A Tale of Two IT Procurements

Recently, the President of the United States, the most powerful person on earth, the man whose finger rests on the nuclear button, struck a bold blow for . . . procurement reform?

“There are a whole range of things that we’re going to need to do once we get [the Affordable Care Act (ACA) rollout] fixed—to talk about federal procurement when it comes to IT and how that’s organized,” the president said on November 4, speaking to a group of donors and supporters.

People are clamoring for heads to roll, and the president is talking about what just could be the geekiest, most obscure topic ever to clog a federal bureaucrat’s inbox. Procurement reform? Has he gone off the deep end?

Well, not really. Among the causes of healthcare.gov’s difficulties, the federal process for purchasing goods and services could rank right up there with toxic politics, lack of funding for ACA implementation, and management goofs. Let me explain why, from personal experience.

From 2009 to 2011, I served as National Coordinator for Health Information Technology. My job was to implement the HITECH ACT, which aims to create a nationwide, interoperable, private, and secure electronic health information system. As national coordinator I had to lead a lot of federal contracts.

This is how that went.

Our staff would decide what services we needed, write a request for proposals (RFP), and send it off to a totally independent contracting office. That office could be within the Department of Health and Human Services (DHHS), but if the DHHS office was too busy, the RFP could go almost anywhere: the Department of the Interior, the Department of Housing and Urban Development, the Department of Education—whatever contracting office had time to process the work.

Officials extensively trained in the details of federal procurement, but lacking familiarity with our programs or field of work, would put the RFP out to bid. An expert panel—over which we had minimal control—would evaluate the responses. Months later, the contracting office would present us with the signed contract.

The winner was usually picked from a group of companies with considerable experience working the federal procurement process. If we weren’t happy with the firm, or with their later performance, there was virtually nothing we could do about it.

Getting out of this shotgun marriage meant months of litigation, during which the funds would be frozen and the work itself would grind to a halt.

After I left the federal government, I became Chief Health Information and Innovation Officer at Partners Health System, a large, Harvard-affiliated integrated health system in Boston. There, I was responsible for managing another procurement: the choice of a new electronic health record (EHR) for the organization’s many hospitals, clinics, and clinicians. This was one of the largest capital projects in Partners’ history. The experience was very different.

I led an expert team of clinicians, IT professionals, financial analysts, clinical managers, materials management professionals, and external consultants who spent months comparing the EHRs and companies that were finalists in our selection process. The vendors’ CEOs and senior leadership met with our CEO and senior leadership to get a sense of whether there would be mutual rapport and trust.

To see the EHRs in action, we visited other customers who were using the products. The vendors came to Partners and demonstrated their EHRs to hundreds of clinicians, IT professionals, and managers, who then voted on which they preferred.

That’s not all. A Partners team visited their headquarters and interviewed their key staff. Our technologists and programmers consulted with theirs to assess in detail their skill and vision for their product. We evaluated the companies’ culture, financial plans, and likely stability over time. After we selected a leading candidate, we spent weeks going over the contract line by line with IT professionals, lawyers, and contracting experts.

The Partners’ procurement process may not be perfect, but the federal process is clearly broken. I have no personal knowledge of how it worked in the case of the federal website, but I can’t imagine that it was up to the task of managing the subtle, technologically complex, and unprecedented challenge of putting together a consumer-friendly website that would seamlessly connect multiple legacy state and federal databases that were never designed to work together.

Why is the federal procurement process so dysfunctional? Mostly because the public seems not to trust government officials. The process is designed to protect against conflict of interest and fraud. When faceless, interchangeable contracting officers manage the process behind closed doors, companies find it difficult to cozy up to federal officials for the purpose of influencing outcomes.

There is something to be said for these boundaries when it comes to routine procurements such as furniture, transportation services, maintenance, and cleaning. But when it comes to the incredibly complex technical work that is required in the IT space, a process in which the people overseeing a project don’t have a chance to interact with potential vendors in advance is a prescription for the kind of problems we are now witnessing with healthcare.gov.

Does this mean that the federal government is incapable of managing complex IT projects? Not at all. In our military and intelligence, we have information technology capabilities that amaze the world. Their consequences, for good and ill, now rival healthcare.gov for time and space in the American media.

President Obama is absolutely right: the federal government can and should do better in IT procurement, at least in the domestic arena. In the information age, this is a core capability for government at all levels. And given the entrenched nature of the current process, it may take an order from the commander in chief to get meaningful reform under way.

David Blumenthal, M.D., M.P.P., is president of The Commonwealth Fund, where this post originally appeared.

