How Insurance, Socioeconomic Barriers Derail DME Care: Sunir J. Garg, MD

Insurance step protocols are forcing retina specialists to begin treatment for diabetic macular edema (DME) with agents known to be less effective for patients with moderate or worse vision loss—while prior authorization delays compound the damage by deferring care that most patients will eventually receive anyway, according to Sunir J. Garg, MD, FACS, FASRS, professor of ophthalmology at Wills Eye Hospital and Thomas Jefferson University, and partner at Mid Atlantic Retina.

The problem is not limited to payer policy. Racial, ethnic, and socioeconomic disparities in DME outcomes are rooted in neighborhood-level conditions, such as food insecurity, variable Medicaid coverage, and lack of direct transit routes, that create friction long before a patient reaches the retina clinic. For hourly workers without flex time, a single missed appointment can cascade into a pattern of undertreated disease that erodes vision over years.

In an interview with The American Journal of Managed Care^® (AJMC^®), Garg described how payer-driven constraints shape his treatment decisions, where disparities are most pronounced along the care pathway, how his high-volume practice manages the operational demands of sustained DME care, and what single change would most meaningfully move the needle on long-term outcomes.

In part 1 of the interview, Garg discussed how to build adherence support into the first visit and how to keep working-age adults engaged in a long-term treatment regimen.

This interview has been lightly edited for clarity.

AJMC: How much of your treatment selection today is driven by efficacy data vs what you think a specific patient will actually be able to sustain over the long term?

Sunir Garg, MD: It's a mixture. A lot of our decision-making nowadays, unfortunately, is dictated by insurance companies and step protocols, which are quite common in our area and affect a lot of retina specialists nationally. For some of our patients, we know we have to start with an agent like bevacizumab, which, for patients with moderate vision loss from DME or worse, we know isn't as good as other agents. Trying to explain to the patient that the next few injections might be spinning our wheels can be frustrating.

Part of it is also driven by co-pay. For patients who have a 20% co-pay, that matters, and with foundation support becoming variably available, that's definitely an issue.

For other patients, it's really great to have drugs that work well and keep them good with less frequent treatments. We know from earlier work that aflibercept 2 mg works great for patients with diabetes. We have aflibercept 8 mg, which I use a lot nowadays, because it's a drug I'm very familiar with, and it hopefully lets me space patients’ treatments out a bit more,¹ which is really valuable. Faricimab is another agent we've been using with good success. I’ll start with different drugs depending on what the next option in line is. If that gives me the results I want in a timely fashion, great. But if not, I will rapidly move to one of my newer agents, simply because I don't want a patient coming in every 4 to 6 weeks for a long period of time when I know I can do things that will often get their OCTs [optical coherence tomography] dry quicker and hopefully keep them there longer.

AJMC: DME outcomes data consistently show worse results along racial, ethnic, and socioeconomic lines. What have you found are the specific points in the care pathway where those disparities are most pronounced?

Sunir Garg, MD: It's a really complex issue, and there are a lot of them. Often it starts with where patients live. In Philadelphia, we have very affluent neighborhoods right next to areas where a substantial percentage of adults and kids are below the poverty line—sometimes within a mile of each other. There are pockets of Philadelphia, generally in minority communities, where there's no access to grocery stores, or people don't have the money to buy nutritious food, or they don't have a working refrigerator. There are a bunch of barriers to a healthy diet, which we know matters for all of our patients.

Many of these patients also have variable health insurance. Some Medicaid products aren't taken in many places, or the wait times are very long, or access to an endocrinologist to help optimize their health is limited. Then there are co-pay issues with drugs, and even if you get in to see the doctor, the appropriate drug for you may or may not be covered by your plan. A lot of times when these patients come into our office, they're hourly workers with pressure to be at work for their shift. They don't have the luxury of flex time. They have to take the bus a lot of times, because maybe they don't have access to a car, and sometimes there is no direct route. Any or all of those things can create points of stress in getting the care they need.

