U.S. flag An official website of the United States government
  1. Home
  2. News & Events
  3. Speeches by FDA Officials
  4. Harnessing Real World Evidence for Safety and Innovation - 11/19/2018
  1. Speeches by FDA Officials


Event Title
Harnessing Real World Evidence for Safety and Innovation
November 19, 2018

Speech by
Scott Gottlieb, M.D.
Commissioner of Food and Drugs - Food and Drug Administration ( May 2017 - April 2019 )

Harnessing Real World Evidence for Safety and Innovation
Remarks by Scott Gottlieb, M.D.
Commissioner of Food and Drugs
Reagan-Udall Foundation

(Remarks as prepared for delivery)

Good morning.  It’s a pleasure to be here.

Our longstanding goal for medical care is to ensure that the right drug or device is delivered to the right patient at the right time.

This vision is increasingly achievable.

It’s finally at hand.

Innovation in drugs and medical technologies are becoming available that make these personalized approaches more routine.

It includes targeted therapies, tissue agnostic drugs, next generation sequencing platforms, 3D-printed medical devices, and regenerative medicines like gene and cell-based therapies.

These are just some of the new approaches that are enabling more precise and personalized medical interventions.

In more cases, these products can offer outright cures for chronic ailments that once required a lifetime of medical care. 

We want to efficiently advance these opportunities for patients in a timely, safe way. This requires us to modernize our framework for the development and review of products in ways that take measure of the unique attributes of the kinds of treatments that are coming along.

The Reagan Udall Foundation (RUF) is a critical part of these efforts. And we’re deeply appreciative of the work done by this organization.

For instance, we’ve partnered with the RUF to make the agency’s active postmarket surveillance program for drugs -- our Sentinel system -- available to support public health research. And to help sponsors test new approaches for evaluating safety signals, for implementing postmarket studies, and for assessing the impact of risk management tools.

These efforts highlight a critical opportunity we have -- to advance healthcare, and promote advances in how we develop new medical products and how patients access them. And that opportunity is the growing convergence between digital and traditional healthcare.

When people think about digital health they think about practical tools for patients. But for us, this convergence is playing out across the entire continuum of development, review, and patient care.

It applies to the kinds of medical products that are being developed. And how those products are being delivered to patients. Those opportunities are becoming obvious. But it also applies to how we develop new products. And how the FDA goes about reviewing them.

I want to touch on all of these opportunities -- and announce today some of the new steps that FDA is taking to promote this convergence.

The opportunities offered by digital health starts with the way products are developed. To take one example; our ongoing, agency wide digital pre-market IND submission program allows us to create new visual analytics for monitoring safety issues in the pre-market setting.

FDA’s Center for Devices and Radiological Health (CDRH) is working with the National Evaluation Center for Health Technology to test new computer modeling of patient avatars, which can reduce patient exposure to potentially harmful products.

And, across our medical product centers, we’re exploring how we can harness real world evidence through electronic health records, apps, and other mobile health technologies to collect more reliable, continuous data on products’ benefit and risk profiles in the pre- and postmarket.

These approaches are still in their early stages. But they’re already helping our reviewers translate science and data into the evidence we need to make more informed and efficient regulatory decisions.

It’s hard to overstate how quickly the mHealth field is changing the efficiency by which products are developed and reviewed -- and the kinds of clinical opportunities we’re ultimately able to offer patients.

In just the past three years, more than 150,000 mobile health apps have been launched for smart devices.

We recognize the enormous opportunities apps present for increased use of digital tech that’s coupled to the delivery of drugs and biologics.

In new ways that can support patient health and well being, improve compliance, and help patients share accurate data with caregivers.

But we’ll never fully realize the potential for digital tools to improve health until we adopt an efficient framework for integrating apps and software into drug development, review, and product approval.

This starts with taking the same forward leaning approach that we’ve adopted for software that may be a medical device and make clear we apply some of the same approaches to apps that are coupled to drugs.

