The Magazine

The Guinea Pig State

Oregon’s quarter-century of failed liberal health care experiments

Feb 24, 2014, Vol. 19, No. 23 • By MARK HEMINGWAY
Widget tooltip
Audio version Single Page Print Larger Text Smaller Text Alerts

Summarizing the Oregon Health Plan, the British Medical Journal explained, “States tend to favour two approaches to control costs in Medicaid: they either pay providers less or reduce the number of people eligible. Oregon rejected both methods and instead opted for a new, bold approach: it would ration the benefits covered under Medicaid.” Kitzhaber believed that rationing health care would be the best of both worlds—it would allow the state to expand Medicaid services to more people and control costs at the same time. The state would draw up a list of diagnoses and treatments prioritized for cost, necessity, and effectiveness and would pay for as many items on the list as it could afford in a given year. When the law first went into effect, the list ranked 696 medical conditions, and the state announced that anyone on Medicaid would be covered for the first 565 of them. 

Of course, the transparent rationing of medical care has political ramifications, and in this instance national politics intervened. In 1990, the merits of Oregon’s rationing plan sparked a federal debate. Oregon’s proposed changes to its Medicaid program required a waiver from the federal Health Care Financing Administration (HCFA, later the Centers for Medicare and Medicaid Services). In 1991, the first Bush administration denied Oregon’s request for a waiver. “Some Oregonians suspected that the plan was denied because George Bush, about to wage a presidential campaign against Bill Clinton, was afraid to be labeled the ‘rationing president,’ ” observed the New England Journal of Medicine. The suspicion was not paranoid. Tennessee senator and soon-to-be Democratic vice presidential candidate Al Gore had contributed an article to Academic Medicine excoriating the Oregon Health Plan. “Oregon’s decision to ration health care to its poorest women and children,” he wrote, “is a declaration of unconditional surrender just as the first battles are being fought over the future of our health care system.”  

In hindsight, the objections to Oregon’s plan to ration health care from national Democrats seem to have been a matter of political positioning rather than principle. In 1993, the newly installed Clinton administration approved Oregon’s Medicaid waiver. In fact, the waiver had a patina of bipartisanship—both of Oregon’s moderate Republican senators, Mark Hatfield and Bob Packwood, were enthusiastic supporters of the Oregon Health Plan.

One key component of the plan, however, never received a federal waiver: the requirement that employers provide health insurance to all employees. John W. Saultz, of the Department of Family Medicine at Oregon Health and Science University, described Oregon’s failure to implement the employer mandate as a betrayal of the law’s supporters. “Even before the OHP was implemented, .  .  . it had already broken faith with the most idealistic of us,” wrote Saultz in Family Medicine. “The state never seriously attempted to get the necessary federal waiver to create the employer mandate portion of the program. Thus, from the start, the OHP became a Medicaid experiment rather than a serious attempt to achieve universal health care coverage.” Clearly, the state was having second thoughts about whether forcing small businesses to provide health insurance would generate inordinate political fallout. 

Then there were two important secondary issues that Oregon had trouble addressing. First, ranking medical procedures by their cost, effectiveness, and necessity is a complicated and imperfect process. How do you compare two potentially lifesaving procedures and decide which should be ranked higher on a list? Indeed, the state’s first attempt at ranking medical procedures was alarming. According to Lancet,

The list ranked 1,600 medical treatments and contained serious flaws. The major difficulties related to inaccurate cost and effectiveness data. This led to some widely criticized rankings: reconstructive breast surgery was ranked more highly than treatment for open fracture of the thigh, and treatment for crooked teeth came higher than treatment for Hodgkin’s lymphoma. Transplantation was again near the bottom of the list as was treatment for AIDS—primary care was near the top.

The second problem was that the state lacked the insurance infrastructure necessary to expand coverage. Even with rationing, the state sought to further reduce costs and to oversee implementation by herding Medicaid patients into tightly regulated health maintenance organizations (HMOs) contracted by the state. Though they have come to be widely loathed by American health care consumers, HMOs were touted in the 1990s as the answer to controlling health care costs. In fact, Saultz writes that he and his colleagues at Oregon Health and Science University were so enthusiastic about what was happening that they formed their own HMO specifically to help provide coverage under the Oregon Health Plan. This and other HMOs would be paid a “capitation rate”—a flat fee for every patient they enrolled, regardless of how much or how little medical care the patient required. Initially, doctor participation in the plan was high, and access to care was not a problem. 

Thanks to a good economy and the fact that the Oregon Health Plan was a high priority for Kitzhaber’s gubernatorial administration, the first few years of implementation are seen as relatively unproblematic. But it didn’t take long before the system started breaking down.

