PEDIATRICS Vol. 105 No. 3 March 2000, pp. 496-501

Pediatricians' Attitudes, Beliefs, and Practices Regarding Clinical Practice Guidelines: A National Survey

Glenn Flores, MD^{*, §}, Mina Lee, MD, Howard Bauchner, MD^{*, §}, and Beth Kastner, MPH

From the Divisions of ^* General Pediatrics and  Pediatric Emergency Medicine, Boston Medical Center; and ^§ Boston University Schools of Medicine and Public Health, Boston, Massachusetts; and  Department of Pediatrics, Yale University School of Medicine, New Haven, Connecticut.

	ABSTRACT

Top Abstract Methods Results References

Background.  Clinical practice guidelines are increasingly being used for a wide variety of medical conditions, but not enough is known about physicians' attitudes and beliefs about guidelines, how often and under what circumstances they are used, and factors associated with their acceptance.

Objective.  To determine practice guideline attitudes, beliefs, practices, and factors associated with use among a representative national sample of general pediatricians.

Study Design.  Cross-sectional mail survey.

Subjects.  Random sample of general pediatrician members of the American Academy of Pediatrics residing in all 50 states and Puerto Rico.

Survey Instrument.  Twenty-four multiple-choice, Likert scale, yes-no, and open-ended questions about pediatric clinical practice guidelines.

Results.  From 1088 respondents, 461 specialists were excluded; the remaining 627 general pediatricians were mostly male (61%), white (81%), and in group practice (62%) in a suburban location (48%). Practice guidelines are used by 35% of pediatricians, in part by 44%, and not at all by 21%. Over 100 different practice guidelines are used, most commonly for asthma (77%), hyperbilirubinemia (27%), and otitis media (19%). Common reasons for use of practice guidelines include standardization of care (17%) and helpfulness (10%). Commonly cited problems with practice guidelines include failure to allow for clinical judgment (54%), use in litigation (16%), and limitation of autonomy (5%). In multivariate analysis, the odds of practice guideline use were greater among pediatricians in health maintenance organization practices (odds ratio [OR]: 9.1; 95% confidence interval [CI]: 1.2-68.0) and those who were nonwhite (OR: 2.3; 95% CI: 1.1-4.8), but lower in those with more weekly patient visits (OR: .7; 95% CI: .5-.9). Features most likely to lead to practice guideline use include simplicity (16%), feasibility (12%), and evidence of improved outcomes (10%). Most pediatricians agree that practice guidelines improve outcomes (89%), are motivated by a desire to improve quality (94%), and should not be used in litigation (82%) or disciplinary actions (77%), nor be motivated by a desire to reduce costs (73%).

Conclusions.  Most general pediatricians use practice guidelines, but no specific guidelines, except those for asthma, are used by >27% of pediatricians. The results of this study suggest that practice guidelines are most likely to be followed if they are simple, flexible, rigorously tested, not used punitively, and are motivated by desires to improve quality, not reduce costs.  Key words:  practice guidelines, pediatrics, attitude of health personnel, quality assurance, health services research.

Clinical practice guidelines (CPGs) increasingly are being used to manage a wide variety of medical conditions. CPGs are being used as educational tools, quality assurance guides, standards of care, and in medical liability cases. Limited data suggest that adoption of certain rigorously tested, explicit CPGs can improve health processes and outcomes,¹ including shorter hospital length of stays² and reduced costs and utilization of resources.³ Unfortunately, the clinical utility of most CPGs has not been convincingly demonstrated, putting health care providers in difficult positions each time a new CPG is released and advocated.

Not enough is known about physicians' attitudes and beliefs about CPGs, how often and under what circumstances CPGs are used, and factors associated with their acceptance. The 1 published study that investigated CPG attitudes and use among pediatricians⁴ examined only 4 CPGs, was limited in sample size, and did not extensively investigate factors associated with CPG use. The goals of this study were: 1) to determine CPG attitudes, beliefs, practices, and factors associated with use among a representative national sample of general pediatricians; and 2) to describe pediatricians' perspectives on the role of CPGs and how to promote their use.

	METHODS

Top Abstract Methods Results References

Subjects

We obtained a list of 2000 randomly-selected members of the American Academy of Pediatrics (AAP) who reside in the United States (including Puerto Rico). A survey and self-addressed stamped envelope were mailed to each subject. A second mailing was sent to those subjects who had not mailed back a response within 6 months. Surveys returned by pediatric self-identified subspecialists were excluded from additional analysis. We chose to exclude subpecialists because we were interested in focusing on the attitudes and practices of general pediatricians.

