To those who know me, it will come as no surprise to know that when it came time to choosing up sides for basketball during P.E. in grade school, junior high school and high school, I was always one of the last to be chosen. Tall, skinny, I stood there with the overweight guys withstanding the humiliation of being picked last. However, when we were outside and picking the track teams, I was always the first or second chosen. I wasn't coordinated but I was fast. (This is actually, still the case.)
The economic stimulus bill recently passed by Congress and to be signed into law by President Obama perhaps tomorrow, contains a provision for $1.1 billion for research to be conduct. s. The goal of the provision is to allow scientists to compare treatments for the same condition and determine which is better.
An article in today's New York Times by Robert Pear explains it this way – what therapy would treat neck pain better – surgery or physical therapy? What would the best combination treatment for mild depression of drugs and talk therapy?
There is little question that comparative effectiveness is an important field of study. If a new drug is only just equal to or not better than an existing older and less expensive treatment, prescribers and patients should know that. But to the degree that comparative effectiveness outcomes begin to drive reimbursement decisions or even treatment decisions along the lines described in the New York Times article, then it ceases to be a useful tool. Getting blood pressure under control sometimes requires a hit and miss and combination of several drugs to get results. And the idea that there is a "best" combination therapy for mild depression implies that all people are alike – that all people would make good basketball players. Well, they aren't and they don't.
There are people called "outliers" who, for whatever reason, are going to be different and they will respond to therapies that may not do any good for most patients, but which work for them. Comparative effectiveness may be a great tool to help define parameters. If however, treatment and reimbursement decisions are guided by comparative effectiveness research, then the outliers may be left standing on the sidelines, which is ok in basketball, but not in medicine.
One of the myths surrounding comparative effectiveness is that it conforms to a one-size-fits-all way of thinking about drugs, devices, and services. It does not. If conducted properly with appropriate sensitivity analyses, comparative effectiveness assessments stratify patients into sub-groups, assessing the comparative (cost) effectiveness per group. Statins, for example, have been shown to be highly effective for high-risk groups and less effective for low-risk groups. As with all scientific evidence determined on the basis of averages, however stratified, this evidence should be used as a guide and not a mandate in the reimbursement and prescribing decision making process.
If, as you say, comparative effectiveness is a useful tool, then where else would it demonstrate its use value other than as a guide to inform the reimbursement and prescribing decision making processes?
Currently, in the United States, the reimbursement decision-making process is as opaque as can be. Particularly among branded products, we really have no idea why products are on one formulary, off another, and why cost sharing is high for one drug and low for another. Besides an alarming lack of transparency on the part of third party payers, this can also be attributed to the evidence gap that exists on the comparative effectivess, safety, and cost of many drugs, across many therapeutic classes. As a nation we must do something about this. And, a federal comparative effectiveness institute may bring about the kind of evidence and transparency that is lacking in the current (non) system of decision making regarding reimbursement.
Joshua Cohen
Joshua you make a good point, but you apparently have more faith than I do that the analyses will be done in a way that they are useful for reimbursement while at the same time take into account the fact that not all patients are made alike. It may be a myth that comparative effectiveness studies are not a blunt instrument, but in the amount of money set aside for these studies is not huge and the variable factors that go into the treatment of some conditions, such as depression or high blood pressure, are very great in number. I’m therefore not as confident as you may be that this instrument may be too blunt for reimbursement decisions, leaving some outliers without effective treatment.
Mark, you’re correct that if wielded bluntly comparative effectiveness analyses could be detrimental to health outcomes. For example, there is substantial inter-patient variability among SSRIs. Here, coming up with recommendations based on averages of patients is indeed unhelpful. But, one must remember that it’s comparative effectiveness assessments that pointed to this patient inter-variability. The problem is that in many other therapeutic classes there’s a dearth of knowledge. And so perhaps having a national clearinghouse of vetted information which all stakeholders could use to their advantage would move us in the right direction.
Mark, the last half of this post has a different tone from your usual posts when I read it. Can you confirm that this extract from your blog applies to this post, or (if not) acknowledge the source. “What you will NOT find here. While I work for an agency, you will not find their thoughts here. There is a link here to the agency’s Web site, in case you are interested. But the thoughts and analyses expressed here are mine, which the agency is kindly letting me to post in this venue. But the point of view expressed here is my own and not those of my firm. My postings are not screened or approved.”
Australian perspective: we have been applying comparative effectiveness research to reimbursement for 15 years, and the world hasn’t ended yet. Comparative research is currently being done by pharmaceutical companies all the time, it’s all in the picking of the comparators.
Mark, I share your fears. Thyroid patients in the UK have in fact recently been impacted by just this sort of blunt instrument. The Royal College of Physicians (RCP) earlier this month issued extremely rigid diagnostic and treatment guidelines for hypothyroidism stating that “thyroxine is the only treatment that should be given for this condition”. Patients have already been forced to switch, even though many derive greater benefit from other medications.
Additionly, the RCP demotes doctors to lab technicians, stating that “patients with normal T4 and TSH do not have primary hypothyroidism and even if they have symptoms which might suggest this should not be given thyroid hormone replacement therapy”. In other words, lab tests trump clinical evidence.
Thyroid patient advocate Mary Shomon writes “[t]hese guidelines make diagnosing and treating thyroid disease a cookie cutter process, that leaves little decision-making to the practitioners.” Of course, it is the patients who will suffer most from this process, especially the “outliers”.
Please read Mary’s article “Thyroid Shakeup in the United Kingdom: British Guidelines Severely Limit Hypothyroidism Diagnosis and Treatment Options”.
http://thyroid.about.com/b/2009/02/18/united_kingdom_thyroid_guidelines.htm#gB3