Problems with Guidelines

Methodological problems with guidelines abound. They have been enumerated elsewhere [2, 3], and we will briefly review them here. Guidelines are pieced together by committees, often referred to as working groups. A defined body of literature is identified. This is reviewed by the group and forms the scientific underpinnings for the guidelines. Time constraints compel establishment of a cutoff date after which no further publications are considered. This necessarily means that many guidelines will be obsolete by the time they are released. At the very least, it ensures that guidelines will be trailing edge. Why should a busy clinician, hoping to provide the benefits of the latest research to his patients, bother looking at guidelines based on research that is over 5 years old at best? It


© 2007 S. Karger AG, Basel 0253-5068/07/0251-0036$23.50/0

Richard Amerling, MD

Beth Israel Medical Center

350 East 17th Street, Baird Hall, 18th Floor

New York, NY 10003 (USA)

Tel. +1 212 420 4070, Fax +1 212 420 4117, E-Mail [email protected]

is much more time-effective to perform a quick literature search on PubMed, or read a current review on UpToDate.

The Heisenberg uncertainty principle states that we can never know anything with certainty because every act of measurement distorts that which we seek to measure. Nowhere is this more so than in medical experiments, where the placebo groups always seem to do much better than the general population. In many disciplines, particularly nephrology, the database upon which guidelines are based is shaky. There are few randomized, controlled, prospective trials in nephrology, and those that have been published are usually inconclusive. In nephrol-ogy, we would literally be unable to practice if we were limited to randomized, controlled trials for guidance. Nephrology guidelines are largely opinion-based. The makeup of the working groups then assumes major importance. We will discuss this more later.

Even randomized, controlled trials are subject to interpretation and opinion. They invariably look at large populations and results may not fit with the specific patient in the physician's practice.

Guidelines remain untested. Have guidelines influenced practice and raised the standard of care? Since 1989, the Agency for Health Care Policy and Research has spent hundreds of millions of dollars to measure 'what works' and to develop clinical guidelines. As of 1994, the agency could not point to a single example of its work affecting clinical practice. In 1996, the agency stopped working on practice guidelines and left that to professional organizations, which have developed more than 1,000. It now concentrates on 'how we can reduce inappropriate variation' [AM News February 24, 2003]. Guidelines will exert a major influence when they become actively implemented as clinical performance measures and when Pay for Performance (P4P) gets underway. This is one of the dangers to medical practice alluded to earlier. Outside parties can only dictate medical practice by adopting guidelines that we have created and turning them into mandates via P4P. We are witnessing the evolution from centralized payment for care to centralized prescription of that care! It is already well underway in the United Kingdom [4].

In response to claims for payments (which must be accompanied by diagnosis codes) submitted by us, we have received written questions from insurance companies if special tests recommended in KDOQI guidelines for specific stages of chronic kidney disease had been performed. It is of interest that there are no published studies of hard end-point outcomes of intervention based on the recom mended testing. The next stage, of course, is litigation should a practitioner fail to order theses tests. Another reason we need tort reform in America.

0 0

Post a comment