, attempting to harness the power of the smartphone to transform how scientists carry out research studies.

Getting started

Smartphone apps are being increasingly used for personal health purposes, both in tracking health and fitness and for detecting illness.
But while these apps can provide users with a perspective on their health, many experts have been critical about the worth of these perspectives, questioning whether they are actually beneficial to users' health.
Last year, Apple launched their Health app and a tool for developers called HealthKit, allowing them to develop software able to pool together health data collected by other apps. The aim of this was to create a range of more empowered health tracking programs, supposedly giving users greater insight into their health than before.
In March, Apple took this work a step further with the launch of ResearchKit, an open source framework that allows developers to create apps specifically for conducting medical research studies. In addition to using data obtained by the Health app, ResearchKit also allows researchers to obtain informed consent from participants remotely.
Screenshot from Asthma Health study app.
The Asthma Health app helps people with asthma treat their condition while helping researchers investigate new ways to personalize treatment.
Image credit: Icahn School of Medicine at Mount Sinai
Dr. Eric Schadt, from Icahn School of Medicine at Mount Sinai, NY, explained to MNT that electronic consenting would eliminate many of the limitations that the traditional consenting process brought to medical research.
"Traditional research has been constrained to informing potential participants about a study, the risks, the benefits and so on, by sitting directly with the participant (or on the phone) to step them through the paperwork and other material to complete the informed consent," he said.
For each participant enrolled into a study, it takes around 30 minutes for a researcher to go through the consenting process. The fact that the process is so time-consuming limits the number of people that can be signed up for a research project due to the amount of time available to researchers.
"With electronic consenting these limitations are all addressed," said Dr. Schadt. "Potential study participants can be stepped through the informed consent using their iPhone, advanced multimedia can be used to more efficiently inform the participant as to the details of the studies as well as the risks and potential benefits."
In order to prevent users from just skipping through the informing stages of the consent process, the apps feature questions to ensure that users can only consent to take part if they fully understand what they are signing up for.
Apple and the researchers behind the first five apps developed using ResearchKit that were launched in March have been very vocal about how the consenting process is a marked improvement on how studies have traditionally been set up.
Dr. Schadt has been involved with one of these first five apps, Asthma Health - an app that combines geolocating data from the phone with air quality information to help users avoid areas where asthma symptoms could be triggered, in turn, collecting data that the developers hope will allow them to discover new ways to personalize treatment of the condition.
On the first day the app was launched, Asthma Health was downloaded by 2,500 people. It would typically take researchers around 1-2 years to recruit this many subjects for a study. In the first month, 7,500 people with asthma had signed up.
This number pales in comparison, however, with the 11,000 people who downloaded the MyHeart Counts app within 24 hours of it becoming available. An element of this may well be the novelty value of being able to sign up to research studies with newfound ease, but initial signs are certainly promising.

Lots and lots of data

The other aspect of ResearchKit that Apple and the app developers are keen to promote is how much data will be obtained by studies built using the framework.
"For the researcher, especially disease research, it is all about the phenotypes," explained Dr. Schadt. "Today, cohorts assembled for disease collect very minimal numbers of phenotypes generally relating directly to the disease, and those phenotypes are collected infrequently - at most, once a year."
Studies conducted via iPhone app will differ in this respect. "With the health apps being built we can collect phenotypes that relate across the disease and wellness spectra and we can do it at high frequency, every day, or 10 or 100 times a day," said Dr. Schadt.
There are a number of inbuilt devices in iPhones that can prove useful to researchers. Data obtained by the microphone, gyroscope, accelerometer, touch screen and the global positioning system of the smartphone can all provide data relevant to research projects.
"All of these variables can be collected numerous times a day over any number of days or years, thus providing a far more detailed profile of the user than could ever be reasonably generated in a medical setting," stated Dr. Schadt.
Not only is the variety of forms of available data beneficial but the frequency by which it is collected is great for researchers too. Many studies are limited by only taking measurements at the baseline - the starting point of the study - and then one or two further times after that.
Infrequent measuring can make results unreliable. Let's say a study wants to examine the effects of a particular diet on body mass index (BMI), and then measures the participants' BMI at the beginning of the study, 6 months later and then after 1 year.
The results of such a study might then show minimal change in BMI after this period. But, if the two periods where BMI was measured came shortly after holidays where the participants drank a lot of alcoholic beverages and did not exercise, the results would inevitably be skewed. This skewing would be exacerbated if the study used a small number of participants.
Such a problem - the ebbing and flowing of symptoms as Apple describes it - is solved by gathering data far more regularly than many studies are able to.
In addition to this, the pool of potential volunteers is much larger with ResearchKit than it is for most studies, typically confined to a small geographic area. The ubiquity of the iPhone means that volunteers can sign up to take part no matter where they live in the world.
This also allows researchers to have access to more varied study populations, although they will be relying on participants that have access to smartphones, still a restricted demographic.

An open source framework for all

What is possibly the most interesting aspect of ResearchKit, however, is that Apple have made the whole thing open source. The core framework can be used by anyone who wishes to use the software to make their own study, and the codes for the five initial apps launched alongside the framework are also available for examination and customization.
Group of three people using smartphones.
While initial access to ResearchKit is confined to iPhones, thanks to the open sourcing of the framework, it will eventually be available to users of every kind of smartphone.
For a company that is renowned for keeping as many aspects of its products as exclusive to the brand as possible, this was a surprising move, but one that could prove incredibly beneficial.
ResearchKit was made available to developers to use from April. Although the initial apps were only available for iPhone users, the fact that the software is open source means that eventually studies will be accessible to users of Android and Windows products.
Science Practice had a go at trying out the newly available framework and commented on how quick it was to construct simple consent forms and participant surveys. Already, too, a small online community has developed in which new ideas and suggestions are being shared on how to use and improve the framework.
A major concern raised with ResearchKit is how private and secure the obtained data will be. Considering how data from the Health app could be loaded up onto the iCloud and how that platform has been hit with several high-profile hacking incidents, the concern is a valid one.
Dr. Schadt informed Yahoo! Tech that "the only people who have access to the data are the investigators of the study," stating that the data is encrypted and meets all industry standards for the transfer of sensitive data. "Apple never touches the data," he reported.
Being an open source framework may also help with privacy, according to Adrian Gropper, chief technology officer of the nonprofit group Patient Privacy Rights.
"Open source encourages people to report bugs in the software and get them fixed," he told Bloomberg Business. "The gold standard is open source because security by obscurity has been shown not to work."

Work in progress

"We believe that these studies represent an entirely new way to perform human health research, putting the participant truly at the center with frequent feedback and control of how their data are used," Dr. Andrew Trister.
"Furthermore, the power of these pervasive devices on quantifying health on an individual is tremendous, opening up real opportunities for truly personalized medicine."
Dr. Trister is a senior physician with Sage Bionetworks, a nonprofit research organization that has worked with Apple to develop ResearchKit.
There are still a number of limitations that the studies conducted by these apps will be subject to. Data provided by the instruments in a smartphone may not be as precise as those obtained using specifically-designed medical instruments. With the study being conducted remotely, it is also harder for investigators to assess the suitability of participants for their study.
Developers might argue that the sheer volume of participants now available to researchers may overshadow these limitations. At this stage, before the results of any of the studies have been processed, it feels as though the positive of having increased and easy access to large numbers of participants is the main point to take from the project.
The most reliable studies are those that are conducted over a long period of time and, as a result, it may be some time before we can effectively gauge the contribution that ResearchKit will make to the field of medical research.

No comments:

Related Posts Plugin for WordPress, Blogger...

Popular Posts