Health Populi

Thursday, November 8, 2007

Using evidence-based medicine to lower health costs - a view from the CBO, and HR 2184

Evidence-based medicine (EBM) can reduce health spending, according to two analysts at the Congressional Budget Office. Note that there's a piece of legislation in Congress to fund EBM research.

The very level-headed essay in the November 8th issue of the New England Journal of Medicine explains the positive role that EBM could play in preventing Medicare and Medicaid from consuming 20% of the GDP by 2050.

Peter Orszag and Philip Ellis, the CBO's Director and senior analyst, respectively, talk about the merits of marrying EBM -- which generates information about the effectiveness of medical treatments -- with rewarding clinicians when they adopt proven, effective health care practices.


In today's less-than-transparent and under-IT'd practice environment, Orszag and Ellis estimate that less than 50% of all medical care in the US is based on evidence -- conversely, that most health care delivered in the US is not based on "firm evidence of effectiveness."

Furthermore, reimbursement incentives can subtly drive physicians to prescribe and use treatments that, at the margin, provide greater compensation -- thus often motivating doctors to take the expensive route, which may not necessarily be the most effective course of treatment.


The CBO team argues that expanding research into comparative effectiveness would go a long way toward aligning patient and physician incentives and eventually reducing costs in the system. This would occur when payers (whether Federal/public or commercial/private sector) pay physicians based on evidence.

To drive down costs even further, payers could require a payment by consumers who wish to receive the more expensive, less effective treatment -- thus putting the patient into a true consumer role by having to decide: "is the added expense I would bear really worth the expected benefit I would experience?"

In May 2007, Congressman Thomas Allen (D) and Congresswoman Jo Ann Emerson (R) introduced H.R. 2184, which attempts to improve the effectiveness and reduce the cost of health care. The bill will allocate about $3 billion in funding over five years for comparative effectiveness studies conducted by the Agency for Healthcare Research and Quality (AHRQ). Drugs, biologics, medical devices and other therapies will be evaluated against other available treatments with regard to effectiveness, safety and cost. Research is expected to be funded by the Federal government, health insurers and large self-insured employers.

According to Fitch Ratings, the passage of this bill would have little impact on the health care industry in the near term but, in long run, would have "a significant impact on the health care industry."

Health Populi's Hot Points: Among the various approaches bandied about by health reformers, the EBM proposal is rational, dispassionate, and sensible. Furthermore, it is do-able -- if we allocate, today, the resources required to do large-scale research into effectiveness. There are a lot of 'ifs' around this -- such as what entities should do the research, and who finances said research. But considering the prospect of Medicare and Medicaid consuming 20% of the GDP, which is the current share of GDP accounted for by the entire Federal budget -- we can't afford not to make this investment.

1 Comments:

  • Ms. Sarasohn-Kahn,

    Thank you for broaching this topic. I agree, effectiveness research could lead to great strides in efficacy and efficiency. However, for effectiveness research to become a reality, at least two significant paradigms will need to be shifted or replaced.

    First, the tacit knowledge of physicians must be made explicit. That is, doctors must distill their medical knowledge into algorithms for specific conditions, wherever possible. Many conditions may be treated with 'rules-based' care. The reason more treatments are not expressed as algorithms has to do with the free-agent model of the medical profession, more so than the complexity of care.

    Those who suggest that their specialized knowledge is irreducible are making the Luddite's argument. Technology/innovation create a flow of ability from the highly-specialized and theoretical to the broadly-generalized and practical. We can easily collect subjective/objective outcomes data, already. If we began codifying the nature of treatments, effectiveness research could become quite a straightforward exercise.

    Here's how we've expressed treatment algorithms where I work:

    1. Tx - Define the physical intervention as rules or steps. (We can use CPT codes, for now. However, I believe this is an artificial and restrictive taxonomy.)
    2. Rx - Define the chemical intervention.
    3. Ix - Define the information therapy as 'key concepts' the patient should understand.


    Expressed as an algorithm, these treatments can be pushed down to extenders, and perhaps eventually become self-care.

    Physicians are highly-trained and have definitely earned the right to practice medicine as they see fit, but not without knowing the results. It is incumbent upon physicians to know the outcomes of the nostrums they proffer. However, the medical community has largely outsourced quality control to the tort system. Changing this entrenched mindset within our healthcare system may be the hardest part of successful reform.

    Secondly, we will need to change who we consider to be researchers. To use Clayton Christensen's Theories of Innovation parlance, our current research community has 'overshot' the needs of those using the research data. The randomized control trial is the 'gold standard', but what it gains us in statistical power, it takes from us in time, cost and specificity. Is the delay between research finding and clinical practice still 17 years? I think we definitely want an RCT when bringing a new molecule into our pharmaceutical arsenal, but with today's information technology, we could run good prospective studies around episodes of care, right now.

    If clinicians expressed their care algorithmically, and collected outcomes data, we'd have an army of clinical researchers and our body of medical knowledge would grow every time you went to your doctor. If we pushed treatment protocols out to extenders, the army would quadruple. What a powerful system! Imagine a centralized database of treatment algorithms and outcomes data. I start to see truly personalized care when I think on this scale.

    Now my head hurts.

    Thanks for letting me borrow your venue.

    Ben

    By Blogger MedManager, At November 8, 2007 11:19 AM  

Post a Comment



<< Home