• Health care typology and efficiency

    In my latest post on the JAMA Forum I offer a typology of health care that clarifies when various types of cost control are appropriate. The typology is summarized with this chart, which the post explains. (Its structural, though not chromatic, bilateral symmetry reminds me of a butterfly. Could be just me.)

    I freely admit the chart and typology are not perfect. It’s pretty hard to simplify health care down to something that can be expressed in a readily-grasped chart and a 1,000-word post. On-and-off, I spent months on this chart, getting feedback from the smart people I acknowledge in my post. The earlier versions were much more complex, and far less clarifying. In the spirit of “all models are wrong, but some are useful,” I assert that this is of some value. Read the post, and let me know if you agree. If not, why not?

    By the way (#1), more and more I am disliking the term “cost control” since, in my view, it’s a doubly misunderstood term. People usually use “cost control” when they mean “spending reduction” or “spending growth moderation.” I think the emphasis should be on efficiency, not spending or its growth per se. But we don’t have a nice catch phrase in health policy discourse that unambiguously evokes an efficiency focus. That’s a shame.

    The second level of misunderstanding is an ironic one. “Cost control,” with a literal interpretation of what “cost” is, is actually closer to “efficiency enhancement” than some may realize. We should want to maximize health (output) per unit of real resources (input). This is a measure of cost. That is, we should want an increase in efficiency though a reduction in cost at any level of spending, but I think too few people understand this. “Cost control” ends up being an overloaded term to which people yoke whatever collection of ideals for and understandings of health care they harbor. That makes it useless for precise communication. I hope my typology and discussion thereof helps clarify some of the nuances, at least for the six of you who read it. (Come on, spread the word, and make it 12.)

    By the way (#2), in response to my “I (partially) quit” post, I was dismayed to hear that some people weren’t paying as close attention to our (mine and Aaron’s) JAMA Forum and AcademyHealth posts as they do to those on TIE. That’s a mistake. Our JAMA and AcademyHealth posts are among our most developed and highest quality posts. They’re worth reading. You’ll find all of TIE’s AcademyHealth posts here. You’ll find all of the JAMA Forum posts here, even those by non-TIE authors. Go read them (or at least those by me and Aaron, though others are good to), staring with my latest.


    • The devil is in the details.

      How do you define what’s effective? Is a two month median improvement in survival a sign that a drug is working? At a cost of $100,000/year? What is the benchmark for evaluating this (QALY)? What can we afford to spend? Should we have a budget for health care spending?

      I’m offended by any drug costing more/year than the median income of citizens/year. Particularly since most of these new drugs are marginally effective. By inspection this is unsustainable pricing and one way or another should end.


    • As I read, my thoughts drifted to an old mentor:

      “Doing nothing, is doing something.”

      There is never one treatment option. There are two. If there is a patient or surrogate, unless life hangs in the immediate balance, there is a shared decision opportunity.


    • I’m not sure I understand when there is ever one treatment option. OK, if we are positive nothing can be done, then we are down to one choice. But if there is a treatment available then there are two options because “do nothing” or “watch and wait till more is known” is always on the table.

      If someone comes in with strep throat and I say “We are 100% positive it is strep throat. We should go with antibiotics to reduce your discomfort, transmission risk and dangerous complications such as rheumatic fever.” there still needs to be shared decision making about whether to go through the trouble of treatment at all. Financial costs, time costs, antibiotic resistance costs might lead to a reasonable decision of “watch and wait” even though ANTIBIOTICS = THE ONE CURE.

      So what say you: are there any diseases that don’t qualify for shared decision making taking into account a “more than one option” threshold is intrinsically present?

      • Viewing “no treatment” as an option doesn’t change anything. It’s a choice, so the patient should be involved. Sometimes, however, there isn’t an option other than death, basically. Show me someone who encounters the health system and opts not to have her fractured hip repaired or any doctor who would view no repair as a viable option. Even if it has ever happened, it’s such a tiny minority it’s not worth worrying about in a model at this level of abstraction.

    • I have reservations about hip repair. Folks with moderate to severe dementia cannot participate in physical therapy, pull bandages and contaminate wounds, and dont follow instructions well (“you cant get up and go to the bathroom without assist”). In cases like this, I have not sent folks to the OR, but to hospice. The surrogate presents clear and convincing evidence patient would not want the surgery. The decision is not easy however; you can imagine the gray zones. Also, its not so much the surgery, but in whom the problem occurs: old, frail folks.

      Situations like acute renal failure in need of urgent dialysis (uremia and fluid overload), sepsis with leg gangrene from diabetes (amputation), bacterial meningitis, and acute aortic dissection are more apropos. Literally, they are do or die.

      Hope this helps