• The best ad for health services research in some time

    In the past few years, I’ve had a number of meetings with philanthropic organizations. They usually have a big donor, or some money themselves, and they want to hear about what we do to see if it’s something they’d like to fund. I’ve gotten pretty good at pitching our work, but inevitably someone asks me, “but what disease are you going to cure?”

    It’s frustrating, because obviously we’re not going to do that. I’m not a basic science researcher. I’m a health services researcher. I’m not trying to cure a disease; I’m trying to get the health care we know works to as many people as possible in the most efficient manner.

    I get that it’s attractive to be on the front lines of a major breakthrough. If you had funded the bit of research that finally ended a major disease, your name would be quite well known. But those instances are rare. Really rare. I also get that it’s much easier to understand that a researcher caring for a child with cancer needs your support more than my clinical decision support system does. I may have improved the way we screen and care for maternal depression, or domestic violence, or smoking in parents, or anemia in children, but these tales are about baby steps. They’re not as exciting or as immediately compelling. They don’t have the impact of a drug or procedure in saving a life.

    I bring this up because Ezra Klein has a longform piece out on a care model in Pennsylvania, not too far from where I grew up, that is using nurses to manage Medicare patients better. There’s no new science. There’s no new medicine. It’s health services work – reorganizing the way we care for patients – and it’s making a huge difference in terms of quality and costs. Medicare is about to shut it down. Here’s why (emphasis mine):

    Brenner puts it more vividly. “There is a bias in medicine against talking to people and for cutting, scanning and chopping into them. If this was a pill or or a machine with these results it would be front-page news in the Wall Street Journal. If we could get these results for your grandmother, you’d say, ‘Of course I want that.’ But then you’d say, what are the risks? Does she need to have chemotherapy? Does she need to be put in a scanner? Is it a surgery? And you’d say, no, you just have to have a nurse come visit her every week.”

    That’s the problem with health services research in a nutshell. When I tell people that we get results from talking to people, from getting the pieces of the system better coordinated, by advising physicians better, just by improving communication – then their eyes glaze over. They don’t see it as “real” research, or “real” care. It’s soft. But it works.

    The basic science guys looking for a cure? They’re trying to hit a grand slam. And once in a blue moon, they may succeed. It will be exciting; everyone will talk about it. But it’s unlikely to happen soon. My group, on the other hand? We’re just hitting singles over and over again. We’ve gotten really good at it. We’re not flashy, and we’re not as exciting, but day in and day out, we’re putting up real results. At some point, you have to decide which strategy is going to win more often.

    Go read Ezra’s whole piece. It’s worth it.


    • I’ve been on the boards of several non-profits, and we always go the same message from donors. “We are willing to fund projects with a distinct beginning and end, and with goals that can be explained in simple language.” It is almost impossible to get donations for operating expenses, i.e. for the daily execution of interventions that we know work. Part of this is the “sexiness factor.” A lot of it is that donors are rightfully leery of making open-ended commitments, and don’t have confidence you could find alternative funding sources if they decided to pull out.

      Its sad, and in my view, its really stupid. But its the reality of the world we live in.

    • Interesting piece, but I don’t think it got at the core issues from a health services perspective:

      Why did HQP work and 15 similar programs failed?

      Why did Medicare shut this program down? I don’t think his explanation was sufficient. The people in charge of these decisions ARE in part health services researchers – I don’t think you need to convince them of the value of better system design compared to a pill.

      • All programs aren’t alike. HQP is less telephone-based and more nurse based. Saying all communication formats are the same is like saying all medications are the same. The differences between HSR interventions are just as nuanced as the difference in drugs are.

        If we ran 15 different trials on 15 different drugs, and one trial worked, would you stop using that drug? Would you say this was an indictment of all the drugs?

        I haven’t seen a comprehensive report on why HQP is being defunded, but I have no problem guessing why. I also know that no industry or outside funder is likely to pick up the slack.

        • I agree that the core issue is that all programs aren’t alike. But that’s exactly why it’s appropriate to be careful with health services research.

          Reading Mr. Klein’s description of HQP, it’s certainly not exactly the same as the other pilots, but there’s a lot of similarity. To my understanding, all of these ideas have been tried before in different combinations. Sometimes they work, sometimes they don’t. It depends on how good your people are, how appropriate the community is for the intervention, how much local stakeholders are willing to buy in. Essentially it’s a new intervention every time, even when the core concepts are the same.

          The best feature of pills is that they’re standardized. They’re made in one facility and shipped everywhere. Of course patient populations matter, but it’s much more reliable and predictable than HQP type interventions.

          To make a basic science analogy, I work in cellular therapy. We are often described as designing a new drug for every patient. Just like health services research, it’s highly variable from place to place – different centers have different levels of success based on the people involved and slight differences in how its implemented. The core concepts are the same, but the end results are very different. And regulatory agencies are RIGHTFULLY very skeptical.

          Thus, I come to the same conclusion as the Medicare people in Mr. Klein’s article. It makes more sense to me to focus on incentives than to prescribe particular programs. There’s not one particular path to improved healthcare organization that seems to work reliably for every place. Instead, it seems to make more sense to me to focus on ACO’s, ending fee for service, not paying for readmissions, etc. Rather than expanding this program across Medicare, say that you’ll reward people who come up with the same results, no matter how they do it. Researchers like yourself will slowly build a set of best practices over time, but the local implementation will vary drastically and can’t be effectively prescribed by CMS.

