Objectives

1. How the ACEP Clinical Policies Standardize and Improve Patient CareAndy Jagoda, MD, FACEPProfessor of Emergency MedicineMount Sinai School of MedicineNew York, New York

2. Objectives • Introduce the process of how clinical policies / practice guidelines are developed • Discuss the medical legal implications of practice guidelines • Use examples from practice guidelines on brain injury and headache to demonstrate applications to patient care

3. ACEP and Clinical Policies • Committee formed in 1987 • Meetings with DM Eddy • Fatal flaw: decision to concentrate on symptoms or complaints • Topics chosen from complaints with high frequency, high risk, or high cost • New directions

4. Clinical Policies / Practice Guidelines • Over 3000 in existence • ACEP: 15 • Chest Pain 1990 • Sunsetting - no longer distributed • Archive – reviewed and kept on website • National Guideline Clearinghouse: • www.guideline.gov • Over 550 guidelines registered

5. Why are clinical policies being written? • Differentiate “evidence based” practice from “opinion based” • Clinical decision making • Education • Reducing the risk of legal liability for negligence • Improve quality of health care • Assist in diagnostic and therapeutic management • Improve resource utilization • May decrease or increase costs • Identify areas in need of research

6. Guideline Development: Time and Cost • Time: 1- many years • Cost: • ACEP: $10,000 • AANS: $100,000 • AHCPR: $1,000,000 • WHO: $2,000,000

7. Interpreting the literature • Terminology • Mild traumatic brain injury • Patient population • Children vs adults • CT + vs CT - • Interventions / outcomes • Any brain lesion • Lesion requiring ns intervention

8. Critically Appraising Clinical Policies • Why was the topic chosen • What are the authors’ credentials • What methodology was used • Was it field tested • When was it written / updated

9. Do clinical policies change practice? • ACEP Chest Pain Policy: Emergency physician awareness. Ann Emerg Med 1996; 27:606-609Clinical policy published in 1990 • 163 / 338 (48%) response to survey • 54% aware of the policy • Majority of those aware did not know content • Wears. Headaches from practice guidelines. Ann Emerg Med 2002; 39:334-337 • 60% of practicing EPs use narcotics as first line medications • Canadian Headache Society. Guidelines for the diagnosis and management of Migraine in clinical practice. • Can Med Assoc J 1997; 156:1273-128US Headache Consortium. www.aan.com/public/practice guidelines

10. Cabana et al. Why don’t physicians follow clinical practice guidelines. JAMA 1999; 282:1458-1465 • Review of 76 articles dealing with adherence • Barriers to physician adherence identified: • Lack of familiarity (more common than lack awareness • Lack of agreement • Lack of self-efficicy (lack of access to intervention, lack of resources / support / social systems) • Thrombolytics in stroke • Lack of outcome expectancy (lack of confidence that an intervention will change the outcome) • Amiodarone in v-fib • Patient related barriers (inability to overcome patient expectation) • Ottawa ankle rules

12. Medical Legal Implications • Clinical policies can set standards for care and have been used in malpractice litigation • May protect against “expert” testimony • Regional practice vs national “standards” • Steroids in spinal trauma • Clinical policies developed using flawed methodology may be challenged • Consensus / Policy statements

13. “Do the authors seriously believe that patients with a first seizure can be discharged from the ED after a serum glucose and a pregnancy test without additional lab testing? This flies in the face of common sense and would perhaps be considered malpractice in some parts of the country.” Journal Reviewer 1995

14. Medical Legal Implications • 1994 Physician Payment Review Commission • 32 cases reviewed where guidelines were used to demonstrate departure from “standard of care” • 259 insurance claims carriers: 6.6% cited guidelines • 980 attorneys surveyed: • 75% were aware of practice guidelines • 36% reported cases with important role • 25% reported that they had influenced a decision to settle or not take a case

15. Deposition of Dr. X in a case of missed meningitis Q. Do you read the policies of the American College of ER physicians? A. I don’t recall reading that policy. Is it something published by ACEP? Q. Yes. A. I don’t recall reading it.

16. Deposition of Dr. X in a case of missed meningitis Q. So if torodol releives a headache, does that cause you to believe the patient does not have meningitis in a patient in whom you are suspecting meningitis a a possible cause of their headache A. It’s an indicator that would decrease the likelihood. Q. If torodol relieved their headache, would you rely on that as a factor in ruling out meningitis? A. It is part of the package.