Survey Instrument

The survey consisted of 24 questions about the characteristics of the pediatrician and his or her clinical practice, familiarity with and use of specific CPGs, information sources for learning about CPGs, attitudes and beliefs about CPGs, reasons for CPG use/nonuse, and perceptions about the benefits and disadvantages of CPGs. Respondents were asked about their familiarity with 9 major pediatric CPGs. These 9 CPGs were selected because they fulfilled the following criteria: 1) they had been released at least 3 years before the survey; 2) they had received widespread attention; and 3) they were judged as important during a recent conference on pediatric CPGs.⁵ To identify the potential spectrum of guidelines used for a given condition, we asked subjects to name any CPG they were familiar with and/or used. The format of question responses included multiple choice, Likert scale, yes-no, and open-ended.

Data Entry and Analysis

Data were entered and initially analyzed using Epi Info 6.0.⁶ Multiple logistic regression was performed in a stepwise fashion using SPSS, ⁷ with -to-enter set at .15. The outcome variable in logistic regression was use/nonuse of CPGs, and was dichotomized as used or partially used versus not used. Independent variables examined in the multivariate analysis included pediatrician gender and ethnicity; years in practice, the number of patient visits per week (dichotomized at the median as 114 vs <114 visits/week); and the proportion of patients insured by managed care or health maintenance organizations (HMOs; dichotomized at the median as 45% vs <45%), by private insurance (dichotomized at the median as 25% vs <25%), and by public insurance (dichotomized at the median as 20% vs <20%). Separate multivariate analyses were performed for each of the 3 patient insurance categories.

	RESULTS

Top Abstract Methods Results References

Of the 2000 subjects contacted, 1088 returned surveys, for a response rate of 54%. Surveys from the 461 respondents who were subspecialists were excluded from additional analysis.

Characteristics of Pediatricians and Their Practices

Table 1 summarizes selected characteristics of the 623 general pediatrician respondents and their practices. Most were male (60%), white (81%), and in group practice (62%) in suburban locations (48%). Subjects had practiced for a median of 14 years, and saw a mean of 114 patients per week. The insurance coverage of patients seen by study pediatricians was most often managed care (median = 45%), followed by fee-for-service (25%) and Medicaid (20%).

Sources of Guidelines and Beliefs About Their Origin

Pediatricians identified 30 sources for obtaining CPGs, including national organizations (43%), colleagues (25%), journals (23%), mailings (14%), managed care organizations (10%), publications other than journals (9%), their own practice (5%), lectures and conferences (4%), drug companies (3%), insurance companies (3%), and making up one's own CPGs (2%).

When asked who creates guidelines, pediatricians most often said national organizations (32%), don't know (12%), physicians (10%), local hospitals (10%), specialists (9%), experts (8%), and committees/task forces (7%).

Clinical Practice Guideline Familiarity and Use

Most pediatricians (82%) said that they were familiar with CPGs (Table 2). When asked to name a CPG, subjects most often mentioned asthma, hyperbilirubinemia, and otitis media with effusion. Of 9 major CPGs, participants were most familiar with those for asthma, health supervision, hyperbilirubinemia, and Early and Periodic Screening: Diagnosis and Treatment.

CPGs are used by 35% of pediatricians, in part by 44%, and not at all by 21% (Table 2). Participants reported using 103 different CPGs, most commonly for asthma, hyperbilirubinemia, and otitis media. Among those using CPGs at least in part, most stated that they were extremely helpful (25%) or helpful (57%), but few (1%) said that they were not at all helpful.

A surprising diversity of CPGs may be used for a given condition. For example, participants reported using 10 different CPGs for asthma, including those from national organizations (the National Heart, Lung, and Blood Institute), regional quality improvement organizations (the Institute for Clinical Systems Integration), local institutions (hospitals and clinics), and publications (Pediatric Review and Education Program). Pediatricians also reported using their own personal asthma CPGs, those for specific aspects of asthma management (oral steroids, inhaled steroids, peak flow meters, and nebulizations), and unspecified general asthma CPGs.

Reasons for Use/Nonuse and Reported Problems Regarding Clinical Practice Guidelines

Pediatricians who use CPGs (at least in part) cited 35 reasons for their use (Table 3), most commonly standardized care/uniform management, helpfulness, the condition is commonly encountered, quality assurance, common sense/logic/practicality, and utility as reference guides/reminders. Pediatricians who do not use CPGs cited 36 reasons for nonuse (Table 3); those most often mentioned were cookbook medicine/do not allow for clinical judgment (41%), they are followed only loosely/in part (20%), and my practice is the same as the guidelines (18%).

