The insulin cliff: what happens to diabetes patients when Medicaid disappears

Consider a patient at a community health center somewhere in the United States. She has type 2 diabetes, manages it with metformin, and sees her doctor every few months. Her hemoglobin A1c - the three-month average of blood sugar levels - is reasonably controlled. She has Medicaid, which covers her medications and office visits.

Then she loses that coverage. Maybe she missed a renewal form. Maybe her state changed eligibility rules. Maybe she got a part-time job that pushed her income just above the threshold but did not offer insurance. The reason varies. The result does not.

Within months, her A1c climbs. Her doctor adds insulin to her regimen - a more intensive, more expensive, more logistically demanding medication. She cannot reliably afford it. Her disease, once stable, begins to slide.

That trajectory is not hypothetical. It is what researchers at Oregon Health and Science University found when they tracked more than 39,000 adults with diabetes across community health centers in 20 states. The study, published March 20 in JAMA Health Forum, provides some of the clearest evidence yet that losing health insurance does not merely create a paperwork problem - it creates a medical one.

Matched at the starting line, diverging fast

The study's design is central to its strength. Lead author Nathalie Huguet, an associate professor of family medicine at OHSU, and her team did not simply compare insured and uninsured patients, which would have introduced confounding from baseline health differences. Instead, they identified patients who experienced insurance churn - defined as two or more consecutive uninsured visits, indicating sustained coverage loss - and matched them with patients who were virtually identical at baseline but maintained continuous coverage.

Same starting A1c levels. Same medication profiles. Same demographic and clinical characteristics. The only difference was that one group lost insurance and the other did not.

The divergence appeared quickly. Patients who churned out of coverage showed worsening glycemic control. They were prescribed insulin and other high-intensity diabetes medications at higher rates. They experienced more acute diabetes complications during the study period.

More medication, less ability to afford it

The increase in insulin prescriptions after coverage loss creates a painful contradiction. Insulin is the medication most likely to be prescribed when oral drugs are no longer sufficient to control blood sugar. But insulin is also expensive - list prices for some formulations run into hundreds of dollars per month - and it requires supplies (syringes or pen needles, glucose monitors, test strips) that add further costs.

Community health centers offer discounted medications on sliding-fee scales, and some patients may have accessed insulin at reduced cost. But the gap between being prescribed insulin and consistently using it is enormous for patients without coverage. Insulin requires refrigeration, regular dose adjustments, and blood glucose monitoring. It is not the kind of medication you can take inconsistently without consequences.

The study did not directly measure whether patients filled or adhered to their prescriptions. But the combination of more intensive medication orders and worsening blood sugar control suggests that the prescription was not translating into effective management.

Acute complications as early signals

Patients who lost coverage also experienced more acute diabetes complications - episodes like diabetic ketoacidosis and severe hypoglycemia that require urgent medical attention. These are not the long-term sequelae of poorly managed diabetes (amputations, kidney failure, blindness), which develop over years. They are the immediate consequences of blood sugar swinging out of control.

Huguet was careful to note that the absence of significantly higher rates of long-term complications during the study period should not be interpreted as reassurance. Those complications take years to develop. What the data capture is the beginning of a trajectory - a patient whose disease is becoming harder to manage, whose treatment is intensifying, and whose risk of serious harm is rising.

The implication is that longer follow-up would likely reveal worse outcomes: more amputations, more dialysis, more hospitalizations. The current data show the leading edge of that curve.

Timing and the Medicaid reckoning

The study arrives at a particular moment in American healthcare. The pandemic-era continuous enrollment provision, which prevented states from disenrolling Medicaid recipients regardless of eligibility changes, has expired. States have been conducting redeterminations, and millions of people have lost Medicaid coverage - many for procedural reasons like failing to return a renewal form, rather than because they no longer qualify.

This study builds directly on Huguet's earlier research showing that patients with diabetes are disproportionately likely to experience insurance instability. The two findings form a chain: diabetes makes you more likely to lose coverage, and losing coverage makes your diabetes worse. It is a feedback loop with no natural stopping point.

Rising insurance premiums compound the problem. For patients who lose Medicaid, marketplace plans are often unaffordable. Many face months or years without coverage - a gap during which their disease can deteriorate in ways that are costly to reverse and, in some cases, irreversible.

The safety net under strain

Community health centers serve as the primary care backbone for low-income Americans. They provide care regardless of patients' ability to pay, using sliding-fee scales and federal funding to bridge the gap. But they rely heavily on Medicaid reimbursement to sustain their operations.

If Medicaid enrollment shrinks, these clinics face a compounding problem: more patients without coverage needing more intensive care, with less revenue to pay for it. Jennifer DeVoe, a professor of family medicine at OHSU and study co-author, pointed directly at this bind. Clinics need more support, not less, to keep the damage of insurance loss to a minimum.

Without that support, patients are more likely to end up in emergency departments - a more expensive setting for the health system and a worse one for the patient. Emergency care can stabilize a diabetes crisis, but it cannot manage a chronic disease. The long-term costs of emergency management far exceed those of continuous primary care.

What the study leaves uncertain

The study has clear limitations. It is observational, and while the matched design strengthens the case for a causal link between coverage loss and worse outcomes, it cannot eliminate all confounding. Patients who lose insurance may also be experiencing other disruptions - job loss, housing instability, food insecurity - that independently affect diabetes management.

The definition of churn as two or more consecutive uninsured visits may miss shorter coverage gaps, and the study cannot distinguish between loss of Medicaid, employer insurance, or other types of coverage. The community health center population is predominantly low-income and may not represent the experience of higher-income patients who lose coverage.

The follow-up period limits the ability to assess long-term consequences. And the study cannot speak to whether restoring coverage reverses the damage or whether the deterioration during the uninsured period leaves a lasting imprint on disease trajectory.

Still, the core finding aligns with prior evidence and clinical intuition. Managing diabetes requires consistent access to medications, monitoring, and clinical guidance. Remove any of those elements, and the disease becomes harder to control. Insurance is the mechanism that provides access. Its absence removes it.