12 replies »

  1. I researched the above link, indicating a med mal action settled by Dr. B:

    Malpractice Information

    This report reflects malpractice payments made in Massachusetts reported to the Board by Massachusetts medical malpractice insurers. You should take into account how long the physician has been in practice when considering the number of payments made.

    Details for Payments in Internal Medicine Specialty
    Date Category of Payment
    4/13/2009 Above Average

  2. Disciplinary and/or Massachusetts Criminal Actions

    Massachusetts Criminal Convictions, Pleas and Admissions

    The Board has no record of felony or serious misdemeanor convictions regarding Dr. Blumenthal.

    Health Care Facility Discipline

    The Board has no record of health care facility discipline regarding Dr. Blumenthal.

    Massachusetts Board Discipline

    Dr. Blumenthal has not been disciplined by the Board.

    Out of State Board Discipline

    The Board has no record of out of state discipline regarding Dr. Blumenthal.

  3. @david_blumenthal You are being disingenuous. Partners was going to buy Epic from the getgo.

    You would have more credibility if you disclosed the details of your med mal lawsuit that you settled before coming to Washington, especially the role that the EHR had in the medical negligence.

  4. Bill – that’s a good point. And there are also lots of fake small business shenanigans that go on in federal contracting.

    But the issue for Healthcare.gov wasn’t the selection process as much as it was the government (i.e. those who planned and executed the procurements) not asking for the right things. Buy a bunch of piece parts and you’ll get a bunch of piece parts. Give the contractors the ability to point fingers at the other guys (or the government) and they will.

    Any contractor who signed on for this job knew that it wasn’t going to be easy, knew that it would occur in a politically charged environment that would challenge planning. Those who accepted the money accepted the challenge, responsibility and risk (financial and reputational).

    I’ve been a part of federal procurement protests from both the contractor and governmental sides. Good contracting teams will make sure that contractors don’t have grounds for protest and will dispatch their claims quickly. Bad contracting teams will allow program staff to inject bias into the evaluation process and provide a basis for later protests (and delays). It all goes back to establishing clear and appropriate evaluation criteria for the procurement and having people on the contract and program sides with experience and education in federal contracting.

  5. One additional element that further complicates state and federal IT procurement processes is the ever-present threat of litigation from one or more of the unsuccessful bidders. This fosters a culture where there is little or no room for subjective input, even if in the form of expert opinion. While it promotes a fair and unbiased evaluation process, it doesn’t drive the selection of the vendor who may be the best suited to carry out the work at hand. The private sector is not immune from this litigious environment, either, as I have witnessed first hand vendors taking non profit health care organizations to court for similar allegations of unfair vendor selection processes.

  6. There are plenty of private sector IT projects that have stumbled and failed as well.

    Something tells me that the Epic implementation at Partners isn’t going to be 100% perfect on day 1…or day 90.

    Some of the issues with ONC procurement under Dr. Blumenthal’s ONC had much more to do with the tight timelines imposed by politics and the lack of experience and infrastructure in ONC for contracting than with government IT contracting more generally.

    Remember that ONC is not an agency but an office in the Secretary’s office. HITECH formalized it but the period after HITECH’s signature was a period of great chaos and pressure at ONC. And neither Dr. Blumenthal nor Dr. Mostashari (nor most of their staff…pulled from either Harvard or NYC) knew much about federal contracting.

    In other words, ONC wasn’t efficient and effective at contracting because ONC lacked experience, expertise and infrastructure / resources / vehicles in this area. And their plans were scuttled by impossible time lines imposed for arbitrary and political reasons.

    What’s to be taken from the Healthcare.gov fiasco, in my opinion, is that agencies (and offices) should develop expertise in federal contracting and establish contracting vehicles to support their current and anticipated needs. Having a close relationship between the program and contracting staff is also important. Too often program staff (CORs) tasked with serving as the technical lead on projects don’t have the skill or time allocated to do it well. There are best practices that can lead to success.

    What should not be taken from Healthcare.gov is that the government should establish different (and new!) processes for contracting for technology (software? websites?).

    CMS has deep cultural issues that greatly complicate coordination internally and externally. Being directed by political leadership in HHS didn’t make matters any better, I’m sure. But ultimately the blame for the shortcomings in Healthcare.gov falls on both the federal contracting staff who should have done better and the contractors who signed on for a job and failed to deliver. Bidding on an (I’m sure) flawed RFP is voluntary!

  7. Nice post. It would have been nice if you’d taken things one stop further and told us what specific reforms you would recommend to the procurement process. You are uniquely positioned to offer insight here.

    Frankly, I’m not comfortable that this administration is willing to take this kind of political risk.

  8. This is useful input from someone who 1) understands IT in depth and 2) has been there. Did the federal IT procurement process prevent CMS from hiring a prime contractor to supervise the 55 subs?