“CE” in medical parlance stands for either Comparative Effectiveness or Cost Effectiveness analysis. Both types of analyses are intended to allow health care providers (physicians, health systems, physician extenders) and patients reach rational decisions about choices in the treatment of specific clinical conditions. Most of the discussion today is on Comparative Effectiveness analysis, without a specific cost component. Comparative effectiveness analysis is critically important in today’s environment when there are often multiple therapeutic options. In the current practice, a practitioner who is familiar with a single modality is more likely to believe in that technique and will frequently, in good faith, recommend his/her most familiar therapy to most of her/his patients. Two examples come immediately to mind:
- Treatment of a patient with coronary disease and mild stable angina: -
a. An interventional cardiologist is likely to believe that angioplasty, usually with a stent, is optimal
b. A cardiac surgeon is likely to recommend CABG, especially if the patient has multi vessel disease
c. Finally, the medical cardiologist will recommend pharmacotherapy in conjunction with exercise training in cardiac rehabilitation.
The results of the COURAGE trial, released in 2007, suggest that from the vantage of trying to prevent myocardial infarction or death, the three therapies are equivalent.
- Treatment of a middle aged man with localized low risk prostate cancer:
a. A urologic surgeon has two potential procedures to offer: open or robotically assisted prostatectomy
b. A radiation oncologist has external beam radiation (IMRT, proton beam, or others) or implantation of “seeds” in the gland.
c. A third option would be watchful waiting.
Again from the vantage point of survival it appears that none of these have a clear advantage (see Watchful Waiting Wins Better Life).
One study identified five hundred potential screening measures for a general population of patients. A busy generalist cannot/should not be expected to be familiar with all potential therapies and come up with his/her reasoned preferences to all clinical conditions that might be seen in any month of practice.
Recognizing that there were few data sets to help with decision-making when competing therapies are present, a Democratic Congress authorized, and Ronald Regan signed, legislation creating and funding the Agency for Health Care Policy & Research. The charge to this agency was to encourage “Health Services Research”. In 1999, a Republican Congress and President Bill Clinton, extended the charge to the agency, but changed the name to the Agency for Healthcare Research and Quality (AHRQ). In 2003, the 108th Congress, with Republican majorities in both houses, and President G. W. Bush, passed the “Medicare Prescription Drug Improvement and Modernization Act” (MMA 11/25/03 PL 108-73), which authorized $50MM on outcomes and clinical effectiveness research (sect 1013). At this time there was a shift in terminology, calling CE research “Patient Centred Outcomes Research” (PCOR). In 2009, as part of the American Recovery & Reinvestment Act, Congress authorized $1.1BB for PCOR, $600MM of which went to directly fund or be administered by AHRQ. Finally, in 2010, as part of the Patient Protection and Affordable Care Act (PPACA, PL 111-148), further funding to PCOR was made into law. Clearly, the history of Comparative Effectiveness research has been non-partisan and supported by Congresses and presidents of both parties.
As currently configured, the goal of PCOR is to encourage thoughtful analysis of various ways of delivering care so that a patient in one part of the country should receive care that is not dissimilar to that from another. It is also intended to help with informed discussion between providers and patients. The PPACA has clear language that the results of CE analysis are not to be used to make guidelines or to limit payment. PCOR results are to be “descriptive not proscriptive”. The results of these analyses are to identify what is known and what is not (shades of Rumsfeld’s – known unknowns and unknown unknowns). There are gaps in what we know about effectiveness as opposed to efficacy, and we need to look at therapies in the light of potential benefits and harms. Many of these data will come from registries (a few of which are already active). While registries don’t have the scientific rigor of a randomized clinical trial, if done well, they can give useful information. Some data should always be better than none. As we get more registry data, our ability to do a better job of data collection will improve.
Why then, is there so much anxiety over government sponsored health systems research while there is almost none over NIH funding for “basic biomedical research”? Many stakeholders may find the reasons in a feeling, that further research is simply not needed. Some people, especially physicians, are afraid that the data will be imperfect and may be misleading. Others are anxious that once the data are available, they will be used to limit payment for some forms of therapies that have been found to be less effective. They believe that even though the current legislation prohibits this, future congresses or executive departments (CMS) may “misuse” the results of the analyses. There is certainly precedent for this feeling. On the other hand, in the absence of CE analysis, some therapies with a large potential for real harm and little return may be used with the best of intentions. Finally, individual providers (health systems and physicians at least) may feel anxiety that something they have invested dollars or training in may be found to be no more effective than another treatment. The more effective treatment may be either less costly or provided by another specialist or institution. Thus, there may be a need for an institution to reconsider use of a facility that has already been built. A physician may need to re-train in another medical field.
It would appear that with both the knowns and the unknowns that exist in medicine in the early 21st century, we should do as much as possible to understand the relative merits of differing therapies. CE research/analysis is clearly one technique to help gain such insight. If the private sector (industry, professional organizations) won’t do it in an unbiased manner, then perhaps an agency such as AHRQ should have our support, counsel and thought.