By Fred Gebhart
Are compounded injections of anti-VEGF biologics for wet macular degeneration more likely to lead to endophthalmitis compared to injections of similar agents manufactured specifically for intravitreal injection?
Testimony before a FDA committee considering the potential scope of regulation of compounded drugs for injection said no compounded product should ever be injected into the eye because of an elevated risk of infection. George Williams, MD, Chair of Ophthalmology at Oakland University William Beaumont School of Medicine asked if that assertion could be confirmed by clinical data.
“Dr. Williams asked the question on Monday,” said William L. Rich, III, MD President-Elect and Medical Director of Health Policy for the American Academy of Ophthalmology. “By Wednesday morning, 30 hours later, we had data on 836,000 consecutive injections into the eye.
“The difference in infection rates, highest to lowest, was 0.03 percent. The endophthalmitis rate was 0.12 percent for Avastin, 0.09 percent for Lucentis and 0.12 percent for Eylea. That’s the power of big data.”
Clinical research in real time is just one of the benefits of IRIS, the nation’s first EHR-based comprehensive registry of eye diseases and conditions. As of September 15, 2015, IRIS had contracted with 10,180 ophthalmologists from 3,555 practices. The data set included 15.6 million unique patients and 51.4 million visits, making IRIS the largest clinical registry in the world.
“Having an answer on the rates of infection associated with the three anti-VEGF agents in clinical use in the United States in 30 hours demonstrates the power of big data in medicine,” Dr. Rich said. “We see those kinds of results every week. In cardiology, clinical registries run by the American Society of Thoracic Surgeons and the American College of Cardiology are the single largest source of peer reviewed literature, more than 400 papers last year alone. We expect to see a similar transformation in ophthalmology.”
Most ophthalmologists recognize the value of participation in a clinical registry. The 10,180 physicians in IRIS represent more than 90 percent of U.S. ophthalmologists using EHRs. When IRIS was opened for registration in April, 2014, the modest goal was to enroll 2,200 physicians by 2017.
“We blew through our 2017 goal in less than six months,” Dr. Rich said. “Response has been so overwhelming that we had to shut down registration for a short period in 2015. Ophthalmologists clearly understand the value of clinical registries and big data.”
Big data is a blanket term for any collection of data sets so large and so complex they cannot easily be processed using traditional data processing tools. Big data typically includes unstructured data from multiple sources, collected in real time, from random sources.
The concept of big data came to public attention in the aftermath of 9/11. The National Security Agency, FBI and CIA were unable to identify the hijackers, but a data mining company unraveled the network in three days using sources such as pizza deliveries and rental records.
“Big data in medicine is not in the same league as the National Security Agency or Amazon,” Dr. Rich said. “But the results are no less transformative. Big medical data will shorten the timeline between science and the adoption of best practices. It is already answering very practical questions like differences in infection rates between different treatments. It provides quality data required for the PQRS value-based modifier, which provides immediate financial benefits. It will ease the maintenance of certification process in the near future. And it does all this at no cost to AAO members and no additional work for physicians or their office staff.”
Early registries required laborious paper entries. Web portals had eased the administrative burden to a degree by the early 1990s, but staff time remained a disincentive. The medical data world changed in 2010 when the American College of Cardiology introduced the Pinnacle Data Registry, that extracts data automatically from electronic health records. AAO is one of ten medical societies that uses a similar system to extract data directly from EHRs.
“We have seen outcome and performance improve continuously from thoracic surgeons, the ACC, the European cataract registry,” Dr. Rich said. “IRIS gives us a chance to improve our own performance and outcomes, which benefits patients and improves their satisfaction. And it avoids all of the penalties that society is imposing on physicians who don’t look at their quality and their outcomes.”
Nearly two decades of experience has yet to demonstrate a down side to registry participation, he continued. IRIS data are not discoverable during legal actions. Nor can IRIS be used by insurers, practices or competitors to compare outcomes. And unlike health insurers and retailers, IRIS has never been hacked.
“Ophthalmology has always been an innovator,” Dr. Rich said. “The first randomized clinical trial was created by ophthalmologists, ophthalmologists created the gold standard in comparative effectiveness trials. And now we have the world’s largest clinical registry. Ophthalmologists understand the power of big data to transform their lives and their patients’ lives for the better.”