Today, I’m pleased to announce we’re taking a step to do just that.

We’re hitting a critical milestone on our collective journey toward advancing patient care by harnessing the power of digital tools to complement the use of prescription drugs.  

There are a variety of ways in which drug-associated apps and software can enhance patient care. But these apps and software can also pose potential risks. So we need a flexible, risk-based framework that protects patients. But a framework that also helps advance the use of digital health in the prescription drug setting as this tech improves.

The opportunities are becoming more obvious.

It includes apps that allow a physician to provide a patient-adjusted, weight-based dosing schedule and then remind patients to take their doses at the correct time. Or apps that help patients track symptoms, and share information with providers. Or that can collect safety information.

We know that when a patient can easily and efficiently record the incidence or severity of symptoms with a few swipes on a phone or tablet, it can improve their ability to communicate with a doctor, and expand the capacity of doctors to help patients manage a disease.

The capabilities are broad.

We’ve approved a drug with an embedded sensor to send a signal to communicate with a device, allowing the app to automatically record when a tablet has been ingested and help track patient compliance.

For challenging conditions like addiction, apps can help support patient adherence to the prescribed medication.  The integration of digital tech into supportive care strategies can promote near real time monitoring of patient outcomes and may improve treatment options. 

But a lot of these technologies haven’t advanced as quickly when it comes to the delivery of medicines because the regulatory pathway has – at times – seemed uncertain. Our aim is to provide that certitude.

And so to advance innovation in digital health, today, we’re establishing a docket in the Federal Register to gather feedback on a proposed framework for regulating software applications that are disseminated by drug sponsors for use with prescription drugs.

Our aim is to establish a bright line between those apps that are coupled to drugs in a way that require their review as part of the pre-market drug application. And those apps that can be safely advanced to patients without pre-market review. Instead, these lower-risk apps can be subject to post-market surveillance and monitoring.

Consider a software product used to measure physical activity that’s branded with the name of a drug that’s indicated to alleviate pain from osteoarthritis. In this case, the drug maker gives the app to patients to record their degree of physical functioning while taking the drug. That information can be used to help inform a patient’s care.

In such a scenario, the software output may be considered prescription drug use-related software, even if its functionality doesn’t meet the definition of a being medical device.  In this circumstance, the app would be covered by our newly proposed framework for how we aim to more efficiently regulate these kinds of digital products.

Under that approach, we expect that the output of most apps used with prescription drugs wouldn’t need to come to the FDA for premarket review.  Instead, in most cases, the software output—screen displays or audio messages—would be considered promotional drug labeling.

That means that the app wouldn’t require review by the FDA prior to sharing it with patients.  This creates a bright line for product developers that want to share useful apps with patients. Like other promotional labeling, copies of the output would need to be submitted to the FDA at the same time the drug sponsor disseminates the app.

With these new concepts, we hope to define a new and more efficient pathway for the development of innovative digital technologies that can help promote the safe and effective use of medicines.

We believe the flexible concepts we’ve put forward will encourage more drug sponsors to advance beneficial and innovative software apps with FDA-approved drugs and biologics. By providing greater clarity about the FDA’s role in reviewing the output of this kind of software, we can help promote the development of these tools.

We recognize that one size doesn’t fit all. Under our proposed framework not all of the FDA requirements would apply every time a digital health tool is employed in relation to a prescription drug.

We know how critical it is for developers to have transparency around our regulatory expectations, especially in rapidly advancing fields like digital health. So, the Federal Register notice makes clear that independent, third party app developers aren’t covered by this proposed framework.

This new framework only focuses on apps shared by prescription drug sponsors.  Other app developers are generally subject to the framework we’ve articulated for software as a medical device. We already have clear rules defining these circumstances. 

But we know that drug makers sometimes want to share apps along with drugs to help improve the use of those medicines. And we know that the rules have sometimes been unclear about how the FDA would treat those tools. So, the aim of our new framework is to provide more clarity and more certainty; while ensuring the safety of these products.

