Risk preference and laboratory use

Risk preference and laboratory use

The Journal of Emergency Medune, Vol 6. pp 79-83. Printed 1988 m the USA . CopyrIght 0 1988 Pergarror Journals Ltd sumer protection. Severa...

268KB Sizes 0 Downloads 3 Views

The Journal

of Emergency


Vol 6. pp 79-83.



m the USA . CopyrIght


1988 Pergarror



sumer protection. Several teas contain naturally occurring coumarins and should be avoided by patients on anticoagulants. Although serious complications are rare, at least four fatalities from herbal tea consumption have been reported in the United States. With the growing consumption of health food, exposures to toxic herbal products can be expected to increase. A list of 27 potentially toxic tea constituents and their clinical toxicity is provided. [Robert L. Wears, MD, FACEP] Editor’s Note: Not all toxins are drugs, nor are they routinely picked up on toxicologic screening. Patients may need to be asked explicitly about herbal tea consumption, since they will generally be unaware of any toxic potential.

0 THE LEGAL STATUS OF CONSENT OBTAINED FROM FAMILIES OF ADULT PATIENTS TO WITHHOLD OR WITHDRAW TREATMENT. Areen J. JAMA. 1987; 258:229-235. Physicians routinely turn to family members to make treatment decisions for patients who are incapable of making decisions for themselves. However, there is no basis in common law for relying on a family member as a decisionmaker for an adult patient unless he has formally been appointed the patient’s legal guardian. The emergency exception to informed consent by definition does not apply when the issue at hand is whether to withhold or withdraw life-prolonging treatment. Fortunately, a new legal standard is emerging to govern the care of patients who have not given advance directives as to their wishes. In a growing number of states the law has embraced medical custom so that physicians may rely on consent to withhold treatment obtained from the families of adult patients who cannot speak for themselves. Patients generally must be in a terminal condition, incapable of deciding for themselves, and the family should be acting in good faith (as far as can be determined by facts reasonably available in the ordinary practice of medicine). There is less agreement on exactly which modalities of treatment may or may not be withheld and on the particular family member or members empowered to speak for the patient. [Robert L. Wears, MD, FACEP] Editor’s Note: Emergency physicians should inform themselves about the legal status of consent obtained from family members in the state in which they practice. Fortunately, the growing number of court decisions that authorize family consent increases the likelihood that courts in other states will follow suit.

0 GEOGRAPHIC VARIATIONS IN MORTALITY FROM MOTOR VEHICLE CRASHES. Baker SP, Whitfield RA, O’Neill B. NEngl JMed. 1987; 316:1384-1387. Mortality rates for motor vehicle accidents were calculated by count for the 48 contiguous United States for 1979 through 1981. These rates were correlated with population density and with average per capita income, based on 1980 censusdata. Mortality was found to be inversely correlated with population density (r= - 0.57), with the sparsely populated middle-western and western inland regions showing high mortality rates. Death rates were also inversely correlated with per capita income, but to a lesser degree (r= - 0.23). The authors speculate that in rural areas, reduced accessto emergency facilities and trauma centers, poorer roads, higher speeds, increased use of jeeps and small trucks, and decreased use of seatbelts may all contribute to the higher mortality rates. [Sari L. Hart, MD] Editor’s Note: This study suggests that location accounts for approximately one third of the observed variation in mortality rates. The mechanism by which it exerts an effect remains unknown.

0 TOXIC EFFECTS OF HERBAL TEAS. Ridker PM. Health. 1987; 42:133-136. Many herbal teas contain toxic ingredients that have occasionally caused serious gastrointestinal, hematologic, cardiac, or nervous system disease. The severity of illness following herbal tea use has ranged from contact dermatitis to fulminant hepatic failure and death. Most of these teas are available in health food stores and there is currently no requirement that their toxic potential be labeled for con-

Arch Environ

q RISK PREFERENCE AND LABORATORY USE. Nightingale SD. Med Decis Making. 1987; 7:168-173. Physicians vary widely in their use of laboratory tests in managing patients; this variation persists after correction

Abstracts-designed to keep readers up to date by providing original abstracts of current literature from all fields relating to emergency medicine-are prepared by the Emergency Medicine Residents of the University Hospital of Jacksonville, Jacksonville, Florida; and the Residency in Emergency Medicine in Denver General, St. Anthony’s, =ZIZZZX St. Joseph’s, Porter Hospital and University of Colorado Health SciencesCenter, Denver, Colorado, with editoriZIZZZGZ al notes by Robert Wears, MD, University Hospital of Jacksonville, and Peter Rosen, MD, Editor-in-Chief, JEM. 0736-467908 79

