So I spent the last couple of days at a disease management conference that focused on diabetes care.

There is general agreement that — at least 15 years since everyone has understood the problem — the health care system suffers from a lack of transparency, information systems, rational incentives, and care quality. Diabetes care is a microcosm of that. Type II Diabetes is a disease that’s primarily caused by years of poor living and poor care (obesity and metabolic syndrome being typical precursors). Once people get it less than 50% of them are correctly diagnosed, and after that the care of diabetics tends to be poor. Only around half get all the recommended tests and care that they need. And yet for a long time (since the DCCT trial back in the early 1990s) it’s been well known that regular monitoring of blood glucose levels can reduce the risks of further damage from diabetes. And those risks are nasty and expensive–blindness, limb amputations and heart disease. Getting diabetics to do all the things that they should do to reduce their dependence on glucose, and control their insulin levels is a great application of the education, monitoring and bullying that is modern disease management.

Disease management really started out as a front for drug company marketing so that they could pretend that they could work with PBMs and wrap services around their pills that would improve patient care. Of course they were also taken by the concept that disease management programs tend to suggest that sick people should take more drugs than they currently do. Of course some of those drugs might be generics….

But when you get beyond the high meaning rhetoric, disease management is complicated and confusing. Within the population with diabetes there are levels of illness, not to mention co-morbidities. Within disease management there are different ways of getting to patients (such as occasional mailers, phone calls, and constant monitoring via telemedicine). Once you get into the management of diabetics (or any other disease management program) it gets more complicated depending on who you are. Integrated systems want to control the costs of their sickest members; health plans typically want to sell value added services to their customers; and employers (and government) want to try to prevent the costs with their disease. But we live in a world where most diabetes disease management is developed for the less sick diabetic patient in a commercial population, while the greatest need — and potentially greatest savings — may be for a much sicker diabetic on Medicare or Medicaid.

But at a practical level, that all means that there is no clear focus on which patients to pursue. Should health plans be looking at their healthy commercial populations, or should they be ignoring them and going after the really sick people in their plans –who may be on their way into Medicare within a few years and give them no return? In the commercial world disease management services for diabetics cost something like $3 pmpm. Intervention using a telemedicine system (like the Health Buddy) can be around $50 pmpm. Obviously you need some pretty immediate savings if you are spending that much, and the VA at least seems to have decided that it is getting a return. But then again, Florida Medicaid in a rather biting criticism of Pfizer Health Solutions last year, felt that the returns from phone-based DM weren’t so great. But overall I came away from the conference no clearer on where on the financial graph the lines of the cost of intervention versus the value of the benefit intersect. And I’m not sure that anyone else really knew either.

What was interesting is how little was known about what the real ROI of different interventions on different types of people. One plan sent out postcards even though they believed them to be ineffectual because a drug company sponsored them. I mentioned to the people next to me that DM had gone full circle and was back to being drugcompany marketing. Even the phone calls may or may not be effective depending on their frequency and what was communicated in the call.

There’s an initiative in Tennessee, run by the Center for Evidence-Based Medicine at Vanderbilt in which the Blues are paying primary care docs to act as educational coaches for diabetics. This seems to be working (although it’s early days) and is having some good results, as are the folks at the VA with their nurse practitioner-led interventions and monitoring. But overall this is an industry that really doesn’t have its story straight as to what works consistently, and what’s worth paying for.

And of course while most payers don’t know if they can look forward to reaping the benefits of a costly intervention down the line, selling DM services will remain problematic. That’s why the Medicare CCIP demonstration projects about to take place are so important. The Medicare population is ground zero for DM especially for diabetics. Let’s hope that the CCIP experience tells us what DM can hope to achieve, and give us a level playing field on which to judge the value of the various interventions.