“Effectiveness” is at the forefront of the health policy debate. Effectiveness is the assessment of whether any particular medical intervention actually advantages patients when prescribed in practice settings. To be considered effective, the intervention must result in a clinically meaningful improvement for an adequate percentage of patients. Furthermore, it must not result in a clinically important adverse outcome in too many. Clearly, “effectiveness” is a value-laden construct. How is “meaningful improvement” defined and by whom? How is “important adverse outcome” defined and by whom? How are “adequate percentage” and “too many” defined and by whom?
“Cost-effectiveness” is even more value-laden. It is legislated to be off-the-table in the machinations of the Affordable Care Act for reasons that vary from fear of rationing to fear of compromising profit margins. But no one can exclude cost-effectiveness from the patient-doctor dialogue. Considerations of co-pays and deductibles often weigh heavily in the valuation of interventions.
The greatest advance in clinical medicine in my time in the practice, fast approaching 50 years, is that today patients and their doctors can assess effectiveness as collaborators. No longer does an imperious pronouncement by a physician suffice. Rather, the patient should occupy the driver’s seat with the physician as navigator. For each option in intervention, the patient asks, “Based on the available science, what is the best I can expect?”
For nearly 50 years, no prescription drug could be marketed unless the FDA was convinced that it had a tolerable benefit-to-risk ratio based on scientific studies. The bar for devices (like hip replacements) and procedures (like liposuction) is not as high, but there is usually some informative clinical science of this nature. The science is generally designed in the hope of demonstrating a favorable benefit-to-risk ration. Hence, patients and interventions are chosen to measure outcomes in the best case. However, make no mistake; neither the fact of FDA approval nor common practice is an adequate response to “What is the best I can expect?” If the best case falls short in your mind, why would you acquiesce to the intervention?
Consider Pill A, which reduces a blood chemical known to predispose to heart damage. The “best case” science says that if 100 adults take the pill for 5 years, perhaps 2 will be spared non-fatal heart damage and 10 will suffer muscle pain. However there is no suggestion that anyone will be spared death before their time. Would you take Pill A? Even if it was free? Consider Pill B, which reduces a blood chemical associated with damage in many body organs and with death before one’s time. Scientific studies of Pill B, some lasting a decade, have proved disappointing both in terms of preventing body damage and causing toxicity. But these studies were done on early versions of this category of pharmaceutical. Would you take a new version or a combination of these agents?
Consider Procedures A and B, both designed to overcome pathology in blood vessels that is associated with organ damage. Procedure A is “minimally invasive”, though it has a 1-2% incidence of catastrophic complications. Procedure B is major surgery with a 5-10% incidence of major complications related to the procedure and a 20-30% incidence of a prolonged period of recovery. Again there is a robust scientific literature saying that neither intervention will improve longevity. The literature that either improves symptoms is highly inconsistent. Nonetheless, these procedures are still offered because they are performed differently today than in the older disappointing studies and they are offered only to candidates thought to be particularly appropriate. If you were deemed “appropriate” would you acquiesce to either procedure?
Pill A is a “statin” for high cholesterol. Pill B is an oral hypoglycemic to lower blood sugar in Type 2 diabetes. Procedure A is stenting for plaques in coronary arteries (or elsewhere) and B is bypass grafting. All are plausible, even sensible, even current common sense. All are sanctioned by august professional bodies and all are covered by health insurance. There is no “right or wrong” answer to the question about patients in general acquiescing to these options. But there is a right answer for you.
Many are capable of rendering the best case science transparent; there are even independent groups of scientists devoted to this exercise. And many more are capable of prescribing and performing according to guidelines and standards. But none of these people can come up with the right answer for you.
That’s why you would be advantaged by a physician who wants to support your valuation of the options. That’s the way you’ll find the shoe that fits.
Categories: Uncategorized
Personally, I think the pair of you are splitting far too may infinitives!
Historians will forever wonder why the English language could not generate a gender neutral pronoun or a gender neutral curmudgeon!
Love the post. Hate the title since it is ungrammatical and irritating to those of us who care about the English language. Since THCB often writes headlines without checking with the author first, Dr. Hadler may not be responsible. As a curmudgeonly grammarian for today, may I suggest several grammatically correct alternatives:
What would you do if you were I? (exactly the same as: This is what I would do if I were you.)
What would you do in my position? (or “if you were in my shoes”)
Or simply use another title. It’s a shame to have a bad headline at the top of an outstanding post.