We’re committed to considering and weighing all comments as we develop draft guidance based on this approach. That includes comments on the examples we’ve used to illustrate when software output is considered promotional labeling; when it may be referenced in required labeling as part of a New Drug Application or Biologics License Applications submission; and when it’s an essential component of a drug-led, drug-device combination product.

As we continue to advance our framework, the goal is to harmonize the FDA’s approach to overseeing software products that are used in conjunction with prescription drugs with our agency-wide approach to digital health.  Our proposal doesn’t change our approach to digital health tools that are developed by software sponsors for use with drugs.  It only applies to drug sponsors who choose to disseminate such software for use with one or more of their own prescription drugs.

Under this proposed approach, some of the prescription drug apps distributed by or on behalf of drug sponsors that may help patients -- and wouldn’t be required to make pre-market submissions -- could include dose calculators, symptom trackers, and medication reminders.

Since we propose to treat most apps as promotional labeling, it means that only that the final display or communication the software delivers to end users – for instance, screen displays, or audio and visual messaging – would be sent to the FDA at the time of dissemination.  In these cases, sponsors would not have to send us the underlying software code.

Sponsors can also leverage a voluntary advisory comment process we have when it comes to promotional material. This way, they can receive the FDA’s feedback before distributing the proposed labeling. In this case, that labeling would be the software output that’s considered – for these circumstances – part of the drug’s promotional labeling.

For certain software output, the new, proposed framework recommends that sponsors use this voluntary comment process.  

We expect that information about prescription drug use related software output would be included in FDA required labeling—and subject to premarket review—only in the following two situations:

First, where the drug sponsor demonstrates that there’s substantial evidence of an impact on a clinically meaningful outcome as a result of the use of the software app with a drug.

Or second, when the app is key to one or more of the intended uses of a drug-led, drug device combination product.  In those cases, we’ll evaluate the safety and effectiveness of the software app as needed.

Finally, we want to provide a framework that supports both robust software innovations related to prescription drugs -- and timely access to affordable generic products. We look forward to comments on how we can achieve these outcomes when software related output is included in the FDA-required labeling for the reference product. We won’t let the existence of software app become an obstacle to timely generic entry.

Digital tools present lots of opportunities when it comes to helping patients make better use of new drugs, and better interactions with providers. But they also present lots of chances for the FDA to improve our review of new drugs, and the process for how they’re developed.

This is especially true when it comes to modernizing clinical trials. We’re going to be focusing more attention in the coming months on how we can use technology to improve the way trials are conducted, and the way information is gathered and analyzed. We believe these approaches can expand enrollment opportunities and lower development costs.

Overly long, overly complex trials can deter patient enrollment, exhaust investigators, push trials out of the community setting, and delay completion of studies so long that their findings are no longer relevant.

Overly restrictive exclusion criteria can also screen out patients with a prior history of health complications.

Travel can be another significant barrier to getting patients to participate in trials. An estimated 70 percent of Americans live more than two hours away from clinical study sites.

Financial barriers – from lost work, or inability to pay for extended childcare – can also discourage patients from participating.

As a result, only a fraction of U.S. patients – about three to five percent in the case of cancer patients -- participate in clinical trials.

This can raise concerns about how efficacy signals from traditional trials can be applied to “real world” patients who may be sicker, older, or less compliant than patients studied at academic trial sites.

We don’t use technology well in clinical trials to collect information and use it to do quality checks on the data that’s collected.

We still do a lot of things manually. And in my view, a lot of these outdated processes are perpetuated by entrenched players like contract research organizations that profit off the old ways of doing things.

Better use of digital tools for capturing and auditing information can disrupt these old legacy approaches. They can provide better oversight, lower development costs, and open up more trial sites, and more providers and patients to opportunities to participate in trials.