          However, this is entirely outside my expertise and I’d love to hear your thoughts, Dr. Carroll.

          • I think much of what you say is correct. But it’s why I’d say that it’s in our best interest to keep HQP going and figure out why it worked and others failed. And, in the market HQP serves, it’s likely more effective to pay for their services than many other things Medicare pays for…

            • Fair enough, keep HQP going, but move slowly in expanding seems to make sense. Thanks for taking the time to respond.

    • I see the same thing in my work in developing countries. Donors are happy to fund buildings and equipment but nothing to maintain the infrastructure or run services. The result is that a lot of equipment sits idle because no one has been trained to fix it and buildings sit empty and crumbling without staff.

    • Have a look at FIg 1, page 22:

      CBO assessment of HQP, among others. CI for cost reduction, although not the end all be all, crosses zero. Ezra purported savings.

      Also, my difft slant on piece:

      • Sure, the CI crosses zero, but 1) the point estimate is well to the left, and 2) the CI is quite wide. That ought to be an argument for increased funding (to resolve the question) rather than defunding.

        • Investing in the program may be a worthwhile goal.

          However, stating the program saves costs (it may), which Ezra conveyed affirmatively, does not stack up with data in the figure, point estimate or not. If I missed something, my apologies.

    • Changes in reimbursement patterns, including penalties for readmissions and systems with prospective payment or sharing of insurance savings will soon make this type of approach far more attractive, even in Manhattan.

      The NHS in Britain has found that a similar approach, using either visiting nurses or frequent (up to several times a week) office visits, saves money in management of diabetes, COPD, asthma, and CHF. Essentia Health System (formerly SMDC) in Northeastern Minnesota and Northwestern Wisconsin has found that a program for severe CHF patients with multiple annual admissions and frequent long stays using intensive nurse contacts and provision of scales that report daily weights by phone or internet saved enough money by avoiding exhaustion of DRG payments that the system was profitable even before the introduction of readmission penalties and ACO payment systems.

      Of course this all flies in the face of the traditional US health care model of heroic intervention in health care catastrophes and will encounter a lot of resistance accordingly.

    • What we have is a health care payment system, not a health care delivery system. Without ever coming right out and saying it, Klein is indicting the payment system, a system run by and for providers (i.e., hospitals and physicians) rather than patients. Why would physicians support a system for compensating nurses rather than physicians; indeed, a system that would deprive the physicians of revenues. What Klein gets wrong, though, is equating the hospitals’ incentives with the physicians’ incentives: while hospitals derive most of their income from inpatients, physicians derive most of their income from outpatients. The incentives for integration (of hospitals and physicians) in ACA won’t produce cost savings, quite the opposite; with hospitals at the center of the integrated model, physicians are more likely to accommodate the hospitals by admitting more patients and by performing more diagnostics in the hospital. Indeed, the co-pay for most diagnostics performed in the hospital exceeds the entire fee for the same diagnostic performed in a free-standing outpatient facility. We spent the past 20 years providing incentives for moving health care services to less-costly outpatient facilities from more-costly hospitals; now ACA reverses that 20 year history and provides incentives for moving those services back to the hospital.

    • In the UK the majority of newborn and well-baby care is accomplished by visiting nurses. By utilizing the very well-maintained data base developed by this system Flemming and his associates were able to tease out the important epidemiological aspects of Sudden Infant Death(SIDS). From these data the “Back to Sleep” programs were instituted which have been shown to dramatically reduce the incidence of SIDS. In the US by contrast none of the highly technological approaches to the problem were even remotely as successful. Sometimes (maybe not so seldom?) simple,well thought out approaches give the best solution.

    • I recommend the answer “everything we know how to cure.”

    • From the 4th Report to Congress on the Evaluation of the Medicare Coordinated Care Demonstration commissioned by CMS and compiled by Mathematica Policy Research, Inc. March 2011:
      Among high-risk participants, HQP’s model of Community-based Advanced Care Management reduced deaths -30.4% (p=0.03), hospitalizations -38.8% (p<0.01), emergency room visits -37% (p=0.05), average monthly Part A and B Medicare expenditures -$511 (-35.5%, p=0.01), and average monthly net expenditures (including program fees) -$397 (-27.6%, p=0.05). This report is available in its entirety at; http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Reports/downloads/Schore_Fourth_Eval_MCCD_March_2011.pdf

      For a full summary of all studies reporting on the HQP program see:

      Note also that this same program has been evaluated by Aetna for 3 years and found to reduce hospitalizations and save money among their high-risk Medicare Advantage members. Aetna has just renewed and expanded their contract with HQP through 2015.
      For more information about HQP visit http://hqp.org

      Thanks for everyone's interest and thoughtful, considered discussion. In short, the HQP program has shown significant net savings, fewer hospitalizations and better health outcomes in higher-risk Medicare beneficiaries. To date it has not saved net dollars among low / moderate risk populations. The model differs from many nurse care management programs in having a more extensive portfolio of preventive services and more rigorous process monitoring to manage service delivery reliability. We believe these characteristics are reproducible and scaleable, but of course that needs to be tested and proven.
      Ken Coburn, MD, MPH
      CEO and Medical Director, Health Quality Partners