For more about the flaws in quality metrics, see also:
http://wp.me/p2bC3h-g7
Dear Karen,
Thanks for the praise of the content. I will not apologize for the title. As another “curmudgeonly grammarian,” I am also fascinated by conditionals. I would argue that the use of the subjunctive voice casts the conditional as a hypothetical and renders “me” permissible.
While we’re having fun with with grammar, it can be argued that “curmudgeon” is gender specific. In that case, only I am the “curmudgeonly grammarian.” You will have to settle for “astute grammarian.”
Have a great 4th.
Nortin
Love the post and insightful comments; thank you.
I left a solo oncology practice to learn how to help patients make their own decisions. After training in clinical-decision-analysis I began an informed-decision-making consult service. I have learned the following: patients are really smart about data.
If we are transparent with them about the precarious nature of the present methods and outcomes of science and take the time needed to help them understand the transparency, they figure out what to do. There is a shift in the decisions they make after being informed — not only of the numbers but, also, how the numbers came about. This is an important addition to informed choice; not just numbers, but how numbers get counted. Patients are wonderful evidence critics.
Turning patients loose as sole decision makers will lead them to tell us what we are worth, and a useful economic plan may follow. It will not happen the other way around. I wonder if we are afraid to turn over decisions as we worry patients may find the value not worth the dollars that now support an indulgent health-care system?
It is a struggle to superimpose any degree of rationality on a “health care” system that is ethically bankrupt yet mired in the entropy of an enormous transfer of wealth and supported by a populace that has long been misled. We missed the opportunity for rational health care by 60 years. Now it’s a Sisyphean endeavor.
I’ve modeled, honed and serially published an approach that pivots on a fiduciary responsibility for cost-effectiveness and still fosters patient empowerment. The most recent print version is a chapter in Citizen Patient. We came close to a pilot demonstration in one of the states several years ago, but there was a push-back even though we’d exclusively indemnify uninsured workers.
Sisyphean or not, we must keep trying. The current approach is not sustainable. Maybe we can inform the Phoenix.
Besides, as I said, I like jousting with windmills.
I am a huge admirer of Dr. Hadler….he and his perspective should be more prominent than Dr. Oz’s! He should be a regular commentator on all the major networks! However, I have a different perspective on fee for service. A few points:
1. Fee for service works poorly now….but that is mostly due to the fact that fee for service cannot work well when there is a third party payment system….insulating the consumer of the service from payment always screws things up and leads to over-consumption.
2. Thus, the way to make fee for service work better is to have the consumer of the service benefit from shopping and financially share the savings from declining service or choosing carefully.
3. The current fad is that accountable care organizations get paid a flat rate per patient…capitation in one form or another. Then the problem will be the patient won’t know whether the recommendation not to have a knee procedure is because it isn’t right for him/her, or whether the doc and ACO system is looking to enhance their profits…..not a big improvement on fee for service in my opinion.
4. We will need a system where patients can chose to participate in a fee for service plan or choose some form of capitation plan. In the legal profession one can choose fee for service or to buy some form of insured plan where you pay an annual premium.
5. Remember, fee for service has been the dominant model since trade and exchange between humans began. It is not inherently bad. In fact, I bet most readers of this use it for their legal services. Yes, your attorney would like to get you to spend lots of money on multiple services….yet in most cases your attorney won’t oversell as he wants to earn your trust and keep you as a customer over the long haul…..and should you think he is overselling you probably will get a second opinion.
Perhaps my view is misguided….. I look forward to comments that point out where I have gotten this wrong. Thanks.
While fee for service is historically the dominant model for trade, I would disagree that we have not evolved from that standpoint. Physicians are one of the few professions that still use the fee for service model. Lawyers bill by the hour, not by the procedure. If I go to the doctor and they spend 15 minutes of their time, but order tests and in office medications they can make both their time based fee plus orders of magnitudes more for the medications and tests. Many physicians make 50-75% of their income from ancillary services and/or infusing medications. Pay physicians for their time and their time will be well spent. Add to this the inability for a patient to know or be able to know if something is needed and the expensive–read ICU treatment–care is at a time that emotions frequently do not allow for a consultation with consumer reports. See my blog: http://www.lakesidemedicalmusings.com
Dr. Scherr,
Thank you for the point about most lawyers charge by the hour. I have considered the point. Charging by the hour is probably the simplist and clearest type of fee for service. All the points I made above apply whether your lawyer charges you for three hours for a will (for example) or just charges you a flat rate for the will.
