Search

Objectives: Undiagnosed human immunodeficiency virus (HIV) infection is not uncommon among patients who seek care in urban inner-city emergency departments (EDs). The optimal method for providing appropriate HIV counseling, testing, and referral in this setting is unclear. A previous evaluation of an outpatient referral system for HIV testing among patients seen in the ED demonstrated that only 11% returned to be tested; of those tested, 7% were seropositive. The purpose of this study was to evaluate the effect of a financial incentive on the proportion of referred ED patients who completed outpatient HIV counseling and testing. Methods: This was a quasi-experiment performed at an urban county teaching hospital. ED patients identified using Centers for Disease Control and Prevention guidelines as being at risk for HIV infection were referred for outpatient HIV counseling and testing at our institution within one week. This study was divided into three study periods. During the first and third periods, no financial incentive was offered for completing HIV counseling and testing. During the second period, a $25 incentive was offered for completing HIV counseling and testing. Results: During the study, 372 patients were referred for HIV counseling and testing. During the control periods, 20 (8%) of 252 patients completed HIV counseling and testing; during the intervention period, 27 (23%) of 120 patients completed HIV counseling and testing (odds ratio, 3.4; 95% confidence interval = 1.8 to 6.3). Of the 47 patients who completed HIV counseling and testing, none (0%; 95% confidence interval = 0% to 8%) tested positive for HIV infection. Conclusions: The use of a modest financial incentive increased the proportion of patients who completed HIV counseling and testing after being referred from the ED. It is unclear whether the use of a financial incentive increases the identification of HIV-infected patients unaware of their status or if it is cost-effective.

The goals of this study were to quantify delays in diagnosing HIV infection in patients presenting for medical care prior to the identification of their HIV infection, and to determine characteristics of visits preceding the identification of HIV infection. This was a retrospective cohort study of consecutive newly diagnosed HIV infected patients in a large urban setting. Patients were predominantly minorities and identified from all inpatient and outpatient HIV tests performed between December 29, 1998 and December 27, 1999. Data were collected from all emergency department and clinic visits for each patient included in this study for a 3-year period preceding the first diagnosis of HIV infection. Two hundred seventy-eight patients tested positive for HIV during the study period. Of these, 76 (27%) met inclusion criteria. Twenty-three patients made a total of 53 health care visits preceding the diagnosis of their HIV infection. The median delay in diagnosis of HIV infection was 2 days (interquartile range [IQR], 1-22 days; range, 0-1093 days) for all subjects, although 19 (25%; 95% confidence interval [CI]: 16%-36%) patients had a delay in diagnosis of 30 days or more. Among those patients who had made prior health care visits, the median delay in diagnosis was 112 days (IQR, 33-690 days; range, 2-1093 days). No specific risk factor, historical clue, physical examination finding, or laboratory finding reliably identified patients with HIV infection. Documentation in the medical record of specific risk factors and clinical characteristics suggestive of HIV infection was poor. This study documented missed opportunities and delays in diagnosis of patients with unrecognized HIV infection. Clinicians must maintain a high index of suspicion for HIV infection in all patients.

The use of confidence intervals in reporting results of research has increased dramatically and is now required or highly recommended by editors of many scientific journals. Many resources describe methods for computing confidence intervals for statistics with mathematically simple distributions. Computing confidence intervals for descriptive statistics with distributions that are difficult to represent mathematically is more challenging. The bootstrap is a computationally intensive statistical technique that allows the researcher to make inferences from data without making strong distributional assumptions about the data or the statistic being calculated. This allows the researcher to estimate confidence intervals for statistics that do not have simple sampling distributions (e.g., the median). The purposes of this article are to describe the concept of bootstrapping, to demonstrate how to estimate confidence intervals for the median and the Spearman rank correlation coefficient for non-normally-distributed data from a recent clinical study using two commonly used statistical software packages (SAS and Stata), and to discuss specific limitations of the bootstrap.

Background. This study reports the epidemiologic features, survival rates, and neurologic outcomes of the largest population-based series of pediatric out-of-hospital cardiopulmonary arrest patients with prospectively collected data.Methods. Secondary analysis of data from a prospective, interventional trial of out-of-hospital pediatric airway management conducted from 1994 to 1997 (Gausche M, Lewis RJ, Stratton SJ, et al. JAMA. 2000;283:783-790). Consecutive out-of-hospital patients from 2 large urban counties in California Results. In 599 patients, 601 events were studied (54% were 3 doses of epinephrine or were resuscitated for >31 minutes in the emergency department.Conclusions. The 8.6% survival rate after out-of-hospital pediatric cardiopulmonary arrest is poor. Administration of >3 doses of epinephrine or prolonged resuscitation is futile.

The purpose of this study is to define the prevalence and significance of apparent life-threatening events among infants in the out-of-hospital setting.This was a retrospective, cohort, outcome study of infants for whom a caregiver activated the emergency medical services (EMS) system. For purposes of the study, an apparent life-threatening event was defined as an episode of apnea, skin color change, or change in muscle tone. Study data characteristics included initial physical appearance, work of breathing, circulation skin signs, pulse rate, respiratory rate, and overall concern for the chief complaint as interpreted by EMS personnel.Sixty (7.5%) of 804 infants encountered by EMS during the study period met our criteria for apparent life-threatening event. Mean age was 3.1+/-3.3 months, and 55% were boys. Of the infants with apparent life-threatening event, 50 (83.3%) infants appeared to be in no distress, 8 (13.3%) infants were in mild distress, and 2 (3.3%) infants were in moderate distress. General physical appearance, work of breathing, circulatory signs, respiratory rate, and pulse rate were not clinically abnormal in the study group as a whole. Critical conditions associated with apparent life-threatening event included pneumonia or bronchiolitis (12%), seizure (8%), sepsis (7%), intracranial hemorrhage (3%), bacterial meningitis (2%), dehydration (2%), and severe anemia (2%). Limitations of the study included retrospective design and inability to follow up study patients beyond hospital discharge.An apparent life-threatening event in an infant can present without signs of acute illness and is commonly encountered in the EMS setting. It is often associated with significant medical conditions, and EMS personnel should be aware of the clinical importance of an apparent life-threatening event. Infants meeting criteria for an apparent life-threatening event should receive a timely and thorough medical evaluation.