In multiple logistic regression (Table 4), CPG use was significantly more likely among pediatricians practicing in HMOs (odds ratio [OR]: 9.1; 95% confidence interval [CI]: 1.2-68.0) and those from non-white racial/ethnic backgrounds (OR: 2.3; 95% CI: 1.1-4.8), but significantly less likely (OR: .7; 95% CI: .5-.9) among those seeing greater numbers of patients per week. No significant associations were found for the number of years in practice, gender, or the regional location of the practice.

Respondents identified 84 problems with CPGs (Table 3); most frequently mentioned were no allowance for clinical judgment (54%), use in malpractice cases (16%), and limiting autonomy (5%).

Recommendations on the Role of Clinical Practice Guidelines and How to Promote Their Use

Subjects offered 77 recommendations on how CPGs should be used. Pediatricians (38%) most frequently recommended that CPGs should be used only as guidelines or general approaches. Other recommended uses included education of both providers and parents (10%), as a reference/review (6%), for adaptation to a provider's population (5%), quality assurance (4%), as an adjunct/aid (3%), as the standard of care (3%), as recommendations (3%), and as the basis for decision-making (3%).

CPGs were shared with parents by 41% of pediatricians. The most common of the 13 ways that participants shared CPGs with parents included general discussion (51%), written information (26%), educational contexts (7%), support of management decisions (6%), quotation of specific guidelines (3%), and informally (3%).

When asked what is the optimal way for the AAP to familiarize practitioners with CPGs, participants identified separate mailings to all members (56%), publish in the journal Pediatrics (20%), publish in a journal other than Pediatrics (20%), compile CPGs into a single book/reference (6%), conferences and meetings (3%), and the Internet (3%).

Pediatricians said that they were most likely to follow CPGs if they were simple to follow, feasible/practical for their practice, demonstrated to improve outcomes, evidence-based, or logical (Table 5). Most agree that CPGs are motivated by a desire to improve the quality of care and are likely to improve outcomes (Table 6). Substantial majorities also stated, however, that CPGs should not be used in litigation or disciplinary actions nor be motivated by a desire to cut costs (Table 6).

	COMMENT

Most general pediatricians are familiar with CPGs and use them at least in part. We found, however, that no specific CPGs, with the exception of those for asthma, are used by >27% of general pediatricians. This discrepancy in part may be attributable to lack of familiarity with CPGs for certain commonly-encountered conditions. Fever and gastroenteritis, for example, are frequent diagnoses in pediatric practices, but only 40% of pediatricians said they were familiar with fever CPGs, and 34% with gastroenteritis CPGs. Practical solutions for enhancing familiarity with CPGs were offered by study participants; most overwhelmingly cited mailings and publication in journals, and a few suggested newer alternatives such as the Internet.

Pediatricians may avoid using most pediatric CPGs because few have been subjected to the rigorous clinical evaluation necessary to demonstrate improved quality and reduced costs. In contrast, recent evaluations of adult CPGs have shown that they can reduce the length of hospital stays for gastrointestinal tract hemorrhage,² and decrease radiography, waiting times, and costs for acute ankle injuries.^3,8 Clinical testing of CPGs is important because, even for CPGs based on extensive scientific evidence, use in actual practice may not improve patient care. Problems with external validity, misapplication, co-morbidity, and lack of acceptance may thwart the clinical effectiveness of seemingly reasonable CPGs.⁹ It has been argued that CPGslike any new medical interventionshould be carefully evaluated for risks, benefits, feasibility, and cost.⁹ Several authors have cogently stated that CPGs should not only be outcome-based but outcome-justified.^5,9,10 Pediatricians in our study echoed this sentiment, reporting that they would be more likely to follow CPGs if they were evidence-based and had been shown to improve outcomes. Similarly, a study of Australian general practitioners revealed that the most important factor in deciding to follow a guideline was whether it was based on evidence.¹¹