Right now, clinical trials are major contributors to the cost of drug development. Costly trials can discourage the development of second and third-to-market innovations. This leaves first-in-class products, particularly for rare indications, the ability to engage in monopoly-pricing power long after patents and other exclusivities have expired.

A recent study estimated average total R&D out-of-pocket cost per approved new compound at $1.395 billion in 2013 dollars.

A soon to be published FDA study found that nearly 50 percent of novel drugs approved from 1991-2000 had a competitor within two years. But it took five more years to achieve the same level of competition for drugs approved the next decade, from 2001-2010.

Novel trial designs incorporating digital tools and real world evidence can help make trials more efficient, more reflective of our diverse population, and more accessible to patients where they live and work. This can be especially true when it comes to second-to-market drugs where the existence of available therapy in a drug class can make it hard to enroll registration trials with follow on innovations.

Technology can also make it easier to disrupt the old way of conducting clinical trials. And disrupt some of the entrenched structures that make site selection, site auditing, and data collection more paper based than it ought to be, and more costly than it should be.

CDRH is working with the Medical Device Innovation Consortium (MDIC) to advance the use of real world evidence (RWE) to advance these and similar opportunities.

Building consensus standards on how best to translate real world data into real world evidence for regulatory decision-making can lead to earlier patient access to safe and effective technologies.

To help support the integration of RWE throughout the total product life cycle for medical devices, we’ve awarded a cooperative agreement to MDIC to manage the National Evaluation System for health Technology coordinating center or NESTcc.

The NESTcc has established relationships with 11 data partners, 150 hospitals, and thousands of outpatient clinics. Taken together, they represent nearly 470 million patient records. NEST’s data partners will conduct testing to assess their system’s capabilities for addressing critical RWE questions across the total product life cycle.

One important role for NEST is developing new methods for active safety surveillance. Active surveillance allows FDA’s reviewers to continuously generate, access, and evaluate large data sets. This includes patient registry data, claims data, and data from electronic health records – to monitor device performance and patient outcomes in routine clinical practice. Where applying these same approaches when it comes to drugs.

In the case of medical devices, active surveillance can allow reviewers, providers, and device manufacturers to make timelier, evidence-based decisions when a potential safety signal is detected.

The FDA has also established a unique device identification or UDI system, which can allow for the more accurate reporting and analyzing of adverse events associated with specific products. This can increase the value of real-world data for safety signal detection.

One of NEST’s active surveillance pilots is VISION – the Vascular Implants Surveillance Intervention and Outcomes Network.

Stents, stent-grafts, and other devices are commonly used to treat diseases of the peripheral circulatory system, like blocked or partially blocked arteries. This pilot is being led by Weill Cornell School of Medicine under a cooperative agreement with the FDA.

VISION aims to improve evidence generation on the safety and performance of vascular devices and procedures by linking patient registry data with state and national claims data to monitor long-term outcomes of patients treated with vascular devices.

Linking registries and state and national claims datasets will help researchers track individual patients over time. It can help validate complication rates like stroke or arterial rupture, and support analytics that can adjust patient outcomes based on patient-specific risk factors.

All of this can help us understand when a device, rather than patient’s overall health status, might be responsible for an adverse event.

NEST is also piloting a national consortium of coordinated orthopedic device registries to develop active surveillance methods for monitoring the safety of arthroplasty devices.

Total joint replacement is the fastest growing elective device-based surgery in the nation, if not in the world, with over 1.2 million hip and knee replacements performed annually in the U.S.

With the aging of the baby boomer generation, higher rates of osteoarthritis, innovative treatment options, and the growing demand for improved mobility and quality of life, orthopedic procedure volumes are projected to reach three million annually in the next two decades.

This pilot will explore how we can improve clinical evidence generation and safety evaluation for orthopedic implants in the U.S. through creation of a strategically coordinated registry network. The pilot will eventually include over 1.8 million total joint replacement procedures. These same kinds of data collection schemes are also being applied to how we gather and analyze data when it comes to drugs.

