I’m delighted that discussion of the issue of avoiding avoidable care is beginning to happen. Its roots go back many years, and I would hope that no one is under the illusion that minimizing, let alone eliminating, avoidable care will be easy.
Beginning about 25 years ago, important work was done in this area by Bob Brook and his colleagues at RAND when they developed and applied the RAND Appropriateness Guidelines for various surgical procedures and invasive diagnostic procedures – e.g., coronary angiography, CABG, carotid endarterectomy, hysterectomy, and others. The RAND criteria were applied to a huge number of clinical scenarios that might lead to a particular procedure, and performing the procedure for patients with that scenario was then categorized as appropriate, uncertain, or inappropriate/unnecessary. A major problem was that very few instances were considered inappropriate/unnecessary – i.e., situations in which evidence or expert opinion indicated overwhelmingly that the risks outweighed the benefits. In the early 1990’s, reviews of patients who had coronary angiography in New York State and also at Harvard Community Health Plan, showed only about 6 percent of the procedures to have been “inappropriate” (Noonan SJ et al: Relationship of Anatomic Disease to Appropriateness Ratings of Coronary Angiography. Arch Internal Med 1995;155:1209-1213). Much larger percentages of persons undergoing such a procedure either were of “uncertain” appropriateness, 16 percent, or “appropriate,” 78 percent. A major question, however, is how many of the procedures are absolutely necessary – i.e., absolutely unavoidable. We do not have good evidence for knowing that, and I fear that we still have not put in place the necessary processes such as shared decision-making and large outcomes databases for gaining that evidence, which, by the way, for the most part will not come from randomized controlled clinical trials (Schoenbaum SC: Toward fewer procedures and better outcomes. JAMA 1993;269:794-796).
The point I want to emphasize is that if a discussion of avoiding avoidable care is limited to the relatively small number of things for which there is blanket evidence that they are always unnecessary, it will be a discussion of the top of the tip of the iceberg. I hope it will quickly become evident that a) we need to develop better evidence of what is necessary in what circumstances; b) we need to learn how to have difficult conversations with each other and with patients about what we do and do not know about benefits, risks, and costs; and c) we need to have difficult conversations among professionals about errors of omission, commission, and risks of malpractice in order to made significant inroads into avoiding avoidable care.