From Breakdown to Meltdown

Early on, it was obvious Oregon’s rationing wasn’t cost-effective. By the time the law was implemented and state officials had agreed on the ranking of 696 procedures, the list was based on subjective judgment and political compromise rather than any rigorous cost-containment formula. A 1996 article in Medical Decision Making concluded there was “virtually no relationship between the [prioritized list] and Oregon’s own cost-effectiveness data.” A paper in the Canadian Medical Association Journal entitled “Rationing medical care: rhetoric and reality in the Oregon Health Plan” observed that the Oregon Health Plan also contained a major loophole used to skirt enforcement of the rationing rules: 

Medicaid recipients continue to receive services that are supposedly excluded by the OHP. In large part, this can be attributed to the reticence of providers when it comes to abiding by the rules. The OHP pays for all diagnostic visits and procedures even if treatment is not covered, and physicians have taken advantage of this loophole to provide uncovered medical services. Patients .  .  . have also been diagnosed with conditions covered by the list in efforts by physicians to secure patients’ access to services below the line.

As a result, the piece concluded, “there is little evidence that the OHP has operated as a model of explicit rationing .  .  . nor has its policy of cutting public coverage for services produced substantial savings.” Nonetheless, the perception that Oregon was “rationing” had some unintended side-effects. The Oregon Health Plan’s mental health and dental benefits were actually superior to those offered under commercial insurance in the state. According to the Cascade Policy Institute, these benefits were tacked on because “concerns about subjecting Medicaid beneficiaries to rationing were calmed by a benefit package that was considered by many to be generous.” 

And even when the state did make noises about enforcing stricter rationing, it was hamstrung by the federal government. To save money by covering fewer of the procedures on the state’s list would require approval by the Health Care Financing Administration. But in 1997, the HCFA started denying Oregon’s requests to move its priority line. As a result, noted Kitzhaber, the Oregon Health Plan had become “a very rich and fixed benefit, with no tools with which to manage the cost.” 

Like the rationing component of the law, the attempt to expand coverage soon started unraveling. Saultz describes his experience running an HMO in the Oregon Health Plan. “Warning signs were occurring as early as 1996. The state resisted increasing the capitation rate to the managed care plans, even as the cost of care increased,” he wrote. “We all assumed that managed care would bring cost controls to bear on the system, but cost containment failed and commercial reimbursement grew rapidly.” The result was a sizable gulf between what private insurers paid doctors and the compensation doctors received for treating patients under the Oregon Health Plan. 

The reduced payments had a pretty dramatic effect on the finances of the state’s health care providers. In the first decade of the Oregon Health Plan, Medicaid’s reimbursements to hospitals fell an average of $130 million a year short of the hospitals’ expenses. By 2003, only 23 percent of the state’s rural hospitals were operating in the black. According to the Cascade Policy Institute, the Oregon Health Plan’s failure to keep up with costs had a major impact on access to care:

Regence HMO Oregon, which began offering plan service in 34 Oregon counties, had pulled out of 23 rural counties between 1996 and 1999. ODS Health Plan stopped offering the state plan in three rural counties because of nearly $1 million in losses. Sure Care of Roseburg offered the health plan in two rural counties but abandoned both by 1999. By 2000, Kaiser Permanente pulled out of the rural counties of Columbia, Linn, Benton, and Yamhill and exited the plan completely in 2002. In addition, by 2000, Regence completely exited the Oregon Health Plan, while Providence Health Plans and ODS Health Plan pulled out of the Portland area markets.  

The Oregon Health Plan’s problems were in many respects predictable, but defenders of the law argue that unforeseen shifts in the political and economic landscape prevented the law from building on its early success. Even those who acknowledge the plan’s failure rarely fail to credit the law with reducing Oregon’s uninsured population. A single statistic pops up again and again: The share of uninsured Oregonians dropped from 18 percent in the late 1980s to 11 percent in 1996, after the plan took effect. Those figures, however, come from the Office for Oregon Health Plan Policy and Research. The Cascade Policy Institute, a right-of-center think tank, however, notes that U.S. census data show the percentage of Oregonians without insurance actually increased between 1990 and 1996, from 12.4 percent to 15.3 percent. Oregon’s uninsurance rate closely tracked the national average, which went from 13.9 percent to 15.6 percent over that period. It’s possible that the state figures are more accurate than the census data. It’s also possible that the state was so eager to prove its expensive experiment in expanding coverage a success that its numbers are suspect. 