Some of those we can help with; some are bigger societal concerns. But often we can figure out ways to get them treated. Being mindful of the additional stressors they may have and perhaps advocate with their insurance company to get them on drugs that might work better and give them a longer interval between treatments is helpful.

AJMC: How are prior authorization and payer restrictions concretely shaping treatment decisions, and in what situations do you feel most constrained by them?

Sunir Garg, MD: They affect us quite a bit and in a bunch of different ways. First, we have to have a lot of personnel dedicated just to managing all of that. In a large group like ours, we have employees whose only job is to help with prior authorizations, and many of our employees have to navigate these waters daily.

Second, some of the authorization pathways are onerous. We're in a large group with a large catchment area, and some insurance carriers require that I be the only doctor who can treat a particular patient on a given day with the drug I chose. If that patient got sick and had to reschedule and come in next week to see one of my excellent colleagues, the insurance company may not provide authorization in real time for a different physician. The patient gets deferred again for another day.

There are other situations where we want to get a patient on a particular drug because the first drug we were required to start with didn't work well, and then the prior authorization didn't come through in time, or the patient showed up to our office and we couldn't get the authorization that day, so the patient has to go home and come back another day. It's a huge problem.

Studies done by the American Society of Retina Specialists and other organizations show that most of the time, patients end up getting the drug they were going to get anyway.² The prior authorization just ends up being an additional burden on physicians and patients that doesn't ultimately change what treatment patients receive. It just delays care. It's an ongoing struggle, and as a profession, it behooves us to advocate on behalf of our patients. It continues to be a big frustration and burden on all of us.

AJMC: Running a high-volume retina practice that serves patients who need frequent visits over years requires real operational infrastructure. What does your practice do to make sustained DME management work, and where do the systems still fall short?

Sunir Garg, MD: It's been a work in progress over the past 20 years since anti-VEGF [vascular endothelial growth factor] agents became available to make a system that works for our practice, and the system that we have works well for our group but may or may not work well for other groups. Having an efficient practice is important. Having airtight systems to manage drug inventory, prior authorizations, reimbursement, and collections is critical, because these drugs are expensive and seem to be getting more expensive each year. Being able to manage all of that and protect the financial health of the practice matters.

On the scheduling side, a lot of times when patients come in for an injection today, if I know I'm going to see them back in, say, 6 weeks, I'll schedule them for the next 2 appointments while they're still with me. That tends to reduce the long gaps where people fall off the radar entirely, because even if they miss the 6-week appointment, our reminder system will generally kick in for the appointment after that. It's been a helpful safety net.

We also have office staff who physically call patients who didn't return for their time-sensitive treatments, and that's helped as well. There's growing interest in using more automated artificial intelligence [AI] algorithms to flag patients more rapidly if they're missing visits. That’s something we haven't yet implemented but are interested in exploring.

AJMC: If you could change one thing that would most meaningfully improve long-term outcomes for patients with DME, what would it be?

Sunir Garg, MD: I would love to be in a position where I can give patients the agents I think would work best for them without all the other barriers in the way, whether it's co-pay barriers, prior authorization barriers, or spinning our wheels with onerous step protocols. We have good data that some of our drugs are definitely better than others in getting patients dry and keeping them dry. We know that translates into better visual outcomes over time, and we know that some of our newer agents allow patients to maintain those good visual outcomes while coming in less frequently. Being able to have that flexibility—to treat the patient in front of me, given the totality of what's happening in their eye and in their life—would be a huge advantage.

References

1. Harp MD. FDA approves extended dosing intervals up to 20 weeks for EYLEA HD in wAMD and DME. Ophthalmology Times^®. April 3, 2026. Accessed June 16, 2026. https://www.ophthalmologytimes.com/view/fda-approves-extended-dosing-intervals-up-to-20-weeks-for-eylea-hd-in-wamd-and-dme

2. Dang S, Parke DW, Sodhi GS, et al. Anti-VEGF pharmaceutical prior authorization in retina practices. JAMA Ophthalmol. 2024;142(8):716-721. doi:10.1001/jamaophthalmol.2024.2217