17. Clinical Policy: Critical issues in the evaluation and management of patients presenting to the ED with acute headache. Ann Emerg Med 2002; 39:108-122 • Does a response to therapy predict the etiology of an acute headache? • Level A recommendation: None • Level B recommendation: None • Level C recommendation: Pain response to therapy should not be used as the sole indicator of the underlying etiology of an acute headache

18. Guideline Development • Informal Consensus • Formal consensus • Evidence based

19. Informal Consensus • Group of experts assemble • “Global subjective judgement” • Recommendations not necessarily supported by scientific evidence • Limited by bias

20. Informal Consensus: Examples • MAST trousers in traumatic shock • Hyperventilation in severe TBI • Oxygen for patient with chest pain • Magnesium level for patients who have had a seizure

21. Formal Consensus • Group of experts assemble • Appropriate literature reviewed • Recommendations not necessarily supported by scientific evidence • Limited by bias and lack of defined analytic procedures

22. Formal Consensus: Limitations • Plain film radiographs after head trauma • Phenytoin to prevent development of epilepsy after head trauma

23. Evidence Based Guidelines • Define the clinical question • Focused question better than global question • Outcome measure must be determined • Grade the strength of evidence • Incorporate practice patterns, available expertise, resources and risk benefit ratios • External validity

24. Description of the Process • Medical literature search • Secondary search of references • Articles graded • Recommendations based on strength of evidence • Multi-specialty and peer review

25. Description of the Process Strength of evidence (Class of evidence) • I: Randomized, double blind interventional studies for therapeutic effectiveness; prospective cohort for diagnostic testing or prognosis • II: Retrospective cohorts, case control studies, cross-sectional studies • III: Observational reports; consensus reports Strength of evidence can be downgraded based on methodologic flaws

26. Description of the process: • Strength of recommendations: • A / Standard: Reflects a high degree of certainty based on Class I studies • B / Guideline: Moderate clinical certainty based on Class II studies • C / Option: Inconclusive certainty based on Class III evidence

27. Evidence Based Guidelines: Limitations • Different groups can read the same evidence and come up with different recommendations • MTBI • t-PA in stroke • Steroids in spinal trauma

28. Concussion in Sports • American Academy of Neurology • Evidence based methodology • Concussion: a trauma induced alteration in mental status, with or without LOC • Confusion and amnesia are the hallmarks • Justifications: • Repeated concussions can cause cumulative brain injury • Provide physicians with guidelines to help overcome the bias in management from athletes, coaches, media, spectators

29. Guidelines for the management of concussion in sports. American Academy of Neurology • Grade 1: Confusion: No LOC or amnesia • remove from event for 20 minutes • 2 grade 1 concussions; no play for one day • 3 grade 1 concussions; no play for 3 months • Grade 2: No amnesia; + amnesia • remove from event • no play for 1 week • 2 grade 2 concussions; no play for 1 month • 3 grade 2 concussions; no play for the season • Grade 3: LOC • hospital evaluation • no play for 1 month • 2 grade 3 concussions: no play for the season

30. Guidelines for Prehospital Management of TBI • Multidisciplinary: Brain Trauma Foundation / Grant from NHTSA • Evidence Based • Prehospital care is the “first link” in appropriate care in TBI • Prehospital providers play a key role in determining the need for trauma center access

31. Guidelines for Prehospital Management of TBI • Identifies the need for focused prehospital research • Establishes need to perform a field assessment including vital signs, GCS, pupils • Guidelines: Hypotension and hypoxia must be prevented • Option: Secure the airway with intubation • Option: Herniation should be treated with hyperventilation

32. ED Management of MTBI in Adults • Multidisciplinary group funded by a grant from the IBIA: ACEP, ASNR, AANS • Evidence based: Three Questions: • Is there a role for plain film radiographs in the assessment of MTBI in the ED • Which patients with acute MTBI should have a noncontrast head CT in the ED • Can a patient with MTBI be safely discharged from the ED if a noncontrast CT shows no evidence of acute injury

33. ED Management of MTBI in Adults • Blunt trauma to the head within 24 hours of presentation to the ED • Any period of post-traumatic LOC or PTA • A GCS score at initial evaluation in the ED of 15 • A nonfocal neurologic exam • Age greater than 15 years

