The passage of the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) represents the most significant change in health care payment policy since the advent of Medicare. There is broad congressional and policy consensus that the present system of fee-for-service payment must transition to a value-based system measuring quality and cost. Upon implementation of MACRA, ophthalmologists and retinal specialists in particular will experience unprecedented changes in reimbursement and the delivery of care. Success in this new model is predicated on data. Retinal specialists will need real-world, real-time data to determine their quality and cost of care. At present, a Qualified Clinical Data Registry (QCDR) such as the IRIS Registry appears to be the most likely mechanism for successful participation.

In 2017, CMS will implement the Quality Payment Program involving two paths: the Merit-based Incentive Payment System (MIPS) and the Advanced Alternative Payment Models (APM). Although at the time of this writing the rules are not yet finalized, it is clear that the vast majority of retina specialists will take the MIPS path. Participation in a QCDR such as IRIS is likely to be the best mechanism for MIPS reporting. A QCDR will facilitate reporting in the four MIPS categories of quality (replacing PQRS and the quality component of the value modifier), advancing care information (replaces EHR Meaningful Use), clinical practice improvement activities, and resource use.

As of April 2016, IRIS had data from 7866 ophthalmologists in 1704 practices comprising 88 million office visits from 24.5 million unique patients. It is projected that by January 2017 there will be data from 130 million office visits and that by January 2018 there will be 40 million unique patients. These data comprise over 350 measures taken directly from the EHR of enrolled ophthalmologists, providing an unprecedented opportunity for analysis of clinical outcomes, practice and billing patterns, and treatment trends.

As just one example, at the end of 2015, IRIS had comprehensive data including disease state, drugs, pre- and post- treatment vision, comorbidities, and complications on 4,572,677 intravitreal injections from 611,818 patients. This is truly Big Data. A detailed analysis of this dataset will be presented, with an emphasis on implications for health policy. Initial findings demonstrate that 47% of injections were bevacizumab, 25% were ranibizumab (either dose), and 28% were aflibercept. Interestingly, these proportions change by disease; bevacizumab comprises 42% of all injections for AMD and 62% for diabetic retinopathy. For AMD, over 20% of patients are treated with two or more drugs, demonstrating the need for patients and ophthalmologists to have access to all available proven therapies.

Data analysis on this scale is a game changer that will facilitate our quest in the challenge to provide high-value retinal care.