Most commentators state our medical system is based on fee for service. My point is that we have a wildly adulterated form of fee for service where an essential element has been removed: the consumer (patient) pays nothing extra (in the short term) for wild overconsumption and assumes more is better (Dr. Hadler clearly points out that this causes great harm to patients).
We have experience that shows that removing this adulteration has salutary effects…reducing wasteful and harmful over-consumption. Catastrophic plans linked to health savings accounts have worked very well and are highly valued by most who have them….people like being in the drivers seat and benefitting….health wise and financially….by wise consumption of health care services. Moving away from fee for service means control flows to government and insurance “experts” who try to tweak the system by imposing new rules for doctors and patients leading to many unintended harmful side effects.
Paul
I guess where we differ is in our feelings about who is best to control the usage of money. I guess my feeling is that the consumer as demonstrated by their love of unproven dietary supplements, weight loss fads and susceptibility to advertising is not the best judge of complex medical decisions. I also think that the big costs in medical care, hospitalization, especially in the last 6 months of life are well beyond the abilities of most patients to be more than lead to a “informed decision”. What would you propose is the copay for a chemotherapy that costs $100,00 for the treatment–one that does not deter appropriate use or limit to wealthy only. What about dialysis? What should the copayment be for this. When you talk of the individual paying for a doctor visit for a cold that is one thing, but expensive treatments that actually is where the bulk of our dollars are spent is another. My big concern for fee for service is not paying for the service of medical care, your doctor’s time, but rather for paying for add ons that generate huge revenue for the physician and DOES affect a physician’s judgement as far as treatment options. As an aside, I was somewhat agast when I went for an eye exam and my opthamologist is selling a diet supplement and diet program. What is up with that. I agree people like being in the drivers seat, but to be honest it is an illusion that they are. If I was to explain to you two chemotherapy options for a cancer you had, would you be able to decide withouth the physician guiding you as to which one you should take? Would most people? See my blog and you will see how I came to these conclusions, http://www.lakesidemedicalmusings.com
I did look at your blog. Respectfully, one area where I think you are off is your assertion that other professions have “evolved” past fee for service. Using one of your examples, if you go to an accounting firm to do your taxes it is “fee for service” even if you are charged by the hour even if the person who does it is a salaried employee. The provider organization still is motivated to charge your the highest hourly rate possible and they still would love to sell you additional services. Ditto for lawyers, architects.
My point is that the villification of fee for service is misplaced….fee for service works fine if the customer (patient) has a financial motivation to use the services prudently….even complicated services. The historical example is the shift from defined pension plans to defined contribution plans…..the “experts” thought it was folly to give employees the power to manage their own money. What happened is that the financial services industry evolved to provide guidance. It also innovated by creating low cost target retirement retirement funds…..and the costs of pension administration dropped dramatically.
We do agree that fee for service is not working well right now…. with patients perceiving each service as no cost or low cost folks will wildly overconsume and docs and hospitals will over treat….in part to avoid malpractice complaints and in part meet patient expectations who interpret more procedures as caring about them.
Paul,
I certainly am not going to stick up for lawyers billing practices, but padding a bill with extra hours, to me, is a whole different ball park. Let’s say your lawyer charges you $1000/hr and add 2 additional hours to your bill for a project that should have taken 5 hours. That is a cost of $7,000 vs what should have been $5000. Yes, $1000/hr is outrageous, but certainly out there. As a consumer, I could not use him in the future because he was too expensive and I was paying for this–your exact point.
Now let’s take an example of an oncologist who recommends that you undergo a treatment for your cancer. His 30 minute consultation is let’s say $400– probably a little high, but I will go with that. The cost of the treatment that he recommends is $50,000–unfortunately not an unreasonable cost. This does not include the hospitalizations….. that you will need on top of the chemotherapy. You have cancer, are emotionally distraut and want to get your treatment started as soon as possible. Whether this treatment cost $50,000 or $100,000 most people could not afford this without insurance. And the number of physicians in most communities that can give this service is small and the cost differential betweem them is little if any. I can’t say for sure what the physician will earn from the $50K, but it probably is around $10K. There is a domino effect to the system for the physician being paid to give treatments. And the money from the treatment far exceeds the hourly rate the physician bills. That has a very different societal affect than padding your hours. Thanks for taking the time to read my blog: http://www.lakesidemedicalmusings.com
I agree with John Scherr that “fee for service” is intrinsically conflictual. Last year I was asked to keynote NOLC (the National Orthopedic Leadership Conference) on the eve of their annual lobbying exercise on behalf of AAOS. The topic; “The Future of Fee-for-Service Medicine.” I argued that there was no future unless remuneration was on a Fee-for-Serving basis; an orthopedist was as valuable for explaining why knee surgery was not needed as for performing the procedure. Then we can put a $ on the “value”.