34. ED Management of MTBI in Adults • Outcome measures in the TBI literature: • Acute traumatic abnormality on CT • Clinical deterioration • Need for neurosurgical intervention • Development of post-concussive syndrome • Outcome measure for this policy: • Presence of an acute intracranial abnormality on noncontrast head CT

35. Is there a role for plain film radiographs in the assessment of MTBI in the ED • Masters 1987 NEJM: Prospective study 7035 pts. • Flawed methodology. 63% with + xray had - CT; 50% with +CT had negative xray • Skull films have low sensitivity for intracranial lesions • Hoffman 2000 Lancet: Meta-analysis • 20 articles reviewed out of 200 identified • Sensitivity .13-.75; PPV of skull fracture in predicting +CT .4 • Specificity .9-.99; NPVof skull fracture in predicting +CT .94 • Recommendation Level B: Skull films are not recommended in the evaluation of MTBI; although the presence of a skull film increases the likelihood of an intracranial lesion, its sensitivity is not high enough to allow it to be a useful screen

36. Which patients with acute MTBI should have a noncontrast head CT in the ED • Various studies in patients with a GCS of 15 report a 5% - 15% incidence of an intracranial lesion • .3-.5 incidence of lesions needing neurosurgical intervention • Stiell 2001 Lancet. Prospective 3021 patients • Outcome: Neurosurgical intervention • 67% had CT; only 33% of the remainder had telephone follow-up • Survey used to determine “insignificant” lesions: patients with those lesions were not followed up • 5 high risk predictors: failure to reach GCS 15 within 2 hours; suspected open skull fracture; sign of basal skull fracture; vomiting more than once; age over 64 • High risk factors were 100% sensitive identifying need for neurosurgery and would decrease CT by 68%

37. Which patients with acute MTBI should have a noncontrast head CT in the ED • Haydel 2000 NEJM; Class I study; 2 phases • Phase I 520 patients to establish predictive criteria • Phase II 909 patients to validate criteria • 7 predictors identified with 100% sensitivity for predicting intracranial lesion. • Use of criteria would decrease head CT by 22% • No follow-up provided after discharge • Recommendation Level A: A head CT is not recommended in those patients with MTBI who do not have HA, vomiting, age > 60, drug or ETOH intoxication, deficits in short term memory, physical evidence of trauma above the clavicle, or seizure.

38. Can a patient with MTBI be safely discharged from the ED if a noncontrast CT shows no evidence of acute injury • Stein 1992 J Trauma. Retrospective • 1339 patients with negative CT, none deteriorated • Dunham 1996 J Trauma Infect Crit Care. Retrospective review of a prospectively collected data base • 2587 patients, no patient with a negative CT deteriorated; those patients who did deteriorate (without initial CT), did so within 4 hours • Nagy 1999 J Trauma Infect Crit Care. Retrospective • 1190 patients with CT and admission • No patient with a negative CT deteriorated (spectrum bias towards sicker patients) • Recommendation Level C: Patients with MTBI who are 6 hours out from their injury and who have a head CT that does not demonstrate acute injury can be safely discharged from the ED

39. Severe TBI Guidelines • AANS / Grant from the BTF • Standards • prophylactic hyperventilation should be avoided • use of glucocosteriods is not recommended • prophylactic phenytoin is not recommended for late sz • Guidelines: • hypotension and hypoxia must be avoided • ICP monitoring is appropriate • mannitol is effective for controlling raised ICP • Options • Hyperventilation may be necessary for brief periods when there is acute neurologic deterioration • AEDs may be used to prevent early posttraumatic sz

40. Huizenga et al. Guidelines for the management of severe head injury: Are emergency physicians following them? Acad Emerg Med 2002; 9:806-812 • 319 / 566 survey responses (56%) to 3 cases • 78% corrected hypotension • 46% used prophylactic hyperventilation • 14% used glucocorticoids • 8% used prophylactic mannitol • Authors conclusion: A majority of emergency physicians are managing TBI according to the guidelines • My conclusion: 7 years post publication, a significant number of emergency physicians are not correctly managing severe TBI

41. Conclusions • Evidence based clinical policies are useful tools in clinical decision making • Clinical policy development must be rigorous • Clinical policies do not create a “standard of care” and do not necessarily override “expert witness” • Clinical policy dissemination continues to be a challenge