The numbers are in, and they aren’t great for some of the sickest people in the country.

On Monday, the Congressional Budget Office scored the American Health Care Act, projecting that the new Republican health reform would make dramatic changes to the American health-care landscape. On the one hand, the law would cut the deficit by over $300 billion over 10 years and reduce average premiums after an initial two-year increase. On the other, the office projected that 24 million additional people would either lose health insurance or choose to go without it in the next decade.

Proportionally, the group of people that would see the most coverage losses under the AHCA is the population of people aged 50 and older. Although they’re more likely to have coverage in the first place, owing to more stable employment and a higher likelihood of public-insurance coverage, CBO estimates show the uninsured rate of people over 50 would skyrocket from around 13 percent currently to just under 30 percent by 2026. Included in that number are near-elderly adults with low incomes, a group that’s most likely among all non-Medicare enrollees to be sick, and one of the most expensive groups to cover in American health markets today.

Essentially, the AHCA works by abandoning a key goal of health reform altogether, and carving out some of the most expensive people to cover—while simply not offering them a more affordable coverage option. The CBO report suggests that while the Republican plan includes an age-adjusted tax credit to purchase health insurance and a fund for establishing reinsurance and cost-sharing programs for states’ riskiest potential enrollees, for low-income near-elderly people those cost-reductions would not be enough to offset the elimination of the Medicaid expansion, the establishment of per-capita caps on Medicaid, the loss of the ACA’s cost-and-income-based premium tax credits, and a new age-rating scheme that allows insurers to charge older people much more for coverage relative to younger people.  

Table 4 of the CBO report shows just how this dynamic plays out:

Congressional Budget Office

Currently, a 64 year old person who makes 175 percent of the federal poverty line ($26,500—about a $13 per hour wage in a full-time job) but doesn’t receive insurance from an employer or public coverage pays around $1,700 per year on average in the exchanges for a health-insurance plan. But that person’s actual premiums are much higher because of age, although Obamacare hides that amount with a premium tax credit that adjusts with cost. Obamacare’s cost-sharing subsidies paid directly to insurers increase the average actuarial value of their coverage to 87 percent—meaning the plan covers 87 percent of health-care costs. Even under Obamacare, that final amount of $1,700 is steep for a person who makes less than the $28,000 average cost of living in the United States.

Under the AHCA, the new age-rating structure actually increases the total premium, while substantially decreasing the amount of federal assistance for paying that premium. It also eliminates Obamacare’s cost-sharing reductions and allows exchange plans to cover fewer services, meaning the same person under AHCA rules will have to pay almost nine times as much for a health-insurance plan that pays for only about two-thirds of health-care services. Put a different way, this hypothetical person would have to pay 55 percent of their pre-tax income just to have inferior insurance. It’s pretty safe to say that nobody in this situation in real life would actually be able to afford health care.

The specific circumstances of 64-year-olds in the exchanges may seem unrepresentative, but note the table shows premium increases for 40-year-olds as well. After age 40, able-bodied people unlucky enough to not have employer coverage will face a steadily worsening affordability curve. Also, as the AHCA defunds enhanced federal rates for the Medicaid expansion and incentivizes states to reduce Medicaid coverage and services, more and more near-elderly people who are close to, or below, the poverty line will lose public-coverage guarantees, and will be forced to choose between paying half or more of their yearly incomes or going without.

As my colleague Olga Khazan notes, even near-elderly people who have employer-based coverage are at risk, because without an employer mandate and with exchange tax credits, some employers will drop their insurance plans. And of course the employers with the greatest incentive to drop coverage will be those with the most expensive employees to cover—including near-elderly low-wage workers.

Republicans behind the AHCA have known about these challenges for a while, and have done their best to spin them as positives. House Speaker Paul Ryan has defended the plan against criticism of the CBO’s estimates of coverage losses, saying that “this report confirms that the American Health Care Act will lower premiums and improve access to quality, affordable care. CBO also finds that this legislation will provide massive tax relief, dramatically reduce the deficit, and make the most fundamental entitlement reform in more than a generation.” Ryan acknowledges the concerns about coverage losses, but recasts the losses as an expansion of choices that reduces cost.

It’s tempting to buy Ryan’s optimistic spin, or temper it by writing off huge premium spikes as uncharacteristic bits of collateral damage for an otherwise useful plan to control costs.The truth of the CBO report, however, is that the entire mechanism of the AHCA’s savings and premium reductions for healthier and wealthier people lies in making coverage too expensive for older, sicker people with less money. Most of the budgetary savings come from cuts to Medicaid funding and the replacement of the Obamacare premium tax credit with a much less generous credit, both of which will lead to less coverage among this older population. And most of the out-of-pocket savings for people who can afford coverage by 2026 rely on expensive enrollees just disappearing from health-care markets. Attrition among near-elderly people with low incomes isn’t an ancillary effect of Republican policy here; rather it is the engine that makes the rest of the bill work.

The irony is that the mid-aughts health-reform momentum that gave the country the Affordable Care Act put the plight of near-elderly people at its center. For decades, researchers have noted that one of the key turning points in a person’s health-care experience is when they turn 65 and are guaranteed Medicare coverage. For many people, that age arrives as an oasis in a health-care desert. In 2004, almost a quarter of all people between 55 and 64 who made less than 200 percent of the federal poverty line were uninsured, a statistic that led Urban Institute and Kaiser Family Foundation researchers to reach the following conclusion 13 years ago:

Extending coverage to this group could be done either through providing access to Medicare, or creating a new group purchasing arrangement. Expanding Medicaid to even subgroups of the near elderly seems less feasible at this time given that the near elderly are a high-cost population and states are struggling to maintain their programs’ coverage in the face of unparalleled fiscal crises. Alternatively, tax credits for the purchase of non-group health insurance are unlikely to help many of the near elderly unless the value of the credit is substantially adjusted for age or health status.

Without employing some of the options outlined in that report—the Affordable Care Act focused on Medicaid expansion and an adjustable tax credit—the AHCA will likely increase mortality among near-elderly people. Researchers have found that people between 55 and 64 who are uninsured are significantly more likely to die than their insured peers. That much should be obvious, as 65 is a somewhat arbitrary age for health purposes, and often the advance of age-linked disorders comes earlier. The median age of several cancer diagnoses—especially for people of color—often comes in the near-elderly range, and heart disease, diabetes, and hypertension often first manifest themselves during this range. And as the opioid epidemic grips the country, the risk of death from overdose is increasingly rapidly among the near-elderly.

Again, those risks are not simply collateral damage or unintended consequences of the AHCA; they are the ultimate endpoint of all its constituent policies. Republicans could fix these consequences, but that might mean risking some of the more positive reports of effects on the budget and premiums, and in essence really creating the “Obamacare lite” kind of program so dreaded around the Capitol. And if whispers of Republican plans to raise the Medicare age or reform that program have any amount of truth to them, the AHCA could land as the first blow of a one-two punch. It would deeply weaken health-care coverage and health for near-elderly people even as future policies delay the promise of guaranteed care past the age of 65. The end result of that system would be—simply—more deaths.

Politically, the AHCA faces risks because it tends to negatively affect the kinds of people who voted for President Trump and the new Republican majorities in the first place: rural, older Americans with higher health costs and more health problems. But beyond the politics, if any American health reform should be judged on how much it betters the health of the sickest people and how much it lowers the costs of the most expensive people, the AHCA fails on just about all the merits.