Last October I spent 4 days waiting for my insurer to approve a continuous glucose monitor refill. Four days. I track my glucose response to every meal I eat. I have 14 months of data showing how this device helped me drop my fasting glucose from 98 to 84 and raise my HRV from 52 to 68. My endocrinologist wrote the prescription. The insurer's algorithm said no, then said maybe, then said yes after my doctor's office spent 45 minutes on hold. I am a healthy person with good insurance who knows how to work the system. Prior authorization reform is not enough. The whole mechanism needs to go.
I realize that sounds extreme. Maya Okafor would say I am projecting my optimization obsession onto a policy question that affects 330 million people. Fair. But the data is not just my data. Physicians average 39 prior authorizations per week, burning 13 hours on the process. In Medicare Advantage alone, 7.7% of requests were denied in 2024. Over 80% of those denials were overturned on appeal. Read that ratio again and tell me the system is functioning as a cost-containment tool. It is functioning as a friction generator that bets most patients will give up. And most do: only 1 in 10 denied requests were appealed in 2022.
The ROI on Delay Is Negative
I think about health interventions the way I think about compound interest. Every day of delay costs you returns. A CGM sitting in a warehouse while paperwork clears is a day of glucose data you do not get back. Scale that logic to a patient with inflammatory bowel disease waiting for a biologic. The ACG found that 70% of IBD patients face insurance barriers to first-line medications. Their survey showed 83% of GI clinicians linked prior authorization delays directly to hospitalizations.
Hospitalizations are expensive. The prior authorization that supposedly saved the insurer money on a $2,000 monthly biologic just generated a $15,000 ER visit. The system's own economics do not hold up under scrutiny.
Indiana Governor Mike Braun got this right: "If your doctor says you need it, you'll get it." Critics call that naive. I call it a minimum viable product for a healthcare system that currently ships broken software and promises a patch next quarter.
Reform Timelines Are the Problem
The CMS Interoperability and Prior Authorization Final Rule took effect in January 2026. Good. Standardized electronic prior authorization infrastructure? Targeted for January 2027. The Improving Seniors' Timely Access to Care Act has 248 House co-sponsors and 64 Senate co-sponsors, a supermajority, and it still has not passed. Health plans serving 270 million Americans made voluntary simplification commitments in 2025. Voluntary.
I will grant the reform camp one thing: the March 31 transparency mandate requiring payers to report denial rates and turnaround times is genuinely useful. Sunlight matters. But transparency without teeth is a dashboard nobody checks. I have a dashboard for my sleep scores. It did not fix my sleep. Changing my evening protocol did.
The pattern here is familiar to anyone who has tried to optimize a broken system from within. You add monitoring. You add reporting. You add APIs. You push the real deadline out another year. Meanwhile, the underlying incentive structure stays intact: insurers profit from denials because most patients never appeal, and the cost of processing appeals falls on physician offices already spending $26.7 billion annually on this bureaucracy.
AMA Chair David Aizuss put it plainly: "When I write a prescription, I consider it a recommendation because I don't know what's going to be filled." A physician's prescription is a recommendation. That sentence should make your stomach turn.
I am not arguing for zero oversight. I am arguing that prior authorization, as a category of intervention, has failed its own success metrics so thoroughly that incremental reform amounts to optimizing a protocol that never worked. You do not keep tweaking a supplement stack that tanks your bloodwork for 3 straight quarters. You stop taking it. You try something else. Senator Whitehouse's approach, exempting providers who bear financial risk for outcomes, points toward a better architecture: align incentives instead of adding gatekeepers.
My CGM arrived on day 5. I lost 4 days of data I will never recover. Somewhere a patient with a serious diagnosis lost something they cannot get back either, and the system that delayed them is promising to work better by 2027.
