Obama Asks Bloggers For Help With Health Care System Overhaul
Show me the quadrants

The politics of healthcare reform are dizzying. I, and probably most readers in the HCIT industry, have been focused on how to achieve goodness as defined by the ARRA-HITECH definitions. We're also mindful that significant payment reform, if it happens, will clearly lag our new cost implications. As much as we might want to be noble, a persistent negative cash flow is discordant with keeping the lights on.

But those issues have fallen off the public and media radar. They have been replaced by large policy issue, “Who is going to pay for healthcare reform?”

As if these issues are separate (clearly they are not), what do we get for our newly allocated dollars? Does this lead to new net costs or new net savings? Douglas Elmendorf of the CBO concluded, " the legislation significantly expands the federal responsibility for health-care costs." And that is consistent with many other CBO analyses. With limited dollars and a portfolio of spending opportunities, where is the dialogue on the economic impact of ARRA-HITECH? Not the rhetoric. Where is the empirically informed, quantitative modeled decision matrix?

What if the ARRA-HITECH Stimulus dollars led to 38 million fewer heart attacks through the use of EHRs? Suppose that led to downstream net savings over three years that dwarfed the new costs?

What if we had data that was convincing to the Congressional Budget Office on explicitly which initiative, Cholesterol or Blood Pressure management led to that net savings? And, over what timeframe? And, which one of the two was more likely to actually increased the national rate of healthcare expenditures, given what the Government knows about Medicare beneficiaries today and existing guidelines, with and without projections to 2025 and beyond?

Well, at the same conference that David Blumenthal spoke after the 2008 presidential election, the 2008 AMIA meeting, another doctor gave the plenary speech and addressed exactly those questions. That other doctor and his collaborators have studied these implications of guidelines. They have applied these modeling tools with payers and care delivery organizations. They've worked with several notable organizations that have substantially similar challenges being addressed by the Obama health-care plan. Those organizations include Kaiser-Permanente of Hawaii, the CDC, and an employer-based, white collar, 30,000 member corporate health plan. Robert Wood Johnson Foundation has funded over 15 million dollars to using their model-based approach. All of these organizations were intensely concerned with economics, that is, how to effectively spend money to save money on healthcare. With attention to quality, of course.

I'm talking about Dr. David Eddy's talk concerning the Archimedes model. There are a number of articles and recorded presentations by him and others that detail the econometrics that are clearly integral to using EHR technology effectively to save money and lives. And, per the first paragraph, where we're likely to achieve financial savings over the next five to ten years. And, where we're not.

There are other models that have relevant predictive power and utility other than Archimedes. The point here is not which model or modeler. The point is that EHRs are only a hammer, not a house. Healthcare legislation that focuses on nuances of the hammers is not going to get us the house that's being envisioned. That vision will be exemplified and evaluated using models like Archimedes.

In the example provided by Eddy at AMIA, Cholesterol management had improved quality of care, combined with net savings, for the population the model was applied to; blood pressure management had a clear net increased cost. Cholesterol management produces a quadrant one result in the Quality versus Cost matrix above, for the study population Archimedes was applied to in Eddy's example. If that result applied to relevant US populations, Where should the government direct the definition of meaningful use as a result?

Eddy also pointed out that it's possible to go even further toward the top left, i.e. more cost savings and higher quality, by focusing on higher risk people. Can we achieve such personalized guidelines and benefits in a practical way in the paper world? Are these savings the stuff the EHRs are being touted as delivering?

So, Mr Obama, thank you for inviting me to blog about how important it is for all of us to seize the day. I am undecided on the issue of who should pay. I am very decided, however, that it's possible, and in fact likely, that costs and quality will not improve without the kind of modeling described here. And it's resultant transparency. Let's not confuse quadrant one and four in the diagram above.

The immediate opportunity in front of us to use HCIT to reduce cardiovascular disease by 50-60% with a concurrent cost savings to our government and healthcare system. The models to get there have decades of work behind them and were essentially non-existent last time massive healthcare reform was attempted. I'm hopeful that our definitions of meaningful use are both informed by models like Archimedes, and that we'll publish our economic and quality assumptions, expectations, and annualized performance. I don't see any other scientifically defensible basis to undertake any costly initiative for which no one can be certain. And there's no way to protect good legislation from political defeat without a compelling story that's better than "no decision" in the eyes of the voters and the CBO.