Introduction

In 1996, the American Society of Anesthesiologists published “Guidelines for Sedation and Analgesia by Non-Anesthesiologists” (₁) (the Guidelines). These were designed to help both anesthesiologists and non-anesthesiologists to develop uniform institutional policies for the practice of sedation and analgesia (“conscious sedation”) as recommended by the Joint Commission for Accreditation of Healthcare Organizations. Like the other practice parameters adopted by the A.S.A., these guidelines were developed using evidence-based methodology as recommended by the Agency for Healthcare Policy and Research (₂). However, in contrast to all of the other ASA practice parameters which relate to the practice of anesthesiology, these guidelines specifically apply to the practice of non-anesthesiologists. While non-anesthesiologists were involved in their development, both as members of the Task Force and as Expert Consultants, there was no way to predict how the guidelines would be received by those larger community of practitioners whose clinical practices would be directly affected. The goal of the present survey was to determine whether the Guidelines have influenced the practice of sedation and analgesia by non-anesthesiologists.

Materials and Methods

With IRB approval. surveys were sent to 280 non-anesthesiologist practitioners of specialties where sedation and analgesia (conscious sedation) are commonly administered for diagnostic or therapeutic procedures (Table 1). Practitioners were selected at random (by a dice-roll procedure) from the ABMS directory of board certified medical specialists (₃) and the American Dental Directory. (₄) For each specialty we selected 20 practitioners, half from private practice and the remainder from academic practice. In addition to the survey, each mailing contained a copy of the A.S.A. Guidelines for Sedation and Analgesia by Non-Anesthesiologists (the Guidelines) and a stamped, pre-addressed return envelope.

Figure 1

Table 1: Demographics of surveyed non-anesthesiologist practitioners of sedation and analgesia (“conscious sedation”). Individuals who indicated that they do not administer sedation to patients are excluded

To establish the demographic characteristics of our respondents, we asked each individual to identify his specialty, practice type (private vs. academic) and the number of patients to whom he/she administers sedation on an annual basis. To determine whether changes in practice were directly related to familiarity with the guidelines, we specifically asked whether respondents were aware of the guidelines before receiving the questionnaire. We then asked a series of yes / no questions relating to the individual recommendations made by the Guidelines. Respondents were asked whether they had followed each recommendation before the Guidelines were published in 1996, and whether they currently follow them (January, 2001). We also asked which medications they had used for sedation before the guidelines were published (1996), and which ones they use at the present time. Finally, we asked about the costs involved in implementing any changes in practice required by the Guidelines.

Statistical Analysis

For each respondent, we determined the total number of practice recommendations which were followed both before publication of the Guidelines and during current practice. Repeated-measures ANOVA determined whether there was a significant change in the number of recommendations followed, as well as whether this change was related to an individual's previous familiarity with the guidelines. If overall significance was present, post-hoc binomial tests of proportions determined the significance of changes in the frequency with which individual recommendations were followed. For contingency tables, we used chi-squared tests. P<0.05 indicated significance.

Results

A total of 280 surveys were mailed; 40 were returned by the practitioners and an additional 14 were returned as undeliverable by the post office. Six of the returned surveys were excluded because the practitioner does not administer sedation, leaving us evaluable data from 34 respondents. Twenty four of these were in academic practice, while the remainder were private practitioners. The median number of patients sedated by each respondent was 100 per year (interquartile range 50-950). Fifty six percent of the respondents were aware of the guidelines before they received the survey; this fraction did not differ between academic and private practitioners.

Prior to the publication of the A.S.A. Guidelines for Sedation and Analgesia by Non-Anesthesiologists (the Guidelines) in 1996, respondents followed 16.6±3.4 (±S.D.) of the 21 recommendations enumerated in Table 2. At the time of the survey in 2001, this had increased to 18.2±2.8 recommendations (P=0.003). Among individuals in private practice the number of recommendations followed increased by 2.8 (from 15.2±4.5 to 18.0±4.0) while among academic practitioners, the increase was only 1.0 (from 17.3±2.8 to 18.3±2.2, P=0.09 vs. private practice). The number of recommendations followed increased by 1.8±3.5 in individuals who were previously aware of the guidelines and 1.1±1.6 in individuals who were previously unaware of the guidelines (P=0.48). Post-hoc testing revealed that these changes were related to significant increases in the proportion of respondents who followed five of the Guideline recommendations (Table 2). For those individuals who monitored blood pressures during sedation, the median interval between blood pressure measurements was 5 min (interquartile range 3 - 5 min) which did not differ between 1996 and 2001.

