The USPSTF, having been in existence for over 20 years, serves as a model of recommendation development in the United States, especially because of its adherence to detailed methodologies and the restrictions it places on conflicts of interest. Clinicians, health plans, and payers have come to rely on the regular reports from the task force to update their practice, payment, or coverage policies regarding clinical preventive services.
The IOM Committee on Clinical Practice Guidelines defined high-quality guidelines as having a number of attributes, including validity, reliability, reproducibility, clinical applicability and flexibility, clarity, development through a multidisciplinary process, scheduled reviews, and documentation (IOM, 1992). Over time there have been noted improvements in the capacities of some clinical and professional organizations to develop robust, evidence-based guidelines (Jackson and Feder, 1998). Nevertheless, the overall quality of clinical practice guidelines is highly variable, and in fact, the quality is often very poor (Shaneyfelt et al., 1999). Shaneyfelt and colleagues (1999) assessed the quality of 279 guidelines produced over the period of 1985 to 1997 and assessed their quality against a set of 25 standards. The investigators found that the mean number of quality standards satisfied over that period was 11 (43 percent). For example, less than 10 percent of the guidelines described formal methods of combining scientific evidence and expert opinion. The investigators also evaluated the guidelines in accordance to their specification of purpose (75 percent compliance), definition of the patient population involved (46 percent), pertinent health outcomes (40 percent), method of external review (32 percent), and whether an expiration date or scheduled update was included (11 percent). Overall, the investigation found significant improvement over time, but each guideline still only met 50 percent of the standards, on average, in 1997 (Shaneyfelt et al., 1999). For some, this variability in guideline quality called for greater transparency in guideline reporting and more rigorous peer review (Cook and Giacomini, 1999).
An evaluation of 86 guidelines developed in 11 countries (which did not include the United States) concluded that the guidelines produced by government-funded agencies and established guideline development programs were of higher quality than guidelines produced by specialty societies (Burgers, 2003). This finding was consistent with the conclusions of Grilli and colleagues (2000), and also with Hasenfeld and Shekelle (2003), who found that the 17 guidelines produced by the AHCPR from 1990 to 1996 were of a substantially higher quality than those subsequently produced