Uncategorized

Stop the Presses: A Disease Management Program Worked

I am known in the disease management and wellness fields as a naysayer, critic, curmudgeon, and/or traitor…and those are only the nouns that are allowed to be blogged across state lines.  This is because I am driven not by wishful thinking but rather by data.  The data usually goes the wrong way, and all I do is write down what happened.  Then the vendors blame me for being negative — sort of like blaming the thermometer because the room is too hot — because they can’t execute a program.

However, the nonprofit Iowa Chronic Care Consortium (ICCC) apparently can execute a program.  They reduced total diabetes events by 6% in the rural counties they targeted.  This success supports a hypothesis that in rural (presumably underserved) areas, disease management fulfills a critical clinical gap:  it provides enough basic support that otherwise would not be provided even to those who actively seek it to reduce near-term complications and exacerbations.

This result will likely produce its own unanticipated consequence: because many people now believe (thanks, ironically, to some of my own past work) that disease management doesn’t produce savings, there will be widespread skepticism about the validity of this study.  Quite the opposite:  this “natural experiment” is as close to pristine as one could hope for in population health, for five reasons:

  1. There was no participation/self-selection bias because outcomes were measured on all Iowa Medicaid members.
  2. The program was offered in some Iowa counties but not others, so there was no eligibility or benefits design bias, Medicaid being a statewide program.
  3. The program encompassed only one chronic condition (diabetes) rather than all five common chronic conditions normally managed together (asthma, CAD, CHF, and COPD being the other four).   Since all five conditions were tracked concurrently, whatever confounders affected the event rate in one of those conditions should have affected all of them.   And event rates in the four other conditions did indeed move together in both the control and study counties.   Just not diabetes.
  4. The data was collected exactly the same manner by the same (unaffiliated) analysts using exactly the same database so there is no inter-rater reliability issue.
  5. Both groups contained hundreds of thousands of person-years and thousands of events.

As one who has reviewed another high-profile “natural experiment,” North Carolina Medicaid, and found that the financial outcomes were the reverse of what the state’s consultants originally claimed (incorrectly, as they later acknowledged by changing their answer), I can also say that natural experiments in population health don’t harbor some as-yet-unidentified confounder that causes the study population to outperform the control population.

The analysis of the data, like all good population health analyses (meaning very few of them), is fully transparent.  I will share it gratis with any researcher in a nonprofit organization, and I welcome private or public criticisms.   I’ll respond to the latter right beneath this posting.

So what did ICCC do to achieve this enviable result?   Much of the intervention was standard procedure.  To begin with, a technology platform provided by a population health vendor used an interactive voice response (IVR) system to populate a web-based decision support system with self-reported daily diabetes self-management activities.  Unfortunately, my conclusion — one in which the New England Journal of Medicine concurs — is that IVR by itself doesn’t improve chronic disease outcomes.

But in addition to this standard procedure came some much more personal and community-based interactions, which likely accounted for the improvement due to the lack of other resources in these rural areas.     A patient centered care manager developed a personal connection with no more than 250 participants at a time (vs. twice that in a typical program), providing each with what might now be called health advocacy.  Clinical variances from “normal” were flagged, enabling the care manager to easily identify participants who needed extra support, coaching or referrals to their physician for medical care.  Quarterly behavioral health screenings identified those with co-morbid depression for referral into Medicaid’s statewide Magellan Behavioral Health program.

Participants were also referred to community based programs operating in some counties, like Better Choices, Better Health®, the Stanford University Chronic Disease Self-Management Program, to promote development of self-care skills, as well local diabetes education programs, programs that 96% of ICCC program participants had not previously even known existed.  Use of these community-based resources improved medication adherence, attention to diet and self-confidence regarding their diabetes self-care skills.

In a well-served population, even a well-served Medicaid population, little incremental adverse event reduction has been achieved through providing similar additional care management resources and/or referrals to community-based resources.  But this is first time that a rural Medicaid population has been segregated from the total Medicaid population, and the promising result in that specific population shows that there could be an opportunity to disease-manage other underserved rural populations, at least in diabetes.