These same digital technologies also present important opportunities to streamline drug trials and improve data site integrity by using statistical tools to remotely monitor data accrual and integrity over the course of a trial.  To take one example, traditional on-site monitoring of each clinical site to evaluate study conduct and perform source data verification is highly resource intensive and accounts for up to a third of the total clinical trial cost. ,   But, traditional on-site monitoring doesn’t guarantee data quality. Findings must still be evaluated to determine their true impact, and whether additional actions are needed. 

To help ensure data quality and human subject protections, the FDA recently issued final guidance that recommends sponsors use a risk-based approach to monitor clinical investigations rather than a monitoring approach that employs 100 percent source data verification.

Risk-based monitoring focuses sponsor oversight on risks to the most critical data elements and processes necessary to achieve the objectives of clinical investigations. This is another way that the use of digital technologies can make the development process more efficient, less costly, while improving our levels of oversight. It can also help disrupt the lock that traditional CROs have on this process.

This is a lock on research that can serve to drive up costs while making it hard for community sites to participate in clinical trials.

As part of a risk-based monitoring approach, the FDA encourages the use of centralized monitoring.   Centralized monitoring is one component of a risk based monitoring approach. And it can be conducted in conjunction with targeted on-site monitoring.

The FDA has learned from stakeholders’  feedback that risk-based approaches to monitoring -- including employing centralized monitoring techniques -- have not been fully implemented.

Challenges to implementing risk-based monitoring raised by stakeholders also include the belief that if 100 percent source data verification isn’t used, there’s a greater risk that an issue will be identified by FDA which could lead to an adverse regulatory action.

And stakeholders may have uncertainty about the best technological solutions for carrying out centralized monitoring, including challenges with integrating these systems into their operational workflows.  For instance, what would be an appropriate threshold for key patient risk indicators that would warrant a change in monitoring?

We recognize that these are important questions and concerns that we need to be address. And we’ll look at ways to address these challenges more in 2019. We want to better understand the extent to which risk based monitoring has been implemented, the ongoing barriers to full execution, and how the FDA can continue to advance implementation of risk based monitoring -- especially with centralized monitoring.

Building technical and organizational capacity for risk based monitoring and centralized monitoring can help stakeholders and the FDA prepare for the conduct of decentralized and hybrid clinical trials.

This includes trial designs where some tests or procedures could be conducted in a home setting. Take, for instance, a blood draw by a phlebotomist without requiring patients to travel to a central study site.
Additional routine data collection or testing for trials could also be facilitated through EHRs, wearable devices, or smartphones and electronic patient reported outcomes surveys (ePROs). 
Efficient, reliable data capture in an interoperable environment could facilitate pragmatic trials conducted at the point-of-care to satisfy post market study requirements. It can develop better natural histories of disease and response to treatment in diverse care settings. And it can help regulators, sponsors, and clinicians better answer unresolved questions about products long term safety and efficacy.

These are all ways that better capture of data using digital tools and better use of digital tools to audit the accuracy of that information can help lower development costs and expand opportunities.

This is how technology can help disrupt entrenched models for doing things in a way that expands opportunities and creates new efficiencies.

Clinical trials will continue to incorporate more of these technologies, allowing researchers to reduce the time it takes generate reliable insights from the clinic to better inform efficient product development.

FDA reviewers will also harness more of these tools to improve how we collect information more efficiently, and analyze it more reliably, to take more timely, targeted action to mitigate serious health risks.

We’ll have more to say soon on how we plan to allow sponsors to make better use of cloud-based analytical tools in the clinical trial and clinical review process. These new steps we’re working on will help advance the use of digital technologies as a tool for improving the capture and evaluation of data as part of the review process. This is a major part of our effort to modernize the new drug review process.

And we look forward to working with the Reagan Udall Foundation to help seize all of these opportunities -- and deliver the full promise of technology and science to improve lives and advance the public health.

Thank you.   

Back to Top