Objective To compare 2 educational approaches (structured group session vs. individual instruction) of learning application techniques of 2-octylcyanoacrylate (OCA) on wounds repaired in the emergency department. Methods This prospective, nonrandomized, observational study was conducted in an urban hospital emergency department. Medical students, residents, and faculty were trained in the use of OCA either in a standardized group session or individually, based on their availability to attend the group session. Physicians completed a data collection form that included wound characteristics, type of instruction, and number of lacerations previously repaired with OCA. Impressions of time required, difficulty, and postrepair cosmesis were each recorded on a 5-point Likert scale. The 3 scales were totaled to yield a 15-point summary outcome measure. Univariate nonparametric comparisons of measures were performed between physicians with and without group instruction and between those with and without prior OCA experience. Using 35 subjects in each group, this study had a power of 0.95 to detect a difference of 1.5 points in the 15-point summary score, using alpha = 0.05. Results Eighty-one subjects were enrolled; the median summary score was 13 (IQR 12 to 15). There was no statistically significant difference in the summary score, nor any of its 3 components (time saved, difficulty, cosmesis), when comparing physicians with and without group instruction, nor when comparing first-time users to those with prior experience. Conclusion OCA application is an easily acquired skill. Physicians were satisfied with their proficiency in OCA application, regardless of type of instruction received or number of previous lacerations repaired with OCA.

Background: No valid model has been developed to predict survival following out-of-hospital cardiac arrest. The purpose of this study was to develop a prediction model for meaningful survival following out-of-hospital cardiac arrest using variables available during resuscitation. Methods: This was a retrospective cohort study. Consecutive adult cardiac arrest patients were studied between 1994 and 2001. Variables included age, sex, race/ethnicity, arrest location, whether the arrest was witnessed, initial rhythm, whether CPR was performed, patient downtime, paramedic response time, survival to hospital discharge, and Glasgow Coma Score (GCS) at hospital discharge. Classification and Regression Tree analysis was used to develop decision rules to predict meaningful survival, as defined by the patient’s discharge GCS. Results: Of the 754 patients, 16 (2%) survived with a GCS ≥ 13, 15 (2%) survived with a GCS = 14, and 5 (0.7%) survived with a GCS = 15. The decision rule for survival with a GCS ≥ 13 incorporated whether the arrest was witnessed and the patient’s age, resulting in a negative predictive value (NPV) of 99.8%. The rule for survival with a GCS ≥ 14 incorporated the initial arrest rhythm, whether the arrest was witnessed, and the patient’s age, resulting in a NPV of 99.6%. The rule for survival with a GCS = 15 incorporated only the interval between collapse and the initiation of life support, resulting in a NPV of 99.8%. Conclusions: This study reports decision rules for potential meaningful survival following out-of-hospital cardiac arrest with high NPVs for each. Future studies need to be performed to prospectively validate these models. © 2004 Elsevier Ireland Ltd. All rights reserved.

To develop and validate a practical outcome instrument applicable to a broad range of neurologic deficits in children.Reliability testing of a draft version of the Neurologic Outcome Scale for Infants and Children (NOSIC) in 100 children with a wide range of ages and levels of neurologic function was performed. After review of the reliability data by a panel of experts, the NOSIC was revised. Validity and reliability testing of the final NOSIC was performed in a new population of 157 children, 52 with cerebral palsy, motor delay, or language delay. Interrater reliability was assessed using Spearman rank correlation coefficients of two investigators' scores. NOSIC scores were correlated with scores on criterion-standard neuropsychological tests to assess validity.The median NOSIC score for normal children was 98, interquartile range 96-100; the median score for abnormal children was 87, interquartile range 58-96. Interrater reliability of the NOSIC scores of 84 patients rated by both raters demonstrated excellent reliability (rho = 0.77, 95% confidence interval [CI] = 0.62 to 0.88). Correlation of the NOSIC scores of the 127 patients who had neuropsychological testing with applicable criterion standards was rho = 0.63, 95% CI = 0.50 to 0.74.The NOSIC is a practical, reliable, valid, instrument applicable to infants and children with a broad range of neurologic deficits. It should be a useful research tool when neurologic function is an important outcome measure.

On January 17, 1994, an earthquake with a moment magnitude (total energy release) of 6.7 occurred in Northridge, California, leading to the evacuation of patients from several hospitals. We examined the reasons for and methods of evacuation and the emergency-management strategies used. The experience in California may have implications for hospital strategies for responding to any major disaster, including an act of terrorism.From September 1995 to September 1996, we surveyed all acute care hospitals in Los Angeles County that reported having evacuated patients after the Northridge earthquake. Physicians, nurses, hospital administrators, and staff on duty at the hospitals during the evacuation responded to a 58-item structured questionnaire.Eight of 91 acute care hospitals (9 percent) were evacuated. Six hospitals evacuated patients within 24 hours (the immediate-evacuation group), four completely and two partially. All six cited nonstructural damage such as water damage and loss of electrical power as a major reason for evacuation. Five hospitals evacuated the most seriously ill patients first, and one hospital evacuated the healthiest patients first. All hospitals used available equipment to transport patients (blankets, backboards, and gurneys) rather than specialized devices. No deaths resulted from evacuation. One hospital evacuated patients after 3 days and another after 14 days because of structural damage, even though initial inspections had shown no damage (the delayed-evacuation group). Both hospitals required demolition. Some hospitals identified destinations for their evacuated patients independently, whereas others sought the assistance of the Los Angeles County Emergency Operations Center; the two strategies were equally effective.After even a moderate earthquake, hospitals are at risk for both immediate nonstructural damage that may force them to evacuate patients and the delayed discovery of structural damage resulting in permanent closure. Evacuation of large numbers of inpatients from multiple hospitals can be accomplished quickly and safely with the use of available resources and personnel.

