Health and Human Services Secretary Michael Leavitt, left, and President George W. Bush watch volunteers help seniors sign up for Medicare Prescription Drug Benefit in 2006. (PAUL J. RICHARDS/AFP/GETTY IMAGES)

Michael O. Leavitt served as secretary of health and human services under President George W. Bush. He is the founder and chairman of Leavitt Partners, which advises and invests in health-care organizations and consults with state governments on the implementation of the Affordable Care Act.

This past spring, Sen. Max Baucus (D-Mont.) called the impending implementation of the Affordable Care Act “a huge train wreck.” His words caught my attention because the last time the federal government delivered a new health-care benefit to more than 40 million people, I drove the train.

As secretary of health and human services during President George W. Bush’s second term, I faced the daunting task of rolling out Medicare’s new prescription drug benefit. Commonly referred to as “Part D,” the program is considered a tremendous success: Premiums have remained low, the program operates well under its projected budget and 90 percent of seniors are satisfied with their plan.

But in early 2006, there were days when I thought we could crash at any moment. For several weeks, the rollout of Medicare Part D felt like a runaway train — bumpy, uncomfortable, unnerving. Fortunately, the ride ended safely.

I opposed the Affordable Care Act, and I still believe that big changes to the law are necessary. But I’m not hoping for a wreck. That outcome would hurt ordinary people, not just politicians. Avoiding a calamity will be a major test for the Obama administration. To succeed, it should learn from our experiences with Part D — what we did well and where we fell short.

Part D and the Affordable Care Act resulted from contentious negotiations and fierce legislative battles. Both charged the Department of Health and Human Services with creating an insurance marketplace where people could choose among competing private plans. Both involved new regulations and information-technology systems, approval of insurer bids and plans, coordination with federal departments and state governments, and the education of millions of Americans. However, the ACA’s challenges are even greater than those we faced, given the law’s complexity, size and scope.

The Obama administration assures us that it will be ready, but independent analyses and recent delays signal potential trouble. In particular, I see four risks that could cause this train to jump the tracks:

Insufficient education

With the ACA’s initial enrollment period three months away, 78 percent of Americans lack awareness about the law and the changes it will bring. Four in 10 don’t even know the law is set to take effect. So it’s not surprising that 58 percent of the uninsured report that they don’t have enough information to answer the important question: What will the ACA mean for me?

With Part D, the education hurdles were also high. Before the program was implemented, only 21 percent of seniors had a favorable opinion of it, and 66 percent didn’t understand what the reform would mean for them.

So we spent 18 months devising and implementing a campaign to explain the prescription drug benefit, prepare seniors as well as partners — such as community groups, churches, pharmacies, insurance plans and state and local governments — and then sign people up. A national bus tour supported each phase. The summer before enrollment (the same period that the ACA is in now) we logged more than 600,000 miles and visited 48 states. As secretary, I made 119 stops in 98 cities. I learned that with a program like the ACA, you can’t count on Washington to sell it. You have to reach people where they live, work, pray and play.

Compared with Part D, ACA implementation is behind schedule. The administration and its allies have waited until this summer to begin an education campaign, to explain how fine print becomes lived experience.

To close the awareness gap, HHS leaders need to hit the road and prepare those seeking coverage for inevitable technical glitches and delays. They must also remember that the biggest political risk isn’t unhappiness among the tens of millions who don’t have insurance — it’s sticker shock among the hundreds of millions who do, and whose premiums are likely to rise. The administration needs a narrow message for the uninsured and must have a more persuasive message for everybody else.

Technology breakdowns

When the ACA starts delivering benefits in January, its success will depend in part on the technical infrastructure the government has built. The centerpiece is a hub that will connect and transfer data between the in-house systems of various government agencies and private entities.

Imagine an uninsured 27-year-old man living in Texas — a key demographic in one of the administration’s target states.

When he enrolls online, he will pick a plan and submit information through his state’s marketplace system, which is connected to the federal hub. The hub will transfer his information to several federal entities. The Internal Revenue Service will verify his tax information and the Department of Homeland Security will check his citizenship or immigration status. Depending on his circumstances, HHS, the Veterans Health Administration, the Social Security Administration and other federal agencies may be looped in. His insurance provider will receive much of this data so it can collect his premiums, communicate with his doctors and begin providing coverage starting Jan. 1.

These numerous touchpoints make it likely that systems will slow down or break down. That possibility became more evident last month, when the Government Accountability Office revealed that many aspects of the ACA’s information technology effort are behind schedule and untested.

The IT system for Part D linked dozens of databases, hundreds of insurers, thousands of drug plans, tens of thousands of pharmacies and tens of millions of beneficiaries. Although the back end was a complex web of coding that could process and transfer information quickly, the public “plan finder” narrowed down more than 3,000 drug plans to the six that best met the needs of each enrollee.

Normally, a system this complex can take up to three years to create. We built it in nine months, which left little time for testing. When the system started running, glitches caused delays. In the early morning hours of Jan. 1, 2006, an HHS colleague visited a few pharmacies to see how Part D was working.