Now if somebody could please find me a wellness vendor that can “move the needle” in a similar fashion, I’ll be able to say that, in the immortal words of the great philosopher Robert Browning, all’s right with the world.

Al Lewis is the author of Why Nobody Believes the Numbers, co-author of Cracking Health CostsHow to Cut Your Company’s Health Costs and Provide Employees Better Care, and president of the Disease Management Purchasing Consortium.

17 replies »

  1. Many companies blog weekly as well as daily,
    meaning you need some great content to help keep your
    potential customers interested. Are you disgusted with
    each of the “generate income online” scams. A homemade scrapbook
    containing memories might also be wonderful.

  2. Thanks George. it turns out to be much harder to apply my “uniquely entertaining style” to a good-news story but I’m glad you feel I rose to the challenge.

    As for your observations and the others, all good ones, like most stuff I think the explanation is multifactorial so far. I’d also add that ICCC is nonprofit, meaning they can deploy more field resources for the same dollar. So it may indeed be that getting people to the right resources (primary care and others) — which are uniquely hard to come by in rural populations — was key. Can’t say yes or no yet.

  3. Thanks Al for a great article written in your uniquely entertaining style. However it seems to me that what the consortium delivered is what I would call good primary care. The role of the primary care provider includes health advocacy for their patient, although in Australia like the USA it is not something they are actually reimbursed for.
    It would appear that good disease management is really just good primary care. If you reimburse primary care providers to do it, and provide them with access to the relevant information via the right tools, you may actually be able to deliver it without developing a bunch of new entities to dilute the process and add expense.
    Maybe I am just being too idealistic, but I think most good doctors know what good medicine is,and will deliver it if we can get the artificial roadblocks out of the way.

  4. Al…appreciate your lifting up this program as an example of a population health program that worked. Organizing and coordinating the implementation of intervention strategies similar to this is our reason for being. ICCC loves doing this work and looks for opportunities to play these strategies forward across the country. Our nationwide Clinical Health Coach training programs grew out of our positive expereinces and results in population health programs. We believe that truly “inspiring personal accountibility” in participants or patients is the secret sauce for the future of healthcare. The positive results of this program work to help support this pivotal concept. Thanks from our Board, our staff and those that partnered to create and deploy this Medicaid Diabetes Telehomecare program…Bill

  5. Leslie…thanks for your questions. These are very good issues to address. Encourge you to review the report. We, of coure, did not conduct any comparative effectiveness tests of working with rural, low resource areas versus others. We simply defined a rural population. And, because they were Medicaid particpants, one may conclude that they were low income. We have conducted other studies and observed no apparent differences even when a similar care management coaching approach was employed.

    Several of our population health programs have engaged and actively involved physicians throughout the development and implementation of the program. Medicaid did not authorize that engagement in this particular program for defined reasons. However, we encouraged each Medicaid member who did not have a medical or health home to do that and expereinced progress toward that end in the treatment population.

    Based upon expereince, we have observations to share regarding the that are very positive related to engaing physicians and the results a good partnership may achieve. This is particularly true within an integrated healthcare system or an indivudal community. with statewide programs we have expereinced practical challenges.

    Hope tht his is helpful. Most willing to visit with you at any time regarding the program. Regards…Bill

  6. Barry…invite you obtain a copy of the report (access referenced in other responses). Yes, there is a quatification of participant expereince, clinical outcomes and cost/cost avoidance issues.

    Your question regarding the portability to an inner city population is something we would love to test. In an independent study ICC directed, a few short years ago, we found no greater utilization by urban adults(which may not exactly be the same as inner city) than by non-urban (rural) adults in a statewide study. In fact, non-urban adults reported seeing a greater number of health care providers on an annual basis than urban.

    There may be differences in characteristics and behaviors when contrasting a general low income inner city population and the tyipical Medicaid inner city population. And, you will be more knowledgable regarding that in your setting.