Despite the publication of guidelines for the management of pneumonia, significant variation in care continues to exist. While there have been several published reports of quality improvement projects for pneumonia, there are few data on the effectiveness of these efforts in small hospitals. The purpose of this study was to demonstrate that a project implemented by a quality improvement organization in small hospitals would lead to an improvement in care that could not be accounted for by secular trends in the management of pneumonia.Medicare-insured hospital admissions for pneumonia were reviewed from 20 small hospitals in Oklahoma (intervention group) at baseline and after feedback. Project intervention included onsite feedback presentations to the medical staff, samples of performance improvement materials, and comparative measures of performance of predefined quality indicators. A second group of 16 demographically similar hospitals (control group) was selected for review during the same 2 periods. These hospitals subsequently underwent an identical intervention with a follow-up assessment.Statistically significant improvements in process measures were demonstrated in the intervention hospitals, including performance of a sputum (P.01) and blood (P.001) cultures, antibiotic administration within 4 hours of hospital admission (P.001), and administration of the first dose of antibiotic in the emergency department (P.001). These measures in the control hospitals did not change significantly (P =.93, .08, .79, and .52, respectively) during the 2 periods.Improvements in processes of care achieved by the intervention hospitals resulted from activities initiated because of participation in a quality improvement organization-directed project. This study demonstrated the effectiveness of quality improvement activities in very small hospitals.

Previous work has suggested that low-energy biphasic waveform defibrillation (BWD) is followed by less post-resuscitation left ventricular (LV) dysfunction when compared with higher-energy monophasic waveform defibrillation (MWD). To the best of the authors' knowledge, the effect of cardiopulmonary resuscitation (CPR) duration and total ischemia time on LV function after countershock, controlling for waveform type, has not been evaluated.To determine the effect of CPR duration on LV function after MWD and BWD.VF was electrically induced in anesthetized and instrumented swine. After 5 minutes of VF, the animals were randomized to MWD (n = 22) or one of two BWDs (n = 46). If countershock terminated VF but was followed by a nonperfusing rhythm, conventional manual CPR without drug therapy was performed until restoration of spontaneous circulation (ROSC), defined as a systolic arterial pressure60 mm Hg for 10 minutes without vasopressor support. Systolic LV pressure (LVP), LV dP/dt (first derivative of pressure measured over time), and cardiac output (CO) were measured at intervals for 60 minutes postresuscitation. CPR times (times to ROSC) and hemodynamic variables for the three groups were compared. Multivariable linear regression was performed to assess the contribution of defibrillation waveform, total joules, and CPR time on LVP, LV dP/dt, and CO at 15, 30, and 60 minutes postresuscitation.When analyzed as groups, significant differences in median number of shocks to terminate VF, total joules, or CPR time were not observed between waveform groups. Regression analysis demonstrated that increasing CPR time was associated with a significant effect on indices of LV function at 15 and 30 minutes postresuscitation. Global LV function was not influenced by waveform type or total joules.Adjustment for CPR time, a determinant of total myocardial ischemia time, is necessary when defibrillation waveforms are compared for their effect on postresuscitation cardiac function and short-term outcome.

Study purpose: Successful defibrillation is dependent upon the delivery of adequate electrical current to the myocardium. One of the major determinant of current flow is transthoracic impedance. Prior work has suggested that impedance falls with repeated shocks during sinus rhythm. The purpose of this study was to evaluate changes in transthoracic impedance with repeated defibrillation shocks in an animal model of cardiac arrest due to ventricular fibrillation (VF). Methods: VF was electrically induced in anesthetized swine. After 5 min of untreated VF, monophasic or biphasic waveform defibrillation was attempted using a standard sequence of ‘stacked shocks’ (200, 300, then 360 J, if necessary) administered via adhesive electrodes. If one of the first three shocks failed to convert VF, conventional CPR was initiated and defibrillation (360 J) attempted 1 min later. Strength–duration curves for delivered voltage and current were measured during each shock and transthoracic impedance calculated. Animals requiring a minimum of four shocks were selected for study inclusion. Impedance data from sequential shocks were analyzed using mixed linear models to account for the repeated-measures design and the variability of the initial impedance of individual animals. Results: Thirteen animals (monophasic waveform, n =7, biphasic waveform, n =6) required at least four shocks to terminate VF (range 4–6). Transthoracic impedance did not change from the first shock in the 13 animals (46±8 Ω) to the fourth shock (46±9 Ω). In animals receiving more than four shocks, transthoracic impedance likewise did not change significantly from the first to the last shock, which terminated VF. The lack of a significant change in impedance was also observed when animals were analyzed according to defibrillation waveform. Conclusion: Transthoracic impedance does not change significantly with repeated shocks in a VF cardiac arrest model. This is likely due to the lack of reactive skin and soft tissue hyperemia and edema observed in non-arrest models.

The emergency department (ED) and HIV specialty clinics are primary sources of care for persons infected with HIV. HIV disease may be complicated by vague and complex symptomatology, and determining the degree of illness at triage is often difficult. The goals of this project were to characterize the ED presentation of HIV-related conditions, to develop a clinical decision rule to triage HIV-infected patients, and to validate the rule in clinical practice.The study population consisted of ambulatory patients with self-reported HIV infection who presented for care to the ED of a 553-bed public hospital that serves a medically indigent, minority population. An Illness Severity Instrument was developed by an expert panel to serve as the criterion standard for defining medical urgency for HIV-infected patients presenting to the ED for care. Two phases of the study were conducted. Data from the first phase, a noninterventional cohort study, were used to develop a clinical decision rule for the ED triage of HIV-infected patients. The second phase was a prospective validation of the clinical decision rule.During phase I, data from 542 patient visits were collected. Data from 441 (81%) patient visits were used in a classification and regression tree (CART) analysis to produce a decision rule, the Clinical Triage Instrument. During phase II, the prospective validation of the Clinical Triage Instrument, 156 patient visits occurred. Of these, 88 (56%) patient visits were triaged using the Clinical Triage Instrument and could be scored using the Illness Severity Instrument. The Clinical Triage Instrument accurately triaged 45 [51%; 95% confidence interval (95% CI) = 40% to 62%] patient visits, undertriaged 11 (13%; 95% CI = 6% to 21%) patient visits, and overtriaged 32 (36%; 95% CI = 26% to 47%) patient visits. Sensitivities and specificities for determining emergent, urgent, and nonurgent medical conditions by the Clinical Triage Instrument were 56% (95% CI = 31% to 75%) and 84% (95% CI = 74% to 92%), 71% (95% CI = 55% to 84%) and 39% (95% CI = 25% to 55%), and 18% (95% CI = 6% to 37%) and 93% (95% CI = 84% to 98%), respectively. The positive and negative predictive values for determining an emergent medical condition using the Clinical Triage Instrument were 48% (95% CI = 26% to 70%) and 88% (95% CI = 78% to 95%), respectively. The positive and negative predictive values for determining a nonurgent medical condition using the Clinical Triage Instrument were 56% (95% CI = 21% to 86%) and 71% (95% CI = 60% to 81%), respectively.The Clinical Triage Instrument was not sufficiently accurate for clinical use. Until accurate and reliable triage methods are developed, all patients infected with HIV who present to the ED for care should receive timely evaluation and care.