The Oregon Health Plan limped along through the 1990s. Oregon voters approved a hike in the state cigarette tax to pay for the plan, but costs were soon completely out of control. “Its budget swelled from $1.33 billion in 1993-1995 to $2.36 billion in 1999-2001,” according to Willamette Week, which also noted that the Oregon Health Plan had prompted “a surge of public hostility toward HMOs.” As if that weren’t enough, Oregon endured arguably the worst state economy in the country during the recession of the early 2000s. State revenues plummeted even as costs were exploding. 

In 2002, with Kitzhaber term-limited and about to leave the governor’s office, he made one last stab at salvaging the Oregon Health Plan. This time the goal was to expand coverage to an additional 46,000 people, bringing the state nearer to universal coverage. This would be done by creating a two-tier system. OHP Plus would enroll the traditional Medicaid population and cover the prioritized list of services. OHP Standard would enroll the new population—those making as much as 185 percent of the federal poverty level but not otherwise eligible for Medicaid. In order to save money, OHP Standard offered significantly reduced benefits. It also introduced co-payments, including $250 for a hospital visit. Failure to make co-payments would result in a suspension of enrollment in the Oregon Health Plan for six months. 

While these changes were seen as Kitzhaber’s attempt to rescue his legacy, they also reflected a shifting of political winds. Republicans had taken control of the state legislature during Kitzhaber’s second term and had legitimate cause to start agitating for fiscal reforms. The introduction of co-pays and reduced benefits for those in the Oregon Health Plan whose higher incomes otherwise rendered them ineligible for Medicaid seemed to reflect Republican priorities. The two-tier plan drew bipartisan support. 

Washington approved Oregon’s waiver and allowed it to begin implementing the two-tier system in February 2003, a few weeks after Kitzhaber left office. The result was a meltdown in the Oregon Health Plan’s enrollment. The federal waiver application had predicted “no negative impact on access due to cost sharing,” but that turned out to be wrong. After co-payments were introduced, enrollment dropped 53 percent, from 104,000 people in January 2003 to 49,000 by December. By 2004, OHP Standard had only 24,000 participants, and it was closed to new enrollees. The percentage of Oregonians who were uninsured was nearly identical to the percentage before the Oregon Health Plan was implemented.  

The Experiment

The Oregon Health Plan isn’t just a failure; it’s such a failure that it has almost totally undermined the notion that government subsidized health insurance is ever effective. Yet this very assumption—that expanding access to government health insurance programs will control costs and improve health—is a central tenet of Obamacare and the progressive health care reform project. In 2008, Oregon was finally able to reopen enrollment in OHP Standard to an additional 35,000 residents. After the four-year hiatus, however, there was a waiting list of nearly 90,000 Oregonians trying to enroll, so new enrollees were selected by lottery. Of the 35,000 Oregonians who won the lottery, only 60 percent bothered to fill out an application and enroll. 

Still, savvy social scientists realized that Oregon’s health care lottery presented a once-in-a-lifetime opportunity to measure the effects of Medicaid. By tracking the health outcomes of thousands of Oregonians enrolling in Medicaid, along with those of a control group of thousands who were on the waiting list but didn’t get to enroll, they could measure Medicaid’s effects on a number of broadly predictive health indicators such as cholesterol, blood pressure, and diabetes. Former Washington Post health care blogger Ezra Klein touted the study in generous terms. “The gold standard in research is a study that randomly chooses who gets a new treatment and who doesn’t,” he wrote. “That way, you know your results are unaffected by differences in the two populations you are studying.” Moreover, the sample size—over 12,000 Oregonians—was nearly unprecedented.

The “Oregon Medicaid experiment,” as it came to be known, was performed by respected health care experts—including MIT’s Jonathan Gruber, a key architect of Massachusetts’s health care reform and one of Obamacare’s biggest defenders. (In 2010, it came out that Gruber had been promoting Obamacare without disclosing his $297,600 contract with the Department of Health and Human Services.) The Oregon Medicaid experiment announced its first round of results in 2011. They measured only the first year and, importantly, included only improvements in health self-reported by the study’s participants. No objective measures of health outcomes were included. Nonetheless, the findings were mildly positive and were greeted with a chorus of hosannas from liberal health care wonks. 

Klein’s write-up of the study’s first-round results ran under the condescending headline “Amazing Fact! Science Proves Health Insurance Works.” The study’s “findings were irrefutable,” noted the Century Foundation. Aaron Carroll, who blogs at the Incidental Economist, asserted that since this was a randomized controlled trial the link between Medicaid enrollment and better health was practically a given. “An RCT [randomized controlled trial] is pretty much the best way to prove causality. .  .  . We can even start talking causality,” Carroll wrote. 

