Developing a Learning Healthcare System

When do you know all there is to know about something? Do you know everything about your spouse after being married 10 years? Or 15 years? Or ever?

When the US Food and Drug Administration (FDA) approves a new drug, most people incorrectly believe that all benefits and risks are knowable. Because clinical trials are controlled to avoid confounding variables, we may not always know how the medicine will react with other medications used in real-world settings—especially by people with multiple comorbidities. Throughout the life cycle of the drug, additional data about the product’s performance accumulate, and new information regarding the performance and safety profi le is learned.1

Phase III clinical trials to test new drugs typically involve large groups of people, from 100 to several thousand. ² These participants can’t compare to the breadth or variation of the patient population for which the medication is eventually prescribed. Some important safety concerns come to light after the drug is marketed to a wider audience. Also, most treatments work for only 50 to 75 percent of the patients who use them.³

Various entities are trying to take the vast amount of post-market data in existence and turn it into usable information that will expand our knowledge base.

Such knowledge will bring us closer than ever to the promise of personalized healthcare, the delivery of care most appropriate for individual patient needs and conditions. We also know the marriage of health research with real-world application leads to improved health outcomes and will help us curb costs responsibly.⁴

The Institute of Medicine has convened a round table of experts seeking a way to develop a “learning health system that is designed to generate and apply the best evidence for the collaborative health care choices of each patient and provider.”⁵ Such a system would drive the process of discovery as a natural outgrowth of patient care and would ensure innovation, quality, safety, and value in healthcare.

In short, it would be a system that encourages learning throughout the care continuum. It would be a system that involves all stakeholders—from the researcher to the provider, from the patient to the payer.

Two initiatives currently under way that will add to the timely generation of evidence and support the generation of new insight into the safety of medicines and treatment effectiveness are the FDA’s Sentinel Initiative and the Patient-Centered Outcomes Research Institute’s (PCORI’s) advancement of high-integrity, evidence-based information.

FDA Sentinel Initiative

The FDA has used administrative and insurance claims databases to investigate safety questions about regulated products, but not collectively. The agency’s goal now is to create a linked, sustainable system that will draw on existing automated healthcare data from multiple sources. The Sentinel Initiative will eventually give the FDA the ability to actively monitor the safety of medical products continuously and in real time.⁶

Launched in May 2008 by the FDA, the Sentinel Initiative aims to develop and implement a proactive system that will complement existing data collection mechanisms to track reports of adverse drug events.⁷ This initiative will augment, not replace, the FDA’s existing safety assessment tools, such as the Adverse Event Reporting System (AERS) and MedWatch.

To help guide the development of Sentinel, the Mini-Sentinel pilot was created. This database comprises quality-checked data held by 17 partner organizations, including entities such as academic medical centers and healthcare systems with electronic health records, and health insurance companies with administrative claims data.⁸

One goal of the Mini-Sentinel was to have the ability to query electronic health data from 100 million people by July 2012. As of December 2011, Mini-Sentinel had already surpassed this goal and could query administrative claims data from more than 126 million individuals, more than one-third of the US population.¹

Using Mini-Sentinel, scientists can get responses to their questions in a matter of weeks, compared with months or even longer using traditional surveillance methods.⁸ Twenty reports regarding identifi cation of health outcomes have already been prepared for the FDA by Mini-Sentinel collaborators as part of this pilot.⁹

For patients, this project will mean their personal reporting data will not be for naught. Their healthcare data will become part of a comprehensive system with appropriate safeguards to protect the privacy of their personal information.

Patient-Centered Outcomes Research Institute

The Patient-Centered Outcomes Research Institute (PCORI) is an independent organization created by the Affordable Care Act signed into law in 2010. Its primary goals are to help people make informed healthcare decisions and to improve the delivery of healthcare. The Institute will commission research that is guided by patients, family caregivers, and the broader healthcare community in order to produce high-integrity, evidence-based information.¹⁰

The research will be designed to inform healthcare decisions by providing evidence on the effectiveness, benef ts, and harms of different treatment options for different patients. PCORI’s first call for pilot research projects generated more than 850 submissions. The Institute is expected to announce its fi rst round of funding in May of this year.

The Institute has proposed 5 national priority areas for patient-centered comparative clinical effectiveness research. One is the “assessment of prevention, diagnosis, and treatment options.” This research goal will focus on determining which options work best for distinct populations with specifi c health problems. Another PCORI-proposed priority is the acceleration of patient centered and methodological research. That means patients and family caregivers will be included in the design of research that is quick, safe, and effi cient.¹¹

It is exciting to see how quickly PCORI has developed its operational plan, hired staff, and engaged the patient community. Understanding how the initiatives can add to knowledge is still a process in motion. Helping patients understand their role in this learning health system is still in its infancy. It is encouraging that both the Sentinel Initiative and PCORI are committed to finding better ways to provide usable information to patients and providers so they in turn can make informed healthcare decisions.¹¹

As the Sentinel Initiative and PCORI progress, we must continue to focus on the need to present the results in a way that is “available, accessible, transferable, usable, and manipulable” in order for people with chronic conditions to make the best use of this learning health system.⁵ Developing effective strategies to communicate the fi ndings and their limitations to patients will be essential in order for this information to be turned into knowledge.

As First Lady Abigail Adams wrote, “Learning is not attained by chance. It must be sought for with ardor and attended to with diligence.”