But I love jousting with windmills…..
First off, I agree with your suggestion of a shared decision on many patient care decisions, end of life being one of the most important. But the time that it would take to go over the huge amount of data, which is fequently conflicting, with patients and their ability to understand and digest it is unrealistic for most medications regimens. Certainly for procedures more time does need to be spent. There is a reason patients go to a physician who has schooling, training and experience and it is for their expertise and recommendations. I think it is unrealistic, regardless of the amount of time someone spends on the internet, for most people to make an informed, intelligent decision about what medications they should take. I also strongly believe that as long as our Fee for Service system is in place there is an inherent conflict of interest in our physician’s recommendation. No where is this more of a conflict than in the area of chemotherapy treatment of cancer. The bias of informed consent really just turns it into consent–sans informed. I would like us to work on a system that physician compensation was tied just to good patient care and we are more likely to get informed consent from our physicians without a bias. See more about this on my blog. http://www.lakesidemedicalmusings.com
Vik,
I think I am optimistic that providers will move towards Dr. Hadler’s shared decision making model.
The essential element for this to happen is for patients to have “skin in the game” financially…..rather than consuming medical services with other peoples money. When patients share the cost they will shop for value….and the profession will respond.
While I strongly disagree with the ACA because it is based on a premise that improvement to the system will happen because wise bureaucrats will force doctors and patients to do the right thing……the fact that the ACA relies on high deductibles will lead to patients asking their docs the right questions and demanding collaboration from their doc and their medical system.
This is an excellent post. I hope it is true that people will soon find themselves dealing physicians who are not imperious and who are partners in care decisions, not dictators or directors.
Unfortunately, I am pessimistic about this. It will not be so easy to undo an era, spanning nearly 50 years, in which we have promoted and rewarded the physician as entrepreneur and not the physician as excellent clinician.
“Define a “preventable readmission” for me. And when you start talking about “risk adjustment” I’m sure you know how difficult that is.”
oops, didn’t mean to include this in your quoted section. Sorry about that Barry.
“How do we define quality anyway? My perception is that quality encompasses four areas – process (following evidence based guidelines and protocols), outcomes, preferably risk adjusted, patient safety (minimizing infections, preventable readmissions and the like) and patient satisfaction all of which would have to be appropriately weighted and somehow measured.” Define a “preventable readmission” for me. And when you start talking about “risk adjustment” I’m sure you know how difficult that is.
Barry,
Evidence based guidelines and protocols have their issues. If I tell you about the evidence underlying the guidelines and protocols, and mention how they might not apply to your unique situation, you might choose to not follow the protocols. So should I tell you simply to follow the guidelines and protocols or should I engage you in the decision?
What outcomes matter? Death? (what about choosing appropriate palliative and hospice care for a patient that may allow them to die on your shift?) Surrogate markers for diabetes or HTN? BMI? tobacco use? What about those physicians that practice in the inner city? Or rural areas? As I’m sure you know, payors have resisted adjusting for those things.
http://www.nytimes.com/2014/04/28/us/politics/health-laws-pay-policy-is-skewed-panel-finds.html?_r=0
(from the anti-Obamacare NY Times.)
I would love if I cold exclude patients for various reasons from my quality scores, but I can’t, and I don’t think most of us can.
I would love if I weren’t penalized for practicing medicine like I’d want for my family, but I often am.
I don’t know what the answers are, (thats for another post) but I know what the answers aren’t . And they aren’t “quality” as currently defined.
Al,
thanks for the reply.
Maybe I will do a blog post, but Dr. Centor over at medrants is already doing most of the heavy lifting on the topic of medical “quality.”
http://www.medrants.com/archives/7659
pcb,
You should do a blog pot on this. It boggles the mind how the powers that be have decided to measure what they can measure and more-is-better on these quality scales.