Then in May 2013, the second round of results from the Oregon Medicaid experiment was published in the New England Journal of Medicine. The results had been expected in 2012 but were delayed without explanation. Whatever the reason for the delay, the findings would not have been politically helpful to Obamacare defenders had they been released in the middle of the presidential campaign. “Medicaid coverage generated no significant improvements in measured physical health outcomes in the first two years, but it did increase use of health care services, raise rates of diabetes detection and management, lower rates of depression, and reduce financial strain,” concluded the study. Now, this result shouldn’t have been surprising. A raft of studies have come to similar conclusions about Medicaid’s general ineffectiveness. With regard to certain specific health measures, there is actually evidence showing those in Medicaid have worse outcomes than the uninsured. 

But after having overpraised the Oregon Medicaid experiment and credulously accepted its early findings, liberal wonks nearly threw their backs out attempting to hoist the goal posts out of the ground. Suddenly they were alternately pointing out major limitations of the study and seizing on the marginal benefits it suggested Medicaid provides. Having asserted that Medicaid coverage causes better health outcomes, Carroll and co-blogger Austin Frakt were now saying, “Chill, people. [The Oregon Medicaid experiment] is another piece of evidence. .  .  . [I]t didn’t show that Medicaid harms people, or that the ACA is a failure, or that anything supporters of Medicaid have said is a lie.” The New York Times’s news article on the results was misleadingly headlined “Medicaid Access Increases Use of Care, Study Finds.” Mother Jones went with “Medicaid Probably Does Improve Health Outcomes After All.” 

As for the supposed benefits of Medicaid coverage demonstrated by the Oregon Medicaid experiment, it’s worth noting two things. One, Obamacare supporters seem content with the notion that increased health insurance coverage is the goal regardless of whether it provides quality health care or is financially sustainable. Two, the Oregon Medicaid experiment did show a barely statistically significant improvement in outcomes for the seriously mentally ill. But the improvements in mental health are a curious result to hang your hat on. The Oregon Health Plan has long since eliminated or scaled back its once generous dental and mental benefits. According to a 2009 state workforce survey, “two-thirds of physicians report only sometimes or never being able to find inpatient mental health services for Medicaid patients. That increases to three-quarters for outpatient mental health services.” If serious mental illness is the one area of slight improvement in outcomes, it’s in spite of the fact that those in the Oregon Health Plan have a very hard time getting treatment for mental problems. 

But for a great many Obamacare foot soldiers, the reaction to the Oregon Medicaid experiment was outright denial. “Above all, you should bear in mind that if health insurance is a good idea—and you are nuts if you let this study persuade you otherwise—Medicaid is cheaper than private insurance,” observed New York Times columnist Paul Krugman. “So where is the downside?” It shouldn’t be necessary to explicate the massive qualitative differences between Medicaid and private insurance. A Ford Pinto with a leaky carburetor and a brand new Honda Civic are both forms of transportation. The former may be cheaper, but it’s much less likely to get you where you need to go day in and day out. 

If a Nobel Prize-winning economist thinks that there’s no downside to American taxpayers spending $7 trillion on a program that the “gold standard” of social science tells us may result in an uptick in Lexapro prescriptions at best, Obamacare’s defenders would seem to be driven by religious zeal rather than empirical evidence. 

Lessons Learned 

Though defenders of Obamacare and other liberal health care reform projects remain unmoved by the evidence, there are a number of lessons to be learned from Oregon’s 25 years of failure. In 2008, John Saultz, drawing on his years working with the Oregon Health Plan, laid out four: 

First, the initial steps to take are to define health system goals and to tackle the question of defining basic benefits. The method of doing this must be a public process and cannot be created purely by policy wonks or government bureaucrats. Oregon discovered that a public process can work, at least in a small state like Oregon. It remains to be seen how this might work in a state with a much larger and more diverse population or a less progressive public policy community.

Second, it is important to keep in mind that health care reform is hard work and takes time. If you try to go too fast, you will make too many mistakes and public support will falter.

Third, do not underestimate the opposition. Health care spending now constitutes more than 16 percent of the American economy. Lots of money is involved, and lots of people and organizations do not want change to happen.

Fourth, always remember that the devil is in the details. The best ideas in the world won’t help you if you cannot deliver a plan that ordinary people can understand.

 

Though he’s been chastened by experience, Saultz remains a guarded supporter of the liberal health care reform project. So it should be even more dismaying that the very year after Saultz laid out these lessons, Congress took up health care reform at the national level and proceeded to do the exact opposite of everything Saultz recommends here. 