The diversity of CPGs used for a given condition was surprising. In the case of asthma, for example, pediatricians reported using 10 CPGs, including those from national organizations, regional quality improvement organizations, local institutions, publications, and clinicians' own personal CPGs. This was unexpected, given the wide dissemination of both the original and revised versions of the National Heart, Lung, and Blood Institute's guidelines for the diagnosis and management of asthma.^12,13 It is also noteworthy that none of the respondents reported using the AAP's published practice parameter for the management of acute asthma exacerbations in children, even though all respondents are members of the AAP.¹⁴ The reasons for pediatricians' use of multiple alternatives to such widely-publicized CPGs are not clear. Consistent with what study participants identified as general CPG problems and reasons for nonuse, the aforementioned national asthma CPGs may be perceived by pediatricians as not adequately subjected to clinical testing, not sufficiently allowing for clinical judgment, or differing little from actual clinical practice. Additional research might elucidate why clinicians choose to use such a diversity of alternative CPGs instead of well-publicized national CPGs for a given condition.

Four published studies have examined clinician attitudes and beliefs regarding CPGS,^4,11,15,16 but none focused on the prevalence of CPG use, or practitioners' reasons for use/nonuse. Surveys of Australian general practitioners¹¹ and American internists¹⁶ confirm our finding that many clinicians view CPGs as good educational tools. Consistent with our data, objections to CPGs being overly rigid and "cookbook medicine" were frequent among American clinicians^4,16 but less common among Australian general practitioners.¹¹ Two studies^11,15 also supported our finding that clinicians are quite concerned about the use and abuse of CPGs in medicolegal contexts. As in our study, Australian general practitioners identified scientific evidence of efficacy as one of the most important factors in choosing to follow CPGs.¹¹

Several limitations of this study should be noted. Although the survey response rate was comparable to other national postal surveys, nonresponse bias may have influenced the results. A skewed sample consisting of respondents who view CPGs either more or less favorably than the national average may have resulted. To ensure confidentiality for participants, we did not mark surveys with identifiers. Although this may have yielded more frank responses, it precluded comparisons between respondents and nonrespondents regarding personal and practice characteristics. Participant comments, however, suggest that the extremes of favorable and unfavorable views toward CPGs were captured and that a reasonable national sample of general pediatricians was obtained. A second potential limitation is that self-reported use of CPGs may not accurately reflect actual use. For example, as many as 8% of internists said that they were familiar with a fictitious CPG for computed tomography of the head,¹⁶ and 49% of Australian general practitioners reported that a fictitious CPG for the diagnosis and management of pelvic inflammatory disease was minimally to extremely useful.¹¹ for self-reported use of CPGs by pediatricians in our study, therefore, might overestimate the proportion that actually use CPGs. Although this is the first study that we are aware of to report partial use of CPGs by clinicians, another limitation is that we were unable to explore what was meant by use of a CPG "in part." CPGs can contain information on diagnosis, therapy, follow-up, and prognosis, and it would be of interest to identify which component is most often selected by those who use CPGs only partially. A final potential limitation is the exclusion of subspecialist responses, leading to a sample that may not reflect the attitudes and practices of pediatricians certified by the American Board of Pediatrics. The study goal, however, was to examine the attitudes and practices of generalist pediatricians; future research might investigate whether there are important differences in the adoption of CPGs between generalists and subspecialists.

In this study, we did not define what constitutes a CPG. We surveyed pediatricians about well-known CPGs, as well as other statements sometimes referred to as guidelines, such as the health supervision guidelines of the AAP.¹⁷ The term guideline has been avoided by the AAP, which labels guides to practice as practice parameters. Several methodological problems in CPG development have been identified,⁵ including what are appropriate sources, whether consensus statements should be considered CPGs, and what represents adequate scientific evidence for inclusion in a CPG. Clinician confusion and lack of CPG acceptance may be an inevitable response to inconsistent terminology and the absence of an accepted definition for CPGs.

CPG use is significantly more likely among general pediatricians in HMOs and those who are non-white, but significantly less likely among those seeing greater numbers of patients. Because many HMOs routinely encourage and internally publicize the use of CPGs in daily practice, it is not surprising that HMO pediatricians are more likely to use CPGs. It is not clear why non-white ethnicity is independently associated with greater odds of CPG use. Pediatricians' responses to the question: "What would make you most likely to follow practice guidelines?" (Table 5) may provide insight on why those with the highest weekly patient visits are least likely to use CPGs. For the busy clinician, it would seem that CPG features of particular importance would include simplicity, feasibility, demonstrated efficacy, logic, appropriateness, and accessibility. Lower use among busier practitioners may, therefore, reflect the dearth of scientific evidence on the efficacy of most pediatric CPGs, and perceived complexity and difficulty in using certain more detailed CPGs.