We sought to create a clinical decision rule, on the basis of variables available to out-of-hospital personnel, that could be used to accurately predict severe injury in pediatric patients involved in motor vehicle crashes as occupants.We analyzed the National Automotive Sampling System database, a national probability sample, using pediatric patients up to 15 years old (occupants only) involved in motor vehicle crashes from January 1993 to December 1999. The National Automotive Sampling System database includes patients from regions throughout the country, weighted to represent a nationwide sample. Twelve out-of-hospital variables were used in classification and regression tree analysis to create a decision rule separating children with severe injuries (Injury Severity Score [ISS]or =16) from those with minor injuries (ISS16). Misclassification costs and complexity parameters were selected to yield a decision tree with appropriate sensitivity and specificity for the identification of severely injured patients, while also being simple and practical for out-of-hospital use. Probability weights were used throughout the analysis to account for the sampling design and sampling weights.Using a sample size of 8,392 children, we constructed a decision rule using 3 out-of-hospital variables (Glasgow Coma Scale score, passenger space intrusionor =6 in [or =15 cm], and restraint use) to predict those patients with an ISS of 16 or more. Internal cross-validation was used to determine the sensitivity and specificity, yielding values of 92% and 73%, respectively, for the prediction of patients with an ISS of 16 or more.Out-of-hospital variables available to field personnel could be used to effectively triage pediatric motor vehicle crash patients using the decision rule developed here. Prospective trials would be needed to test this decision rule in actual use.

When assessing the clinical utility of therapies intended to improve subjective outcomes, the amount of improvement that is important to patients must be determined.1 The smallest benefit of value to patients is called the minimal clinically important difference (MCID). The MCID is a patient-centered concept, capturing both the magnitude of the improvement and also the value patients place on the change. Using patientcentered MCIDs is important for studies involving patientreported outcomes,2 for which the clinical importance of a given change may not be obvious to clinicians selecting treatments. The MCID defines the smallest amount an outcomemust change to be meaningful to patients.1 In this issue of JAMA, Hinmanet al3 report findings froma clinical trial evaluatingwhether acupuncture (needle, laser, andsham laser) improved pain or overall functional outcomes compared with no acupuncture among patients with chronic knee pain. Pain was measured on a numerical rating scale and functional status by the Western andMcMasterUniversitiesOsteoarthritis Index (WOMAC) score. The MCIDs for both end points were based on prior experience with these scoring systems. The MCID for pain was determined using an expert consensus, or Delphi approach,4 while the MCID for function was determined using an “anchor” approach, based on patients’ qualitative assessments of their own responses to treatment.5

The intention-to-treat (ITT) principle is a cornerstone in the interpretation of randomized clinical trials (RCTs) conducted with the goal of influencing the selection of medical therapy for well-defined groups of patients. The ITT principle defines both the study population included in the primary efficacy analysis and how the outcomes are analyzed. Under ITT, study participants are analyzed as members of the treatment group to which they were randomized regardless of their adherence to, or whether they received, the intended treatment.1- 3 For example, in a trial in which patients are randomized to receive either treatment A or treatment B, a patient may be randomized to receive treatment A but erroneously receive treatment B, or never receive any treatment, or not adhere to treatment A. In all of these situations, the patient would be included in group A when comparing treatment outcomes using an ITT analysis. Eliminating study participants who were randomized but not treated or moving participants between treatment groups according to the treatment they received would violate the ITT principle.

Objective: We compared the survival of trauma patients in urban versus rural settings after the implementation of a novel rural non-trauma center alternative care model called the Model Rural Trauma Project (MRTP). Materials and Methods: We conducted an observational cohort study of all trauma patients brought to eight rural northern California hospitals and two southern California urban trauma centers over a one-year period (1995-1996). Trauma patients with an injury severity score (ISS) of >10 were included in the study. We used logistic regression to assess disparities in odds of survival while controlling for Trauma and Injury Severity Score (TRISS) parameters. Results: A total of 1,122 trauma patients met criteria for this study, with 336 (30%) from the rural setting. The urban population was more seriously injured with a higher median ISS (17 urban and 14 rural) and a lower Glasgow Coma Scale (GCS) (GCS 14 urban and 15 rural). Patients in urban trauma centers were more likely to suffer penetrating trauma (25% urban versus 9% rural). After correcting for differences in patient population, the mortality associated with being treated in a rural hospital (OR 0.73; 95% CI 0.39 to 1.39) was not significantly different than an urban trauma center. Conclusion: This study demonstrates that rural and urban trauma patients are inherently different. The rural system utilized in this study, with low volume and high blunt trauma rates, can effectively care for its population of trauma patients with an enhanced, committed trauma system, which allows for expeditious movement of patients toward definitive care.

