Cost, Choice, and Value

The Massachusetts Massacre has everyone stepping back a bit.  The President says that we should “coalesce around those elements of the package that people agree on,” but it is unclear just which elements those might be, given the extreme polarization that has defined the debate.  He suggests that points of agreement might center on insurance reform and cost containment, which are both important goals.  I’m skeptical that a sudden flowering of bipartisanship will allow such agreement, however.  Ezra Klein, on the other hand, has a paring proposal that goes in another direction, and reminds us of why we got into this in the first place: to extend coverage to the uninsured.  If we must narrow our focus, Klein says we should extend Medicare to those over 50, and expand Medicaid to those under 200% of poverty.  This would get lots of people insured, and could well be accomplished through budget reconciliation if no Congressional coalescing is to be had.

However the parsing, paring, and palavering goes, cost control is and will be at or near the health reform debate for years to come.  Two recent articles are worth a look for those interested in analysis of cost-containment strategies.

In his health care speech to Congress, the President suggested that one component of an effort to lower health care costs should be to empower a commission of “doctors and medical experts” to identify and,

encourage the adoption of  . . . common-sense best practices by doctors and medical professionals throughout the system. Wrapped up in that suggestion are notions of adhering to expert guidance in treatment decisions.

The stimulus bill passed in February pushed for scientific assessment of modes of care, providing $1.1 billion for comparative effectiveness research.  The current reform bills further emphasize CER, and would encourage the adoption of proven and promising treatments through professional education and some payment reform.  Harvard Medical School professor Jerome Groopman writes on evidence-based medicine in the latest New York Review of Books.   In his 2007 book, How Doctors Think, Groopman did a great job of explaining the complex and fraught process by which doctors make decisions, and he is fully on board with the notion that there is ample room for improvement.  His new article, however, cautions that the use of panels of experts with authority to impose or even recommend best practices is a dangerous way to go.

Groopman acknowledges the need for health policy folks to consider the bounded rationality of both doctor and patient.  He examines the Obama Administration’s policies on evidence-based practice by contrasting the views of two key advisors: Cass Sunstein, whose view of “libertarian paternalism” incline him to favor gentle “nudges” that may encourage certain behavior while leaving people free to reject the advice if they wish, and Peter Orszag, who is more inclined to employ forceful regulatory standards and financial incentives to achieve cost effective medical practice.  Groopman is compellingly  skeptical of expert claims of definitive standards on what “works” in health care, and cautions that such standards can result in harm to patients who fit uncomfortably into the hard categories defined in such best practices.

Groopman’s analysis seems incomplete for two closely intertwined reasons, and surely as a result of space constraints.  First, he suggests that the administration is faced with a stark choice between

aggressively pushing doctors and patients to do what the government defines as best, or [being] respectful of their own autonomy in making decisions.

Surely there is much middle ground between tying doctors’ hands and respecting complete clinical independence.  And it is not enough to say, as does Groopman, that

Most physicians seek data and views on treatments from peers and, as needed, specialists, and then present information and opinion to patients who ultimately decide.

Maybe so, but physicians are sometimes self-interested, and patients’ choices  are sometimes influenced by advertisements or other considerations disconnected from quality concerns.  For these and other reasons, spending decisions are no longer consigned to the doctor/patient dyad, but increasingly must accommodate the cost-containment interests of third party payers — government, employers, or insurers.

Second, Groopman describes two exclusive categories of procedures: “mechanical procedures” such as the  insertion an intravenous catheter (where he argues that enforcing standards to avoid infections is proper) and all other procedures, where the individual patient’s condition becomes relevant, and where he argues that coercing clinical choices is out of bounds.  It is not obvious that the universe of procedures is so divisible; it is even less clear that the dividing line between the two categories is uncontroversial.

Many questions remain.  Groopman is surely right that we must be cautious in enforcing categorical “best practices;” it is important to create public processes for vetting their accuracy and usefulness.  He is also surely right that public and private health finance rules must accommodate variation in medical needs, and must bend readily when a “best practice” is not suitable for a particular case.  But cost is relevant, and encouraging efficient practice can reduce the cost (and therefore the extent) of coverage.