“How many Part D prescriptions have you been able to fill?” my colleague asked a busy pharmacist in Front Royal, Va.

The pharmacist answered, “None.”

Our runaway train ride began that morning. In the next few days, we began hearing scattered complaints that some seniors were showing up at pharmacies and being told they weren’t in the system. Even though the vast majority of prescriptions were being filled without delay, every quandary had a human story. News coverage of these examples made it seem like the problems were pervasive, and we didn’t have data to determine the extent of the problems.

Our crisis-management team identified our biggest challenges and created metrics to gauge our progress. Armed with better data, we pinpointed IT issues and the agencies collaborated to resolve them.

The Obama team should learn from our mistakes. It needs a dedicated unit to respond to IT problems and offer a set of metrics and the means to track them. These steps can prevent small problems from seeming bigger and big problems from bringing down the system.

Subsidy errors

One way the ACA aims to make insurance affordable is by giving people a federal subsidy to defray the cost of buying coverage on their state exchange. To qualify, an individual’s income must fall between 100 percent and 400 percent of the poverty line (unless you qualify for Medicaid, in which case the threshold is 133 percent).

The process for determining and delivering these subsidies is perilous. For starters, people are allowed to attest to most of the information on their federal application; it is verified later, creating a risk of incorrect subsidies. Additionally, decisions on 2014 subsidies will be made using 2012 tax data — information that could be out of date in a volatile economy.

The administration’s decision to delay the employer mandate was wise, but it further complicates matters. Under the ACA, if you turn down affordable employer-based coverage, you do not qualify for a subsidy at taxpayer expense. By giving employers a year-long reprieve from reporting their insurance offerings, the administration is now unable to verify whether workers should have access to affordable employer-based insurance in 2014. The result is an even greater risk that many who do not qualify will receive a subsidy.

A 2011 study in the journal Health Affairs predicted that many enrollees will either get a smaller subsidy than necessary or receive one when they shouldn’t qualify. The delay in the employer mandate exacerbates this problem and its related conundrum: Should the government make cash-strapped Americans give back subsidy money? If so, how?

We faced a similar problem with Part D. Because our beneficiaries were seniors, we arranged for premiums to be deducted from their Social Security checks automatically. However, a technical glitch incorrectly calculated some payment amounts. Each month, some seniors were charged about $20 too much, while others were charged $28 too little.

The thought of asking seniors to write a check to the government was unimaginable. But allowing some beneficiaries to get a nearly free ride at taxpayer expense was equally so.

The Obama administration may face such problems on a massive scale. Estimates indicate that the average subsidy could be more than $5,000 a year, with many running well into six digits. Fixing mistakes will mean asking low-income people to repay money they don’t have.

There are no easy solutions for these problems. HHS simply needs to be prepared for what is coming.


Implementing the ACA requires thousands of partners working together. During Part D, the states, insurers and pharmacies were not as prepared as they should have been. Although HHS certainly deserved some of the blame, one of the most important things we did in those first tough weeks was assume full responsibility.

Early on, the biggest problem was getting state Medicaid computer systems to connect with federal systems. In 13 days, I visited 20 states and met with 18 governors, and worked hard to secure their support. My mantra: “We own these problems. We’ll find them, fix them and finish them.”

At a critical moment, I was summoned to meet privately with members of the Senate Finance Committee. I told them what I’d told the governors, provided timetables and promised updates. Sen. Chuck Schumer (D-N.Y.), one of our consistent critics, said, “I don’t think we can ask for any more than that.” Candor bought us time.

Three moves will help the Obama administration own the problems and avoid the blame game. First, it should give a point person the overall responsibility and authority to get this job done — someone who can serve as the public face of the implementation effort, and whom the American people can get to know and trust. Second, HHS should organize a joint crisis-management center with insurers — a place to solve problems and establish metrics. When the system goes live, insurers should serve as intelligence-gatherers, providing early insight into problems and helping HHS respond.

Third, the administration has signaled that it will be aggressive in enforcing the ACA’s serious penalties for noncompliance. Fair enough. However, it should work with insurers who have made a good-faith effort to comply. The administration will be more successful if it reassures insurers that they are viewed as partners, not scapegoats.

Although there are many similarities between the ACA and Medicare Part D, there are important philosophical differences. The ACA reflects the belief that government should play a much bigger role in making our health-care decisions, while the drafters and implementers of Part D held the view that government’s role in health care should be limited to organizing a system of competition, where consumers are empowered to make choices and are protected from unfair treatment.

In my view, many of the ACA’s problems stem from this distinction. Its top-down ideology collides with the adaptability required in a marketplace of millions of consumers.

There is a need for smart reform in our health-care system. But if Sen. Baucus is right and the ACA’s implementation ends in a wreck, my hope is that the law’s strengths and weaknesses will become lessons learned in our continued quest for a uniquely American solution to our health-care dilemma.

Read more from Outlook, friend us on Facebook, and follow us on Twitter.