$3.00 + .OO

80 for educational background, practice setting, and patient outcome. The effect of educational intervention on physician laboratory use has been transient at best. This study compared physicians’ test usage in an outpatient setting with their preference for, or aversion to, risk as measured by a risk preference questionnaire. Of 137 physicians surveyed, the group of 91 (67%) who exhibited risk preference in the face of certain loss ordered significantly more laboratory tests than did the group showing risk aversion. This effect was more pronounced in the physicians showing the greater degree of risk preference. This effect could not be explained by differences in level of postgraduate education and did not correlate with physician knowledge as measured by scores on the American Board of Internal Medicine Certifying Examination. No effect was noted for risk preference in the face of certain gain. The cognitive and subliminal processes that determine risk preference in the face of loss, whatever they may be, appear to have a substantial influence on physician test ordering behavior. [Robert L. Wears, MD] Editor’s Note: Emergency physicians are commonly accused of ordering too many lab tests. This paper demonstrates how little we know about physicians’ test-ordering behavior and offers an explanation of the general failure of current attempts to modify it.


shenker E, et al. Ann Emerg Med. 1987; 16:851-856. Accurate evaluation of acute chest pain in the emergency department is often difficult, as the ECG and cardiac enzymes may be normal or nondiagnostic early in the course of acute myocardial infarction. This prospective observational study was done to evaluate the potential usefulnessof serum myoglobin in the early identification of acute myocardial infarction. Fifty-nine patients admitted to a community hospital for chest pain had serum myoglobin levels determined at 0, 3, and 6 hours after presentation to the emergency department. All of the 21 patients ultimately meeting criteria for infarction had elevated serum myoglobin at three hours (sensitivity = 1.O, 95% confidence interval 0.84 to l.OO), and 29 of the 38 patients without infarction had normal levels (specificity= .76, 95% confidence interval 0.60 to 0.89). The authors conclude that serum myoglobin may be useful in the early identification of patients with acute myocardial infarction and may permit for earlier therapeutic intervention in such patients. [John D. Hernandez, MD, FACEPI Editor’s Note: These results should be viewed as preliminary owing to the low precision of the estimates of sensitivity and specificity. Also, the incidence of infarction here (36%) was higher than that experienced in emergency department patients with chest pain (5% is typical). Thus, if the test was generally applied in the emergency department, only about 18% of those testing positive would actually be suffering infarction. Confirmation in a larger study is necessary prior to widespread acceptance.

The Journal

of Emergency



Emerg Med. 1987; 161733-737. Delayed diagnosis and inadequate treatment of carpal navicular fractures can result in delayed union, pseudoarthrosis, or avascular necrosis. This prospective study estimated the frequency of negative initial radiographs and assessedthe validity of three clinical signs of navicular fracture: snuffbox tenderness, pain on supination against resistance, and pain on longitudinal compression of the thumb. Eighty-five patients with a history of wrist trauma who did not have other wrist or forearm fractures were studied; 40 ultimately were shown to have fractures. Eight of these (20%) were not visible on the initial radiographs. Snuffbox tenderness and pain on supination against resistance were present in all patients with fractures and in one patient without (sensitivity= loo%, specificity= 98%). Pain on longitudinal compression of the thumb was present in 39 patients with fractures and in two without (sensitivity = 97.5%) specificity = 97.8%). Navicular fractures have a high incidence of false negative radiographs on initial presentation; patients presenting with any of these clinical signs and a negative radiograph should be properly immobilized and reexamined in ten to I4 days to prevent potentially serious complications. [Robert L. Wears, MD, FACEP] Editor’s Note: There is an old saying to the effect that there is no such thing as a sprained wrist-they’re all occult navicular fractures.





Young MJ, Med. 1987;

McMahon LF, Stross JK. Arch Intern 147:1219-1222. Several clinical prediction rules for identifying patients at high risk of acute myocardial infarction have been developed over the past several years; virtually all of them originated in academic medical centers. This study attempted to validate two such rules in community hospitals by retrospectively analyzing the records of 397 patients admitted to rule out myocardial infarction. One rule based on the initial ECG that had identified patients with a 17 times greater risk of death and a 23 times greater risk of life-threatening complications in an academic center was not associated with a statistically significant ability to identify high-risk patients in the community hospital population. Similarly, a second rule based on the patient’s course in the initial 24 hours in the hospital was significantly less effective than it had been in an academic center. Clinical prediction rules developed in academic centers should be validated in other settings before being generally applied. [Robert L. Wears, MD, FACEP] Editor’s Note: The problem of accurately and safely selecting patients suffering acute myocardial infarction out of the mass of patients presenting with chest pain continues to