Figure 2

Table 2: Percentage of respondents following practices recommended by the A.S.A. Guidelines for Sedation and Analgesia by Non-Anesthesiologists before their publication (1996) and at the time of the survey (2001). Questions are quoted directly from the survey.

Figure 3

The survey revealed that midazolam is the agent most commonly used for “conscious sedation.” The proportion of respondents who used each of the tabulated medications did not change significantly between 1996 and 2001 (Table 3). Of the individuals reporting the use of propofol in 2001, two were oral-maxillofacial surgeons, three were intensivists, and one was a general surgeon. Of the individuals who used ketamine three were emergency physicians, one was an oral-maxillofacial surgeon, one was a general surgeon, and one was a gastroenterologist. Chloral hydrate was primarily used by interventional and pediatric radiologists, pediatric dentists, and radiation oncologists.

Figure 4

Table 3: Percentage of respondents using the indicated medications during sedation and analgesia before publication of the guidelines (1996) and at time of survey (2001)

Figure 5

Of the 26 individuals who responded to the portion of the questionnaire related to the cost of guideline implementation, 15% purchased new equipment, supplies, or pharmaceuticals in order to implement the Guidelines. The median acquisition cost for capital equipment was estimated at $5000. The median estimated increase in costs attributed to continuing to follow the Guidelines was $3000 per year.

Discussion

Although the American Society of Anesthesiologists has adopted 11 practice parameters (guidelines and advisories) over the last decade, there are no previous data to indicate whether they have altered clinical practice or patient outcomes. The guideline development process is rather expensive: From inception through completion, a typical guideline costs ASA between $150,000 and $200,000. (₅) Although the practice parameters have been widely accepted both by ASA members and outside organizations (such as JCAHO), there has been no evidence that the guidelines and advisories have made a significant impact on patient outcomes.

Part of the problem lies in the relatively low incidence of sedation and anesthesia-related complications. A randomized study involving at least 37,000 subjects per group would be required to demonstrate a 50% reduction in the likelihood of a complication whose initial incidence was 1:1000 (?=0.05, 1-?=0.8). This estimate assumes, of course, that the practices recommended in the guidelines would not have been used if the guidelines had not been developed. On the other hand, if the guidelines merely restate common practice, it becomes even more difficult to demonstrate a guideline-related improvement in patient outcomes.

The ASA Guidelines for Sedation and Analgesia by Non-Anesthesiologists are unique in that they are the only ASA practice parameter which were developed to solely affect the practice of non-anesthesiologists. They were developed using a series of testable linkages, relating patient care interventions to patient outcomes. When there were sufficient data in the literature, linkages were tested by meta-analysis; in the absence of sufficient literature, they were tested based upon the opinion of a panel of expert consultants. Our questionnaire was based upon the linkages originally used for developing the Guidelines. We were able to demonstrate an overall increase in the average number of guideline recommendations being followed between 1996 (before the guidelines were published) and 2001 (the time of the survey). However, we were unable to establish that this increase was greater for practitioners who were previously familiar with the guidelines than for those who had not seen them before receiving the survey. There are several possible explanations for this discrepancy:

• A disappointingly small fraction of our surveys were returned by individuals who administer sedation and analgesia (12%). Power analysis reveals that we would need a minimum of 254 surveys from each group (previously aware vs. previously unaware of guidelines) to demonstrate a significant difference in the increase in number of recommendations being followed (?=0.05, 1-?=0.8). Assuming a comparable rate of return, we would have to mail at least 4,200 surveys!

• Practitioners who were personally unaware of the guidelines may have been affected by them indirectly. Hospital and departmental policies may have been changed based on the Guidelines (6) and subsequent mandates by accrediting organizations such as the JCAHO.

• Monitoring equipment may have become less expensive and more universally available between 1996 and 2001.

• Practices may have been changed based on guidelines or standards promulgated by other professional organizations.

Conclusion

In summary, the number of Guideline recommendations being followed by non-anesthesiologist practitioners of moderate (“conscious”) sedation increased significantly between 1996 and 2001. It is particularly encouraging to note that the use of pulse oximetry, immediate availability of a defibrillator, and contemporaneous recording of monitored parameters (vital signs, SaO2, level of consciousness). While these changes may not have been a direct result of the Guidelines, the fact that over half of our respondents were previously aware of the Guidelines suggests that they have had a significant impact among non-anesthesiologists.

Correspondence to

Jeffrey B. Gross, M.D. Department of Anesthesiology (M/C 2015) University of Connecticut School of Medicine Farmington, Connecticut 06030-2015 Phone: 860-679-4261 Fax: 860-679-1259 E-Mail: GROSS@NEURON.UCHC.EDU