Identifying Guideline-Practice Gaps to Optimize Evaluation and Management for Patients With Syncope

Abstract

Syncope is a common and complex symptom that requires efficient evaluation to determine the cause. Recent guidelines focus on high-value testing, but a systematic evaluation of their implementation has not been performed. To this end, we used a mixed-methods approach of surveys, chart reviews, and focus groups to understand current practices relating to the diagnosis and management of patients with syncope and to identify barriers and facilitators to the implementation of guideline-supported recommendations. Surveys were distributed to 1500 providers in the specialties of hospital medicine, cardiology, emergency medicine, and family medicine, and 175 responses were received. Knowledge of class I and III guideline recommendations were assessed with the use of clinical vignettes, which were answered correctly 60%−80% of the time. Chart reviews focused on patient history and testing for syncope. Per the guidelines, < 50% of charts met criteria for bare minimum history and physical examination. Based on the documentation, 25% of echocardiograms and 90% of neurologic testing obtained would not have been appropriate per the guidelines. Self-reported and actual practice patterns were similar in rates of testing. Our results indicate that there remains a gap between guideline-directed management and actual practice for syncope. Focus groups revealed barriers across multiple levels of care that need to be addressed to improve care. Our findings emphasize the need for proactive strategies to improve syncope testing practices, potentially saving millions of dollars in the health care system.

Syncope is a common reason to seek medical attention. Nearly 1 in 2 individuals will have at least 1 syncopal event in their lifetime and up to 13.5% will experience recurrent episodes.¹ The underlying cause for syncope can be challenging to establish in an efficient manner. In an effort to provide guidance on optimizing the evaluation and management of syncope, several groups have published guidelines to management of syncope, including the Canadian Cardiovascular Society,² the European Society of Cardiology,³ and the American College of Cardiology (ACC), American Heart Association (AHA), and Heart Rhythm Society (HRS).¹ These guidelines focused on recommendations for the initial evaluation of patients presenting with syncope and for avoiding unnecessary testing.¹

Adopting a standardized approach to syncope based on guidelines offers an evidence-based opportunity for delivering high-value care. Successful implementation of any clinical guideline requires assessing current practice gaps and potential barriers and facilitators to their implementation. The present study aimed to establish the scope of a practice gap in the evaluation and management of syncope as well as provide some much needed context across subspecialties, practice environments, and provider experience.

Methods

To understand practice patterns in diverse clinical settings, we undertook a mixed-methods multipronged effort that included: 1) a survey sent to more than 1500 clinicians regarding their current practice patterns in the evaluation and management of syncope and a relevant knowledge assessment with the use of clinical vignettes focusing on class I and class III recommendations from the 2017 ACC/AHA/HRS guidelines (Supplemental Table S1); 2) a review of electronic medical records at a large quaternary-care academic medical center to validate the frequency of missed appropriate evaluations and unnecessary tests; and 3) focus groups and key informant interviews of clinicians to understand their practice patterns, awareness of guidelines for syncope, and attitudes toward the guideline recommendations. We sought to elucidate barriers and facilitators to guideline implementation from clinician perspectives. Details of the methodologies are provided in the Supplemental Methods.

Results

Current practice survey

The demographics of the participating clinicians are presented in Supplemental Table S2. The distribution of cases treated by specialty is presented in Supplemental Figure S1.

Orthostatic hypotension was identified as one of the top causes of syncope (Supplemental Fig. S2). The 2017 guidelines recommend that orthostatic vital signs be included as a required part of the physical examination, and the majority of survey respondents (82%) indicated that they obtained orthostatic vitals usually or always (Table 1). Cardiac syncope was another prominent cause of syncope, and electrocardiograms (ECGs) were reportedly obtained by 92% of the respondents usually or always upon evaluation (Table 1). Self-reported use of imaging, particularly head computerized tomographic (CT) scans and carotid Doppler, was also evaluated. Only 26% of respondents indicated that they never or rarely ordered a head CT scan for patients with syncope, whereas 14.3% use it usually or always and 20.1% obtain a carotid Doppler half of the time or usually/always in the evaluation of syncope. The survey also revealed potential overuse of blood tests with 67% stating that they usually or always obtain a complete metabolic panel blood test and 65% usually ordering complete blood counts. Notably, nearly half of the respondents reported testing for illicit drugs.

Table 1.

Testing modalities in evaluation of syncope

Test	Never/rarely	Occasionally	About half of the time	Usually/always	Don’t know
Orthostatic vital signs	1.9%	1.3%	14.3%	81.8%	0.6%
Electrocardiogram (ECG)	0.6%	3.9%	2.6%	92.2%	0.6%
Head computerized tomographic (CT) scan	26%	32.5%	26%	14.3%	1.3%
Echocardiography	23.4%	28.6%	21.4%	26%	0.6%
Carotid Doppler	50%	29.2%	10.4%	9.7%	0.6%
Pulmonary CT	30.5%	62.3%	5.2%	.6%	1.3%
Troponin	24%	30.5%	22.1%	22.7%	0.6%
Complete blood count	11.7%	10.4%	12.3%	64.9%	0.6%
Metabolic panel	7.1%	13.6%	11.7%	66.9%	0.6%
Drug screen	19.5%	31.8%	28.6%	18.8%	1.3%
Stool hemoccult	61%	33.1%	3.9%	1.3%	0.6%
Urine analysis	28.6%	31.8%	13.6%	24.7%	1.3%

Correct answers to clinical vignettes varied largely based on the types of questions asked, with the highest rates of correct answers for hospital medicine clinical vignettes and substantially lower rates of correct answers for cardiology clinical vignettes (Table 2). Ninety percent of respondents correctly selected fluid administration as their first treatment of a patient with orthostasis. Only 60% of respondents recognized the need for an ECG in an emergency medicine scenario where an ECG was indicated, despite the reported routine acquisition of ECGs based on the clinical practice survey. Seventy-seven percent correctly identified the need for inpatient monitoring in a scenario with high-risk features for an arrhythmogenic event, and 78% correctly selected ECG in a scenario with examination findings of severe aortic stenosis. Consistent with the reported practice patterns of neuroimaging, in 2 of the clinical vignettes, inappropriate head CT imaging and carotid Doppler testing were selected by 15% and 53% as initial testing, respectively.