    Most willing to answer any remaining questions you may have. Regards…Bill

  7. Vic…thanks for highlighting a couple of key issues — program cost and transportability. ICCC has been organizing various population health programs for several years, often with very good results. We live for the opportunity to demonstrate how transportable and cost effective they can be.

    With good front end oragnizing and high quality user technology (in this case we used Pharos Innovations Tele-Assurance strategy) the labor intensity is mitigated allowing time for personal care management and coaching of participants.

    If one relies upon Al Lewis’ assessment of the program as a stipulation of validity, I encourage them to review the abstract, report and audit letter; review the outcomes in cost avodance, participant expereince and clinical outcomes; compare that to the actual program cost (which likely could be completed at a much lower cost in another replicated similar program); and, determine for themselves if the cost/value story makes sense for them. I sense that many ACOs, health plans, and state Medicaid programs may find it worth investigating further. Regards…Bill

  8. Susan…great questions which the report may very well address. A copy can be obtained at: http://iowaccc.com/diabetesevaluation/

    The report describes how the matched control groups were created including the criteria and describes a two year look back in control and intervention groups. It decribes more common outcomes measured around the Clinical Vlaue Compass. The event rate analysis is not detailed in the report as that may be a bit much for most readers, including me.

    Most willing to visit with you directly should you have addtional questions. Regards…Bill

  9. Dr. Pullen…encourage you to review the full report to found at:
    http://iowaccc.com/diabetesevaluation/

    When organized with the right partners, it may be the simplicity of the program that is most noticible. The abstract includes a quote that speaks to this issue. The technology leverage in this strategy and the feedback it provides, queues up clincal variances which allow the care manager time to address the most important care management and coaching needs. Thanks for the comments…Bill

  10. Great news and great post. Interesting that the case load was so much lower than in usual disease mgmt programs. I think the hypothesis that the program was more effective in a rural, low-resource area is quite plausible, but how can you tell if it’s that versus better attention from the case managers?

    At some point I hope to learn more about this successful program, esp in terms of how the case managers worked with the PCPs. My own personal experience with disease management has been that it felt more like hindrance than help, but that is anecdote and not generalizable data…

  11. Was there any quantification of net cost savings to the Medicaid program in the counties that implemented DM?

    I wonder if this approach would be as effective with large inner city low income populations where there are lower barriers to emergency room access, poor people often perceive emergency room care as better and more convenient for them than care from a clinic or primary care doctor’s office and the personal connection between a nurse and the patient might be harder to establish.

  12. Hi, Susan,

    Good questions all (and I would expect nothing less from you). The beauty of Medicaid is that THEY control for demographics etc. Eligibility is the same everywhere.

    Also since this was technically a difference-of-differences, unless there was a change in demographics over the period it would be valid. And because the other 4 diseases showed no relative change, it is unlikely that there was any change.

  13. This was indeed an extremely well designed and executed program, but there were so many interventions, from identifying depression to relatively personal case management that exactly reproducing it is unlikely in many other populations. Still nice to get a feeling that something can help.

  14. Al: thanks for this look at a DM program that works. For me, it shows the importance of concentrating on a specific target rather than trying to serve as a catch-all that can deal with a broad range of clinical issues. It does seem, however, that the program was quite labor intensive, which raises the twin issues of cost and transportability to other environments, especially in light of data published recently that coaching is not cost-effective (http://www.biomedcentral.com/1471-2261/13/33). Can you please comment on the cost and transportability issues, as you see them?

  15. Hi Al,
    Always nice to hear about positive outcomes! I am curious to learn more about their study. What was their control group? Did they control for demographic and other variables? What statistical methods did they use to arrive at their outcomes?
    Best,
    Susan

  16. Thank you for some good news in a medical world awash with incessant very bad news- almost to the point of speculation by sober data-driven leaders about imminent collapse of the entire US health care system.

    Dr. Rick Lippin
    Southamapton, Pa