The one that really steams me is asthma. My son has very mild asthma. His pediatrician — supplemented by many mailings from the health plan — kept trying to convince us that my son needed a prophylactic inhaler to go with his rescue inhaler. That was 5-8 years ago (they did this for 3 years) and we never went along with the program…and my son never had a significant asthma attack regardless. But his doctor may not have hit the asthma metrics as a result of this.
Barry,
You often post here about the need to move to “quality” metrics when paying physicians. And here you are supporting the importance of shared decision making.
Shared decision making will often undermine quality metrics. If I engage a patient in the data surraound the decision to screen for colon cancer, or the rationale for meeting some arbitrary A1C goal for diabetes, or the lack of evidence that adding a 4th hypertension medicine in someone close to goal will provide them any benefit (and outweigh the likely increased side effect risk), I risk them choosing a treatment path that will harm my “quality numbers” as a physician.
So should I tell them that they “really need to screen”, or “it’s important that you take another drug,” to meet some external definition of quality care? Or should I engage them as fully as I can in the decision and risk that they choose otherwise once they know the potential harms/benefits?
Which kind of doctor do you want?
pcb,
When I talked about shared decision making, I was mainly thinking about surgical procedures, cancer treatment options and end of life care. I know that the issue around quality in healthcare is highly controversial. How do we define quality anyway? My perception is that quality encompasses four areas – process (following evidence based guidelines and protocols), outcomes, preferably risk adjusted, patient safety (minimizing infections, preventable readmissions and the like) and patient satisfaction all of which would have to be appropriately weighted and somehow measured.
I know that in the UK, primary care doctors can exclude a certain number of patients from their panel for metric measurement purposes based on issues like non-compliance and other factors. Maybe shared decision making could also be coupled with shared savings so the doctors get a bonus if they keep spending down without skimping on necessary care. I define necessary care as care the doctor would want for himself or a family member if faced with the same issues as the patient in front of him.
What’s your prescription for striking a reasonable balance between providing appropriate and beneficial care while controlling costs? Since the docs are the experts here, they’re the ones who should be providing the leadership. No?
This is an excellent argument for (among other things) requiring more math proficiency in high school. There are no easy answers in many of these cases (though I would never take statins absent a clear diagnosis, having been convinced by you, Alan Cassels, Gilbert Welch, and Marcia Angell) and patients have a lot of factors and probabilities to weigh. The difference between today and years ago is that patients are allowed or even encouraged to weigh factors, but that only increases the probability of making the right decision incrementally. You still have innumerate patients, innumerate doctors, perverse incentives, and questionable research to contend with.
Given the latter two, I would say that one excellent piece of advice based on this posting would be, when in doubt, don’t. I’ve never regretted not having something done or passing on a diagnostic or taking a drug, but I’ve several times regretted entering the “treatment trap.”
Thanks again for an excellent posting.
Yes, spectrum of disease is important. Error is not asymmetric. Benefits are not linear. Neither are harms.
One of the most seductive type of thinking in medicine is linear thinking.
X produces Y benefit so 10X produces 10Y benefits.
Or
X produces Y benefit so X/10 produces Y/ 10 benefits.
Linear thinking, and its correlates extrapolation & projection , is simply wrong. Shallow & wrong.
Bottom line: every patient is individual
There is so much co-morbidity and so much genomic variation in people that it seems nearly impossible to select a suitable control group for any randomized clinical trial. I.e. you can’t match the control to the experimental arm in everything except the dependent variable because you simply don’t have all the information about the genome and the patient’s other illnesses. Hence the scientific basis of our trade is going to be flimsy for quite some time. It may be difficult to ever get the Ns we need. This implies that there is still much art in our craft. And if what we are doing and creating with patients has a lot of artistry, the patient has as much “art” to contribute. We can help patients by listening and trying to define and somewhat quantitate their subjective feelings: How much pain, 1-10? Describe your muscle aches? Did your sense of well being stop after taking that drug?
Despite Dartmouth, we may need lots of practice variation to arrive at the best interventions. We still have to document what we find (that works) so that we move the anecdotes ahead toward solid investigations.
Nicely written article, Norton. I substantiallly agree.
It sounds like a good argument for shared decision making not to mention price and quality transparency. What I wonder about is how physicians’ fear of litigation factors into how they present treatment options to patients and even whether to present them at all as opposed to just definitively recommend a course of treatment based both on his perception of the science and the litigation environment in his city or region. If he happens to be the doctor who will actually perform the surgical procedure or intervention, revenue generation potential is likely to factor into the equation as well.