So, how might a balance between financial constraints and patient protection work?  In a Health Affairs article posted  yesterday, Michael Chernew and coauthors examine the growing phenomenon of “value-based insurance” — a structuring of insurance co-payments responsive to the needs of people with chronic illness.  The co-payments imposed by insurers are, of course, intended to reduce demand for health care services (an Orszag, not a Sunstein tool, you might say).  Value based insurance reduces or eliminates these co-payments for services of “high clinical value.”  That is, if an insurer determines that it would rather not discourage utilization for a particular service, it reduces or removes the patient cost-sharing, presumably increasing usage, for cost as well as clinical reasons.  As the authors explain,

The belief that a value-based insurance program will lower health care spending rests on the recognition that the use of high-value health care services reduces the probability of adverse events related to chronic disease and that on a population basis, these events are much more costly than the services aimed at preventing them.

The authors found some evidence that such programs are cost effective, even in the narrow sense of reducing a plan’s health care expenditures.  They suggest that widening the economic lens to consider broader societal goals would only strengthen those conclusions.

The article acknowledges the reality of economic coercion in the clinical setting, and measures attempts to shape the tools of cost containment in a way that protects patients while maintaining cost containment.  One doesn’t have to accept the general wisdom of patient cost-sharing to value attempts to protect patients from untoward effects of its use.

The need to obtain “value” for health care spending and to take steps to restrain health inflation will persist however we come out of the current reform debate.  The discussion will benefit from both the erudite analysis of Groopman and others warning us away from answers that are too easy, and that of Chernew and others who can shine a light on the efficacy of particular cost containing measures.

Mammography, Cervical Cytology Screens, and Rationing

The recent recommendations on mammography and cervical cytology screens by the US Preventive Services Task force and ACOG (American College of Obstetricians and Gynecologists), respectively, have added a new dimension to reform discussions.  Some are inclined to say “gotcha,” suggesting that the recommendations are evidence of a creeping denial of needed care that would follow governmental insinuation into health finance and benefits design.  Others see the reports as serendipitous irrelevancies, unconnected to reform discussions.  The truth is, not surprisingly, more complex.  The irony is that consumers will be more represented in health technology assessment in a public plan than they have been in private insurance.

It seems inevitable that any future health finance system will rely on evidence-based assessments of new (and old) technologies for both quality and cost purposes.  Our experience with the widespread use of affirmatively harmful (e.g., hormone-replacement therapy) and apparently useless treatments (e.g., knee arthroscopy for osteoarthritis) points to the possible risks of rapid  or uncritical adoption of new technologies.  As Sara Rosenbaum and others have pointed out,  (subscription required) we don’t want to confuse population data with individually-applied diagnostic and treatment judgment.   Both reports, to their credit, got this part right, and advised individual patients and physicians to assess each case in context, notwithstanding the general population-level guidance.  But evidence-based population data on the efficacy and comparative benefit of new and expensive interventions will be of enormous assistance in future treatment and funding decisions.

How should such health technology assessment be done, if not by expert panels?  As Bill Sage has observed, private health plans were opaque and inconsistent when they were in the technology assessment business.  (They have pretty much gotten out of that field, leaving cost control to others.)   One criticism of the mammography and cervical cytology reports has been that they should have included a more public process before issuing recommendations.  As the reports are merely advisory, it is not clear that post-publication comment doesn’t get the job done.  Where, as may be the case in the future, such expert analysis has instrumental effect, pre-implementation public process is essential.   Two guides for public health technology assessment advise as much.  The Institute of Medicine, in guidance issued earlier this year for comparative effectiveness analysis funded by the stimulus bill, observed that,

Clinicians and patients do not always consider the same factors when weighing the tradeoffs posed by important health care alternatives.  To ensure that the fruits of CER support consumers’ health care decision making, the CER Program should focus on the questions of patients as well as their health care providers.