Our goal was to compare initial operative and nonoperative management for periappendiceal abscess complicating appendicitis.This study is a retrospective review of 155 consecutive patients with appendicitis complicated by periappendiceal abscess treated between 1992 and 1998. Eighty-eight patients were treated initially nonoperatively, and 67 patients were treated operatively. All patients had localized abdominal tenderness and either computed tomography or intraoperative documentation of an abscess.Our patient population consisted of 107 males and 48 females, with an average age of 33 (range, 16-75) years. Age, gender, comorbidity, white blood cell count, temperature, and heart rate did not differ significantly between groups. For the initial nonoperative management group, the failure rate was 5.8 percent and the appendicitis recurrence rate was 8 percent after a mean follow-up of 36 weeks. The response to treatment of the initial nonoperative group and the initial operative group was compared by length of stay (9 +/- 5 days vs. 9 +/- 3 days; P = not significant), days until white blood cell count normalized (3.8 +/- 4 days vs. 3.1 +/- 3 days; P = not significant), days until temperature normalized (3.2 +/- 3 days vs. 3.1 +/- 2 days; P = not significant), and days until a regular diet was tolerated (4.7 +/- 4 days vs. 4.6 +/- 3 days; P = not significant). Complication rate was significantly lower in the nonoperative group (17 vs. 36 percent; P = 0.008).Initial nonoperative management of appendicitis complicated by periappendiceal abscess is safe and effective. Patients undergoing initial nonoperative management have a lower rate of complications, but they are at risk for recurrent appendicitis.

Objective.In children Methods.Evidence: References addressing pediatric HT were generated from a computerized database (Medline). The articles were reviewed and evidence tables were compiled.Expert Panel: The multidisciplinary panel was comprised of nine experts in pediatric HT.Consensus Process: A modified Delphi technique was used to develop the guidelines. Before the one meeting, panel members reviewed the evidence and formulated answers to specific clinical questions regarding HT in young children. At the meeting, guidelines were formulated based on data and expert consensus.Results.A management strategy was developed that categorizes children into 4 subgroups, based on risk of ICI. Children in the high-risk group should undergo a computed tomography (CT) scan. Those in the intermediate risk group with symptoms of possible ICI should either undergo CT scan or observation. Those in the intermediate risk group with some risk for skull fracture or ICI should undergo CT and/or skull radiographs or observation. Those in the low-risk group require no radiographic imaging.Conclusions.We have developed a guideline for the evaluation of children

Study Objective: Prior laboratory and clinical studies demonstrate that cardiopulmonary resuscitation (CPR) preceding countershock of prolonged ventricular fibrillation (VF) increases the likelihood of successful cardiac resuscitation. The lower limit of VF duration at which time preshock CPR provides no benefit has not been specifically studied. The purpose of this study was to compare countershock and cardiac resuscitation outcome between immediate countershock of VF of 5-minute duration and CPR without drug therapy before countershock in a swine model. Methods: VF was induced in anesthetized and instrumented swine. After 5 minutes of VF, animals received 1 of 2 treatments. Animals in group 1, a "historical" control group (n=20), received immediate countershock followed by CPR and repeated shocks if needed. Group 2 animals (n=11) received CPR for 90 seconds preceding countershock, then continued CPR and repeated countershock if necessary. Drugs were not administered to either group, and resuscitation efforts were discontinued if a perfusing rhythm was not restored within 10 minutes of the first countershock. First shock success rate (defined as termination of VF), the number of shocks required to terminate VF, and the cardiac resuscitation rate were compared between groups. Results: The first shock terminated VF in 13 of 20 group 1 animals and 2 of 11 group 2 animals ( P =.023). All but 1 animal in group 1 developed pulseless electrical activity after countershock. All but 1 animal in group 1 were eventually successfully resuscitated with CPR and repeated shocks if necessary. Four group 2 animals could not be resuscitated ( P =.042). Conclusion: Although effective in improving outcome of prolonged VF, CPR preceding countershock of VF of 5-minute duration does not improve the response to the first shock, decrease the incidence of postshock pulseless electrical activity, or the rate of return of circulation. In this study, CPR preceding countershock resulted in a significantly lower cardiac resuscitation rate. [Niemann JT, Cruz B, Garner D, Lewis RJ. Immediate countershock versus cardiopulmonary resuscitation before countershock in a 5-minute swine model of ventricular fibrillation arrest. Ann Emerg Med. December 2000;36:543-546.]

The optimal initial treatment for selected patients with perforated appendicitis may be nonoperative. For this reason it is important to be able to diagnose perforated appendicitis preoperatively. The purpose of this study was to determine the accuracy of diagnosing perforated appendicitis using only admission factors. The study population was comprised of 366 adult patients who underwent appendectomy for presumed appendicitis during 1997. Admission factors associated with perforated appendicitis were determined using univariate and multivariate analyses. These variables were then used to formulate a rule for the diagnosis of perforated appendicitis. Sensitivity and specificity were calculated for this rule. The admission factors analyzed were sex, race, age, days of pain, temperature, heart rate, symptoms, physical examination findings, and laboratory findings. Multivariate regression analysis revealed days of pain, temperature, and localized tenderness outside the right lower quadrant to be significant (P0.05). Using two or more days of pain, a temperature of/=101 F (38.3 C), or localized tenderness outside the right lower quadrant as criteria to indicate perforation, we achieved a sensitivity of 86% and a specificity of 58% for distinguishing perforated from nonperforated appendicitis. We concluded that (1) perforated appendicitis cannot reliably be distinguished from nonperforated appendicitis based on admission factors, and (2) two or more days of pain, localized tenderness outside the right lower quadrant, and a temperature of/=101 F (38.3 C) define a group of patients with appendicitis who have a high incidence of perforation.

This article describes the design and implementation of the Pediatric Airway Management Project. The project was completed January 1, 1997, and evaluated the effectiveness of endotracheal intubation relative to bag-valve-mask ventilation in improving survival to hospital discharge and neurologic outcome in children, the effect of training on paramedic airway management skills and self-efficacy, the length of time the skills can be retained, and the costs of training and retraining. The main focus of project design was the implementation of a controlled trial comparing methods of airway management for acutely ill and injured pediatric patients in the out-of-hospital setting. To date, this project is the largest prospective, controlled, out-of-hospital study of the care of children ever reported. Barriers to implementation of a study of this size are described.

