The ProofPilot Blog - Design, Launch & Participant in Research Studies

How is ProofPilot Different than ResearchKit?

ProofPilot How is ProofPilot Different than ResearchKit?

Apple stoked the imaginations of researchers everywhere with the possibilities presented by ResearchKit, the open source development framework for iPhone research studies. But almost two years after introduction, the platform has few studies with much traction. Is ResearchKit really changing the world of research? Or is it all just Apple-branded buzz?

Apple is the world’s most valuable company. It makes great devices like desktop and laptop computers, iPhones and iPads. Emerging devices that controls your TV and lighting have been introduced via HomeKit. There are rumors Apple is even getting into the medical device and smart car business.

But the company has mixed results beyond hardware. Apple’s Office Suite, iWork, has never blossomed like Google Docs or Microsoft solutions have. Even after a flood of startups created infrastructure to make publishing electronic magazines easy, Apple quietly closed NewsStand, which was meant to upend the publishing market. Startups like Magster, FlipBoard and Texture created their own technology, filled the void and thrive. Udacity, Coursera, and EdEx innovated beyond iTunesU, Apple’s free online course platform. Apple has even ceded leadership in music and movies. Amazon, Spotify, Hulu, and Netflix have innovated far faster and usurped Apple’s once dominant leadership.

So, if history is to repeat itself int he world of research, another player will revolutionize the research space more radically than ResearchKit.

ProofPilot doesn’t have the PR muscle that ResearchKit has. Yet, it does have a sizable feature lead over ResearchKit. Unlike ResearchKit, ProofPilot requires no software developers to launch a study. ProofPilot is platform agnostic, supporting users on iPhone, Android, and desktop, has extensive retention and engagement tools, an area where ResearchKit struggles, and utilizes advanced machine learning to catch bad actors and outright fraud. ProofPilot even supports interventional studies like randomized controlled trials.

Though ResearchKit and its features fall short today, the platform is open sourced, inviting other organizations to fill its feature gap. One group has already stepped up and is pulling together an Android platform ancillary. Android has a sizable market share lead over Apple in the US and an even larger worldwide. Other groups are trying to create “bake your own app” wizards so non-programmers can launch their own ResearchKit apps, much like Inklings and others did for Apple NewStand when it was around.

While not open sourced, ProofPilot is a platform with a sizable library of interventions, measurements, communication tools, rewards, and statistical processes. And unlike ResearchKit, none of these tools need any developers to integrate them. If that experience isn’t available, ProofPilot is architected to easily integrate — and allow you, and others, to use it again and again — also without developers.

Open-source is enormously powerful. ResearchKit provided a much needed imagination and possibility into what technology can do for research. But there are issues that neither Apple nor open source developers will be able to address with ResearchKit. ProofPilot is taking a different strategic, operational and technical approach, and already does:

ProofPilot eliminates the need for developers, data managers, and IT resources. While ResearchKit is only a development framework, ProofPilot requires no programmers or IT resources. To build a ResearchKit app, you need expensive developers and a secure server for the data. These resources are expensive and difficult to find. Few researchers have the management knowledge to oversee such a software development process — and even fewer have the budget to do so.

With ProofPilot, the Ph.D., and extensive research infrastructure is optional. ProofPilot spent nearly two years developing and defining a visual study protocol designer that allows even research novices to choose templates and fill in the blanks to quickly create a study. Before every study launches, ProofPilot partner Veritas IRB, reviews for regulatory and ethical compliance. All data from the study resides on servers that meet or exceed security and privacy requirements.

ProofPilot provides freedom from the onerous app store download experience. ProofPilot’s participant experience uses next generation technology called the progressive web. It works across any web connected device and creates an app-like experience, without downloading the platform from the app store. To enroll in a study, participants click an invite link. That’s it.

But ResearchKit, must be downloaded from the iTunes store before users than have to engage in the enrollment process. Only a small portion of iOS users download a vast majority of apps creating participant bias. Downloading an app is a several-step process requiring potentially forgotten passwords. Each step and irritant reduces the likelihood someone downloads the app. And they still haven’t started the specific study enrollment process. And even if they are downloaded, the likelihood of an app being opened is very low. Most apps aren’t opened more than twice.

ProofPilot tracks participants engaged in multiple studies to reduce bias. Every ResearchKit app is its own walled garden, meaning participants may engage in multiple studies without researcher knowledge. For the current crop of observational studies using ResearchKit, this isn’t an issue. But with interventional studies, it can create a serious problem: it can dramatically skew your results. To eliminate this problem, ProofPilot allows research designers to make candidates enrolled in other studies ineligible.

Here are some other resources on the topic you might find insightful.

ResearchKit Reaches a CrossRoads and Faces some Tough Questions (late 2016)

Apple’s ResearchKit is Not (Yet) Ready for Primetime (But Soon?)(from 2015)

Why Native Apps Don’t Solve Real Health Research Problems and Make Some Worse (2016)

You'll Also Be Interested In

Other Stories in the ProofPilot Blog