Table 2.

Correct answer response rates by specialty for clinical vignette questions

Vignette	Specialty	Appropriate answer selected
Emergency medicine (EM)	EM	76.5%
	HM	52.4%
	Cardiology	50.0%
Hospital medicine (HM)	EM	62.5%
	HM	55.0%
	Cardiology	84.8%
Cardiology	EM	6.5%
	HM	5.1%
	Cardiology	27.6%

Clinical chart review

Chart review of patients presenting to the emergency department or admitted to an academic medical center with a primary or secondary diagnosis of syncope was analysed on a question-by-question basis. Among the most emphasized aspects of the 2017 guidelines, the majority of the charts reviewed (68%) did not meet thresholds for detailed history and physical examination. Cardiogenic and neurologic causes of syncope were assessed in 88% and 76% of charts, respectively. Documentation of orthostatic vitals was lacking in 33% of the cases. Ninety-seven percent of charts had ECGs ordered and performed. In addition, ~75% of adjunct cardiovascular testing (predominantly echocardiography) and ~90% of neuroimaging (predominantly head CT) were performed in cases in which cardiogenic and neurogenic syncope, respectively, were not high on the differential as scored by the chart reviewers.

Focus group

The interviews asked about clinicians’ familiarity with the guideline and whether they perceived there to be a practice gap. Two overarching themes emerged from the qualitative analysis. First, specific knowledge of the 2017 guidelines was generally low (see Supplemental Data Obtained from Focus Groups). A second identified theme was that most clinicians recognized that they tend to do more testing than they feel is clinically indicated. Providers reported compliance with low-impact testing, such as orthostatic vitals and ECG acquisition, but felt they do more high-cost testing than they know is needed. This is consistent with the high utilization of adjunct testing, specifically laboratory testing, cardiovascular imaging, and neuroimaging.

Discussion

In this study, we found a significant gap between the actual clinical practice of syncope assessment and management vs recommendations in the 2017 ACC/AHA/HRS¹ and 2018 European Society of Cardiology³ guidelines. To establish the practice gap, we used a mixed-methods approach with surveys, chart reviews, and focus group sessions. In particular, we observed: 1) deficiencies in documentation; 2) lack of knowledge of recommendations; and 3) underutilization of inexpensive tests and overutilization of certain expensive tests. Based on the extent of the practice gap identified, even modest reductions in unnecessary testing by adherence to the guidelines would result in substantial savings to health care systems. This is important given that others have found that adherence to the guideline may actually increase hospital readmission rates,⁴ an outcome that was not the focus of the current work.

Clinical vignettes revealed lack of knowledge of current guidelines and may underestimate the issue based on characteristics of the clinicians who elected to respond to the survey. The primary knowledge gap was an overestimation of the prevalence of neurogenic syncope and concomitant overutilization of unnecessary neuroimaging procedures. When comparing interview, clinical vignette, and chart review assessments, we observed differences in practice patterns. For example, > 90% of clinicians selected to order orthostatic vital signs in the clinical vignette, yet our chart review indicated that orthostatic vital signs were missed more than 30% of the time. This suggests that some of the documentation deficiencies reflect a practice gap rather than a knowledge gap.

Our assessment also revealed reliance on high-cost testing over clinical assessment. The impetus for ordering more expensive tests may partly reflect a knowledge gap in understanding the specific recommendations in the 2017 guidelines. However, the inappropriate ordering of tests is likely to be more complex than just a lack of insight or knowledge. A more comprehensive understanding of the rationale will be needed to inform any strategy development to decrease inappropriate utilization of high-cost interventions.

Importantly, our findings highlight a bias of testing over cognitive tasks in clinical practice and management of syncope. Consistently, in chart reviews and clinical vignettes and surveys, we observed an emphasis on testing over documentation of cognitive tasks. This finding likely reflects a larger trend in medicine to rely on objective data (eg, laboratory values or imaging). This practice pattern may be a reflection of the ready availability of testing modalities, lack of confidence in clinical assessment skills, insufficient time to document cognitive reasoning, fear of litigation, reimbursement incentives, or other factors. To narrow the guideline-practice gap, a better understanding of the drivers for testing over cognitive tasks will be needed. A thoughtful approach to designing order sets and care pathways that coincide with guideline recommendations would have the potential to improve documentation of cognitive tasks and augment the use of low-cost, high-impact interventions. In addition, clinical risk tools such as the Canadian Syncope Score may alleviate some of these issues.⁵

In summary, a guideline-practice gap does exist for syncope, highlighted by the overutilization of costly laboratory and imaging tests. Although a knowledge gap was also identified, strategies that primarily or exclusively focus on knowledge deficiencies alone are unlikely to close the guideline-practice gap. Health care systems need to assess additional factors crossing over many different aspects of care delivery that contribute to the guideline-practice gap and apply a multifaceted approach to reduce gaps in guideline-recommended care and current practice in the assessment and management of syncope.