Background The association of cardiac and liver disorders has not been extensively outlined in the literature. Methods A survey of the MEDLINE database was performed to assess the current status of research regarding the association between cardiac and liver disorders. Results Combined cardiac and hepatic disorders occur in 3 different settings: heart diseases affecting the liver, liver diseases affecting the heart, and cardiac and hepatic disorders with joint etiology. The spectrum of heart diseases affecting the liver includes mild alterations of liver function tests in heart failure, cardiogenic ischemic hepatitis, congestive liver fibrosis, and cardiac cirrhosis. The liver diseases affecting the heart include complications of cirrhosis such as hepatopulmonary syndrome, portopulmonary hypertension, pericardial effusion, and cirrhotic cardiomyopathy as well as noncirrhotic cardiac disorders such as high-output failure caused by intrahepatic arteriovenous fistulae. Cardiac and hepatic disorders with joint etiology include infectious, metabolic, immune, vasculitic, and toxic disorders. We propose a practical approach to a diagnostic workup of combined cardiac and hepatic disorders based on recognizing the sequence of appearance of the cardiac and liver disease, presence of features of a multisystem disease, and presence of pathognomonic features. The evaluation of combined cardiac and hepatic disorders takes into consideration the expected benefit of treatment and the risks related to invasive procedures. Accordingly, investigations can be limited to ancillary tests for patients with congested liver and mild alterations of liver function tests, in cardiogenic ischemic hepatitis, patients with cardiac cirrhosis who are proposed for conservative treatment, and multisystem disease involving the heart and the liver. Conversely, comprehensive investigations are recommended when invasive therapeutic interventions are considered for the treatment of hepatopulmonary syndrome, portopulmonary hypertension, or arteriovenous fistulae. Conclusion: Classification of a patient to any of the 3 categories—heart diseases affecting the liver, liver diseases affecting the heart, and cardiac and hepatic disorders with joint etiology—permits the physician to narrow the span of the possible diagnoses and allows for a more simple workup. (Am Heart J 2000;140:111-20.)

Study objectives: To investigate the ability of an emergency department screening protocol to initiate respiratory isolation of patients with pulmonary tuberculosis at ED triage before chest radiography. Methods: We conducted a prospective cohort study with retrospective medical record review of adult patients who presented for care to an urban, university-affiliated hospital in Los Angeles County over a 4-month period. Ambulatory patients were administered a triage screening protocol that used patient-reported tuberculosis risk factors and symptoms in combination with selective chest radiography to screen patients at ED triage for active pulmonary tuberculosis. Results: A total of 10,674 patients were screened; 2,218 were isolated at triage and underwent chest radiography, and 378 were kept in isolation in the ED. The respiratory isolation of pulmonary tuberculosis (RIPT) protocol detected 17 of 27 visits made by patients with unsuspected pulmonary tuberculosis, yielding a sensitivity of 63% (95% confidence interval [CI] 42% to 81%). The estimated specificity was 78%. For each patient with tuberculosis who was detected by the RIPT protocol, 624 patients were screened at triage, 130 chest radiographs were taken, and 22 patients were placed in respiratory isolation in the ED. Patients with undetected pulmonary tuberculosis more commonly had nonpulmonary chief complaints (76% versus 20%; odds ratio [OR] 13, 95% CI 2.1 to 78.3), and only 60% (95% CI 26% to 88%) were ultimately isolated in the hospital. Among RIPT screen-positive patients, radiographic findings predictive of pulmonary tuberculosis were cavitary lesions (OR 84.3, 95% CI 22.6 to 315), upper lobe infiltrates (OR 24.2, 95% CI 9.1 to 64.4), pleural effusions (OR 8.9, 95% CI 2.5 to 31.8), diffuse/interstitial infiltrates (OR 5.7, 95% CI 1.8 to 17.9), and non–upper lobe infiltrates (OR 3.1, 95% CI 1.0 to 9.5). Conclusion: The RIPT screening protocol was only moderately sensitive for isolating patients with pulmonary tuberculosis at ED triage. Future studies should evaluate modified and abridged screening protocols, as well as the cost-effectiveness of triage screening. [Sokolove PE, Lee BS, Krawczyk JA, Banos PT, Gregson AL, Boyce DM, Lewis RJ. Implementation of an emergency department triage procedure for the detection and isolation of patients with active pulmonary tuberculosis. Ann Emerg Med. April 2000;35:327-336.]

ContextEndotracheal intubation (ETI) is widely used for airway management of children in the out-of-hospital setting, despite a lack of controlled trials demonstrating a positive effect on survival or neurological outcome.ObjectiveTo compare the survival and neurological outcomes of pediatric patients treated with bag-valve-mask ventilation (BVM) with those of patients treated with BVM followed by ETI.DesignControlled clinical trial, in which patients were assigned to interventions by calendar day from March 15, 1994, through January 1, 1997.SettingTwo large, urban, rapid-transport emergency medical services (EMS) systems.ParticipantsA total of 830 consecutive patients aged 12 years or younger or estimated to weigh less than 40 kg who required airway management; 820 were available for follow-up.InterventionsPatients were assigned to receive either BVM (odd days; n = 410) or BVM followed by ETI (even days; n = 420).Main Outcome MeasuresSurvival to hospital discharge and neurological status at discharge from an acute care hospital compared by treatment group.ResultsThere was no significant difference in survival between the BVM group (123/404 [30%]) and the ETI group (110/416 [26%]) (odds ratio [OR], 0.82; 95% confidence interval [CI], 0.61-1.11) or in the rate of achieving a good neurological outcome (BVM, 92/404 [23%] vs ETI, 85/416 [20%]) (OR, 0.87; 95% CI, 0.62-1.22).ConclusionThese results indicate that the addition of out-of-hospital ETI to a paramedic scope of practice that already includes BVM did not improve survival or neurological outcome of pediatric patients treated in an urban EMS system.