It is also tempting to say that Oregon’s experience isn’t necessarily comparable to what will happen with national health care reform. The open attempt at rationing made the politics and details of the Oregon Health Plan very different from those of Obamacare. As Saultz notes, however, Oregon’s attempt to define benefits was relatively transparent. Currently, millions of Americans who had their policies canceled after being promised they could keep them are struggling to ascertain what their coverage will be like on the new Obamacare exchanges. And just because Obamacare doesn’t explicitly ration services, as the Oregon Health Plan tried to do, there’s no guarantee that the federal government won’t adopt rationing in the future. 

Obamacare creates two new bureaucracies: the Patient Centered Outcomes Research Institute (PCORI) and the Independent Payment Advisory Board (IPAB). PCORI is tasked with doing research on what medical treatments are most cost-effective. IPAB is supposed to be a board of presidential appointees—Obama has yet to make appointments—confirmed by the Senate who are tasked with lowering the Medicare budget. IPAB’s recommendations automatically become law unless they are overridden by a three-fifths majority in the Senate or Congress passes its own Medicare plan that meets the same spending target. So far, IPAB is expressly forbidden from rationing care—or adjusting Medicare premiums, deductibles, and co-pays. Thus, it has curiously limited powers considering its daunting task of reining in the biggest driver of federal debt. If the federal government ever does want to start rationing Medicare services, the administrative infrastructure will be in place. It is suggestive that in recent years, the Centers for Medicare and Medicaid Services (CMS) has started declining to pay for clinically effective but expensive cancer drugs. (Once again, Oregon blazed the trail: The state generated a flurry of headlines in 2008 after it sent a letter denying a 53-year-old man with prostate cancer chemotherapy but offering to cover the cost of physician-assisted suicide.) 

Former CMS head Donald Berwick, who was recess-appointed to the agency by Obama after GOP senators objected to his explicit endorsement of rationing medical care, is running for governor of Massachusetts. Berwick is trailing in the Democratic primary, but rationing is no longer an unutterable word among politicians in blue states. As everyone from Al Gore on has demonstrated, liberal health care champions are always  against rationing right up until they are for it. Medicare has long been “a very rich and fixed benefit,” and Democrats have shown an absolute unwillingness to adopt market-based cost control proposals offered by Republicans. If Democrats want to control Medicare costs, rationing is the only real alternative. Rationing is also a central feature of the single-payer system, for which the Democrats’ liberal base is increasingly agitating in response to Obamacare’s failures. Obamacare opponents should gird themselves for the day when rationing inevitably becomes part of the national conversation.

Given that Medicare is the biggest driver of federal debt, it makes sense that Obamacare is focused mostly on reining in Medicare costs. Yet Obamacare also aims to expand Medicaid coverage to as many as 20 million people, and there is still no cogent plan to keep the program from busting budgets. How desperate is the Obama administration for solutions? On May 3, 2012, the administration announced it was giving Oregon $1.9 billion to enact new Medicaid reforms. In exchange for the money, Oregon is promising to keep Medicaid costs from rising more than 2 percent a year, a goal that, if met, would save the federal government $11 billion over the next decade. The Obama administration is hoping that Oregon’s reform ideas can be replicated in other states. Indeed, just about every other governor in the country would kill for lavish federal funds to shore up their state’s Medicaid budget. But considering Oregon’s history of failures, a $1.9 billion gamble on its notions of health care innovation has pretty long odds. 

This brings up the final lesson to be learned from Oregon’s attempts at health reform: Opponents of government-run health care will have to remain vigilant, because liberal health care crusaders have shown a resolute and dumbfounding willingness to keep trying the same failed ideas year after year. Here’s how the Washington Post describes the proposal that won over federal authorities: “Oregon wants to move its 600,000 Medicaid population into ‘Coordinated Care Organizations.’ These are health care systems that will accept a flat fee for all care delivered. While remaining within that budget, they will have to hit certain quality metrics, ideally creating financial incentives to deliver the most cost-effective care that can deliver good results,” observes the Post. “If this idea sounds familiar, that’s because it shows up in the Affordable Care Act: Accountable Care Organizations will use a similar structure within the Medicare program.”

Once you understand that “Coordinated Care Organizations” and “Accountable Care Organizations” are just an attempt to rebrand the widely hated HMOs, this should all become even clearer. The details of Oregon’s new proposal also sound eerily similar to what was first laid out by John Kitzhaber nearly a quarter-century ago. 

Meet the new Oregon Health Plan, same as the old failed Oregon Health Plan. 

Mark Hemingway is a senior writer at The Weekly Standard.

Recent Blog Posts