Responses of this national sample of general pediatricians have implications for improving CPG quality and acceptance. Participants repeatedly emphasized that a CPG is most likely to be used when it is simple, feasible, allows for clinical judgment, does not impede autonomy, and has been shown to be effective. Most agreed that CPGs should not be used for medical liability or disciplinary actions, and should not be motivated by desires to control costs. The data suggest that American pediatricians believe that CPGs can improve patient care, but only when the CPGs are flexible, rigorously tested, not used punitively, and are motivated by desires to improve quality, not reduce costs.

	ACKNOWLEDGMENTS

This work was supported by Grant 030878 from the Robert Wood Johnson Minority Medical Faculty Development Program, an Institutional Research Training Grant of the Health Resources Services Administration, and the David and Lucile Packard Foundation.

	FOOTNOTES

Received for publication Jun 21, 1999; accepted Sep 3, 1999.

This work was presented in part at the Pediatric Academic Societies' Meetings; May 3, 1998; New Orleans, LA.

Reprint requests to (G.F.) Division of General Pediatrics, Boston Medical Center, 1 Boston Medical Center Place, Maternity 419, Boston, MA 02118. E-mail: glenn.flores{at}bmc.org

	ABBREVIATIONS

CPG, clinical practice guideline; AAP, American Academy of Pediatrics; HMO, health maintenance organization; OR, odds ratio; CI, confidence interval.

	REFERENCES

Top Abstract Methods Results References

1. Grimshaw JM, Russell IT Effect of clinical guidelines on medical practice: a systematic review of rigorous evalutations. Lancet 1993; 342:1317-1322 [CrossRef][Medline]
2. Hay JA, Maldonado L, Weingarten SR, Ellrodt AG Prospective evaluation of a clinical guideline recommending hospital length of stay in upper gastrointestinal tract hemorrhage. JAMA 1997; 278:2151-2156 [Abstract]
3. Stiell IG, McKnight RD, Greenberg GH, Implementation of the Ottawa ankle rules. JAMA 1994; 271:827-832 [Abstract]
4. Christakis DA, Rivara FP Pediatricians' awareness of and attitudes about four clinical practice guidelines. Pediatrics 1998; 101:825-830 [Abstract/Free Full Text]
5. Bauchner H, Homer C, Adams W The status of pediatric practice guidelines. Pediatrics 1997; 99:876-881 [Free Full Text]
6. Dean AG, Dean JA, Coulombier D, et al. Epi Info, Verson 6: A Word Processing, Database, and Statistics Program for Epidemiology and Microcomputers. Atlanta, GA: Centers for Disease Control and Prevention; 1994
7. Norusis MJ. SPSS for Windows, Release 7.5. Chicago, IL: SPSS Inc; 1996
8. Auleley G-R, Ravaud P, Giraudeau B, Implementation of the Ottawa ankle rules in France: a multicenter randomized controlled trial. JAMA 1997; 277:1935-1939 [Abstract]
9. Weingarten S Practice guidelines and prediction rules should be subject to careful clinical testing. JAMA 1997; 277:1977-1978 [CrossRef][Medline]
10. Brook RH Practice guidelines and practicing medicine. Are they compatible? JAMA 1989; 262:3027-3030 [Abstract]
11. Gupta L, Ward JE, Hayward RS Clinical practice guidelines in general practice: a national survey of recall, attitudes and impact. Med J Aust 1997; 166:69-72 [Medline]
12. Expert Panel, National Asthma Education Program. Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1991. US Department of Health and Human Services Publication No. (PHS) 91-3042
13. Expert Panel Report II. National Asthma Education Program. Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1997
14. American Academy of Pediatrics, Provisional Committee on Quality Improvement Practice parameter: the office management of acute exacerbations of asthma in children. Pediatrics 1994; 93:119-126 [Abstract/Free Full Text]
15. Kelly JF, Helfrick JF, Smith DW, Jones BL A survey of oral and maxillofacial surgeons concerning their knowledge, beliefs, attitudes, and behavior relative to parameters of care. J Oral Maxillofac Surg 1992; 50:50-58 [Medline]
16. Tunis SR, Hayward RSA, Wilson MC, Internists' attitudes about clinical practice guidelines. Ann Intern Med 1994; 120:956-963 [Abstract/Free Full Text]
17. American Academy of Pediatrics, Committee on Practice and Ambulatory Medicine Recommendations for preventive pediatric health care. Pediatrics 1995; 96:373-374 [Abstract/Free Full Text]