In 1996, the Food and Drug Administration released its Final Rule for Waiver of Informed Consent in Certain Emergency Research Circumstances (the Final Rule). The Department of Health and Human Services (DHHS) also released an update of its regulations related to waiver of informed consent in emergency research. These new regulations allow resuscitation research to proceed with a waiver of informed consent under very narrow and specific clinical research circumstances. Waiving informed consent for research participation has profound ethical and scientific implications. However, in unpredictable life-threatening clinical situations for which current therapy is unproven or unsatisfactory, patients usually are unable to consent on their own behalf to participate in clinical trials of potentially beneficial but experimental interventions. Because of the time-dependent nature of most resuscitation interventions, it is usually not feasible to identify and contact the legally authorized representative who can speak on behalf of the patient within the presumed therapeutic window of the intervention under investigation. For such clinical trials to proceed, a waiver of informed consent is usually necessary. Patients who are critically ill or injured and unable to provide meaningful prospective informed consent because of their current life-threatening condition are vulnerable and require additional protections beyond those for research subjects who can speak on their own behalf. The Final Rule and the DHHS-updated regulations incorporate a number of additional patient safeguards that must occur if a clinical trial is to proceed with waiver of informed consent. Specific means of adequately meeting these requirements are not described in the regulations. Although this was intentional on the part of the federal regulators so that individual protocols and research environments would direct the development of these patient safeguards, the lack of specific guidance has led to confusion on the appropriate implementation of the new regulations. This article reviews some of the key concepts of the Final Rule, with suggestions on their purpose and meaning. It also reviews the studies that have been approved to date to proceed with waiver of informed consent, and offers suggestions for the process of implementing the requirements of the Final Rule for research involving patients who are unable to give prospective informed consent.

OBJECTIVE To evaluate the concordance with criteria developed by the study investigators and supply costs associated with placement of i.v. lines and saline locks by paramedics in the out-of-hospital setting. METHODS This was a retrospective consecutive case series at an urban base hospital. Patients were treated by paramedics using one base hospital for medical control during December 1995. Base hospital written records and taped patient calls were reviewed to determine actual i.v. access method used by paramedics, chief complaint, and whether fluid administration was ordered. Indicated method of i.v. access was determined for each patient based on predetermined criteria developed by the investigators. i.v. access methods were ranked by cost of supplies as follows: i.v. line (i.v.) > saline lock (SL) > no i.v. line (No i.v.). An assignment of concordant treatment was made when actual = indicated method, discordant-overtreatment when actual > indicated, and discordant-undertreatment when actual < indicated. RESULTS 452 patients were treated via radio by the base hospital during the study period. 380 of 452 (84%) received an i.v.. 28 of 380 (7%) received fluid resuscitation in the field. 166 of 452 (37%) received concordant treatment; 253 (56%) discordant-overtreatment; and 33 (7%) discordant-undertreatment. Pediatric patients (< or =14 years of age) were more likely to be undertreated as compared with adults, 33% vs 3% (p < 0.001). Patients who had medical chief complaints were more likely to receive discordant-overtreatment as compared with patients who had trauma chief complaints, 61% vs 32% (p < 0.001). 73% of chest pain patients received discordant-overtreatment. Based on these data, the yearly cost of supplies used in i.v. access discordant-overtreatment was $13,735 for this base hospital and $560,000 for the Los Angeles County emergency medical services (EMS) system. 91% of the excess supply cost is due to patients' receiving an i.v. instead of a SL. CONCLUSION Based on study criteria for utilization of i.v. lines vs SLs in the field, paramedics and base hospital personnel often provide discordant-overtreatment of patients by placement of an i.v. when a SL or No i.v. would suffice, resulting in unnecessary costs for EMS systems.

Objective: The aim of this study was to define the incidence of cholangitis in gallstone pancreatitis, in the absence of cholangitis, to identify the clinical predictors of persistent common bile duct (CBD) stones at endoscopic retrograde cholangiography (ERCP) or at intraoperative cholangiography (IOC). Methods: A total of 122 consecutive patients with acute gallstone pancreatitis were prospectively evaluated for the presence of CBD stones as determined by elective ERCP or IOC. Urgent ERCP was restricted to patients with concomitant cholangitis. APACHE II scores and serial laboratory data were obtained. Results: Four (3%) patients had cholangitis and all underwent urgent ERCP successfully. Eighteen (15%) patients without cholangiogram were excluded. The remaining 100 patients underwent elective ERCP or IOC on a mean of hospital day 6.8. Twenty-one (21%) patients had persistent CBD stones. Univariate analysis detected significant differences in serum total bilirubin, ALT, and alkaline phosphatase levels between patients with and without persistent CBD stones. On multivariate analysis, serum total bilirubin on hospital day 2 was the best predictor of CBD stones. A serum total bilirubin level > 1.35 mg/dl had a sensitivity of 90.5% and specificity of 63%. Age, gender, mean APACHE II score, amylase, and presence of CBD dilation on ultrasound were not predictive of CBD stones. Conclusion: In patients with gallstone pancreatitis, 1) cholangitis is uncommon, and 2) the best clinical predictor of persistent CBD stones is serum total bilirubin on hospital day 2.

This article reviews the federal regulations for emergency and acute resuscitation research in effect prior to October 1996, the historical issues that contributed to the development of these regulations, the controversies that arose surrounding the application of these regulations to emergency research circumstances, and the methods by which the regulations were changed. The new regulations introduced by the U.S. Food and Drug Administration (FDA) also are reviewed.