Similarly, a health technology assessment guide created by the European Observatory on Health Systems in 2008 describes well-functioning technical assessment as consultative and transparent:

Social accountability permeates the whole knowledge production and is reflected not only in the interpretation and diffusion of results but also in the definition of the problem and the setting of research priorities.

We don’t want a health system — public or private — that is blind to either sound evidence-based technology assessment or the particular health needs of individual patients.  One advantage to a public system is that the assessment of technologies can and should include a robust public process.  We didn’t get that with private managed care.  The mammography and cervical cytology reports should call our attention to the opportunity for public process in decision making in publicly-funded coverage, and the need for close attention to the implementing regulatory processes if and when a bill is signed.

Here’s an Idea: Asking Doctors about Health Care Reform

The New York Times just published a very interesting article that ties the efforts of the medical community to bring about change in the American health care system with Congress’s attempts to reform health care through legislation.  The article, which details the research of a team of health care providers in the Intermountain Healthcare system in Utah and Idaho, offers insight into what doctors are doing on their own to effect change while waiting for our nation’s leaders to implement the means to better health care for Americans.

As can be seen by American Medical Association’s recent endorsement of the Democratic House bill, and the long time call of the National Physician’s Alliance for reform, there is a consensus among  health care providers for health care reform.

Of course, essential in that reform is delivery system reform. Part of delivery reform is likely to emphasize not only preventive care, a cornerstone of Obama’s plan, but also a careful monitoring and consideration of the outcomes of health care practices.  Although there is debate about the best way to monitor and measure such practices, and some bridle at the prospect of being “confined” to protocols derived from large studies,  the evidence-based medicine model is emerging  as a favored tool with which to analyze how health care providers themselves can produce more cost-effective, life-preserving results. Evidence-based medicine puts protocols in place (which may be overridden at a doctor’s discretion) and relies heavily on the statistical analyses of a health care system’s performance (i.e., patient outcomes from particular practices).  Such is the model executed by the Intermountain Healthcare system highlighted in the Times article.

The protocols ultimately implemented sometimes differ from the usual course of treatment offered by some doctors. The physicians at Intermountain Healthcare admit that it is often hard for doctors to hear that they are doing something wrong– or perhaps “not optimally” would be a better choice of words.  The Executive Director of Intermountain Healthcare Institute for Healthcare Delivery and Research, Brent James, relates that some doctors do not believe the results of the statistical research because doctors are reluctant to change their ways, but that oftentimes when presented with clear statistical evidence doctors change their practices.  He gives the example of obstetricians who were performing elective inductions prior to 39 weeks for pregnant women for the sake of convenience, as the inductions save hours of labor for the mothers and therefore hours of hospital time.  However, an analysis showed that babies born prior to the 39th week of gestation were far more likely to wind up in intensive care. After doctors saw the data, and protocols were put in place, James found that the rate of elective inductions fell dramatically. A similar protocol developed for the treatment of one form of pneumonia was said to have cut the rate of death for that condition by 40% over several years.

Some doctors contend, however, that the medical metrics of evidence-based models are not the best way to bring change in health care practice, both because doctors will feel pressured to follow set protocols without considering other possible treatments and because humans are not statistical data that can be remedied through calculations and formulas. The danger, of course, is in negating the healing art– in throwing the proverbial baby– independent critical thought– out with the bathwater.  Doctors of this school of thought often espouse  revamped medical education as a better way to reform health care practices; after all, the basis of how health care providers develop their practices is the way in which they were/are taught.

And one wonders if there isn’t room for both approaches. If the education of medical students can be changed to incorporate better and cost effective practices based on studied outcomes (perhaps in part culled from the Health IT initiatives), and changed to incorporate greater emphasis on preventive care (coupled of course with a pay system which rewards patient wellness), while still respecting doctor autonomy so as not to prepare a generation of medical robots. It doesn’t sound “un-doable.”

Interestingly enough, medical schools have seen an increase in students applying to their programs.  In response, four new American medical schools have opened.  With the older generation of health care practitioners on its way to retirement, the need for more doctors is imminent.  But, we need doctors that are able to help carry the new ideals and practices of a reformed health care system; reaching into the med school curriculum would seem to make a lot of sense.