Hypothesis testing and the P value it generates are overemphasized in statistical analyses published in medical journals. An alternative, the confidence interval (CI), offers significantly more information to readers interpreting results. There have been many authoritative calls for the report of CIs in place of P values, 1-7 such as that of the International Committee of Medical Journal Editors, whose guidelines for statistical reporting give the following instructions: "When possible, quantify findings and present them with appropriate indicators of measurement error or uncertainty (such as confidence intervals)," and "Avoid sole reliance on statistical hypothesis testing, such as the use of P values, which fails to convey important quantitative information." 8 In this article, part 1, we provide an overview of CIs for the clinician reading the medical literature. We describe the advantages of CIs and explain and illustrate their proper interpretation. In part 2, which follows this article, we provide added information important for clinical researchers, including a precise definition of CIs, a compact reference to methods for calculating CIs in common situations, and an explanation of the difference between CIs and probability intervals. 9 [Young KD, Lewis RJ: What is confidence? Part 1: The use and interpretation of confidence intervals. Ann Emerg Med September 1997;30:307-310.]

In many cases, CIs should be reported in place of P values. The lower limit of a 95% CI is chosen such that the probability of obtaining the value observed in the trial or a greater value is .025 and the upper limit such that the probability of obtaining the value observed or a lesser value is also .025. CIs for data from most clinical trials may be calculated easily with the use of formulas given in this article. A 95% CI is not the same as a 95% probability interval. The 95% CI contains values statistically consistent with the data obtained, whereas a 95% probability interval has a 95% probability of containing the true value of interest.

Purpose: The purpose of this study was to determine factors associated with longer times to transport of emergency pediatric patients requiring tertiary care. Design: Retrospective case series. Setting: Emergency pediatric transport service. Participants: Infants and children transported by the transport service at the University of North Carolina Hospitals at Chapel Hill from January 1, 1988, to December 31, 1990. Main measurements: The time-to-request, the time from patient arrival at the referring hospital to the time when the request for transfer was received, and the ground time, defined as the time between the transport team's arrival at the referring hospital and their departure, were recorded for each transported patient. Results: Three hundred consecutive children 0 to 16 years (61% male) were transferred. Time-to-request was shorter for trauma patients (median 62 minutes, quartiles 29 and 153 minutes) than for medical patients (median 172 minutes, quartiles 83 and 508 minutes) (P = 0.0001). Infants, children, and adolescents had similar times-to-request of 147 minutes, 129 minutes, and 128 minutes, respectively (P = 0.91). Increased ground times were associated with diagnosis category (median of 40 minutes for medical patients vs 29 minutes for trauma patients) (P = 0.0001), with younger age (median of 46 minutes for infants, 35 minutes for children, and 28 minutes for adolescents) (P = 0.0001), and with the performance of major procedures (median of 35 minutes if no procedures were performed, 38 minutes if one procedure was performed, and 54 minutes if two procedures were performed) (P = 0.039). After the transport team arrived, 13% (40/300) of patients required at least one major procedure prior to transport. Conclusions: Increased time-to-request for patients with medical diagnoses, increased ground times for younger patients and patients with medical diagnoses, and failure to perform necessary procedures contribute to a prolongation of the time-to-transport of emergency pediatric patients. The magnitude of the impact of these longer transport times on outcome is unknown.

Study objective: To define the rate of complications of adult varicella, including pneumonia, and to determine the association of such complications with pregnancy. Methods: Retrospective, descriptive case series of consecutive adult emergency department patients with primary varicella seen over a 31-month period in an urban county hospital. Results: A total of 130 patients were identified, with a mean age of 23.9±6.1 years; 69 (53%) were female. Thirteen patients (10.0%) had pneumonia; 6 of these were hospitalized. Two patients admitted with pneumonia had significant underlying comorbid disease and expired from respiratory failure. The remaining 4 admitted patients recovered uneventfully. Four of the 7 patients discharged with pneumonia were known to have recovered uneventfully. Twenty-eight (41%) of the 69 female study patients were pregnant; 7 of these (25%) were admitted, with admission diagnoses of pneumonia (1), active labor (2), and pregnancy with varicella (4). Incidence of pneumonia in pregnant patients was 3.6% (1 of 28; 95% confidence interval, .1% to 18.3%). All 7 admitted pregnant patients and the 19 (of 21) discharged pregnant patients who were contacted recovered uneventfully. No congenital anomalies or perinatal complications were noted in the infants of the 26 mothers with documented follow-up. Conclusion: Hospital admission should be considered for adults with varicella pneumonia, especially if they have significant comorbid disease. It does not appear that pregnant women are at increased risk of developing varicella pneumonia. [Baren JM, Henneman PL, Lewis RJ: Primary varicella in adults: Pneumonia, pregnancy, and hospital admission. Ann Emerg Med August 1996;28:165-169.]

Objective: To determine 1) the percentage of patients with new–onset atrial fibrillation for whom admission is medically justified and 2) whether those patients for whom hospitalization is medically justified can be reliably identified in the ED. Methods: A retrospective, descriptive cohort analysis was performed using consecutive adult patients with new–onset atrial fibrillation seen in an urban, county hospital ED from 1987 through 1992. Admissions were categorized as medically justified if patients were hypotensive (systolic blood pressure

Study objective: To analyze a standardized medical evaluation of alert, adult emergency department patients with new psychiatric symptoms. Design: Prospective, descriptive case series. Setting: Urban, county hospital. Participants: One hundred consecutive, alert patients, aged 16 to 65 years, seen over 9 months, with new psychiatric symptoms. Excluded patients included those with obvious alcohol or drug intoxication, psychiatric patients with previously diagnosed abnormal behavior, psychiatric patients with medical complaints, and overdose or suicide patients. Interventions: The following evaluation was performed: medical history, physical examination, complete blood count, SMA-7 (electrolytes, blood urea nitrogen, creatinine, and glucose), prothrombin time, calcium, oxygen saturation, creatine phosphokinase (CPK) if there was possible myoglobulinuria, alcohol level, urine drug screen (for cocaine, amphetamine, and phencyclidine), cranial computed tomography, lumbar puncture if febrile, and psychiatric evaluation when appropriate. Results were considered significant when they diagnosed the cause of the symptoms or resulted in medical admission. Results: Sixty-three of the 100 patients had an organic etiology of their symptoms. The medical history was significant in 27, physical examination in 6, CBC in 5, SMA-7 in 10, CPK in 6, alcohol and drug screen in 28, computed tomography scan in 8, and lumbar puncture in 3. Conclusion: Most alert, adult patients with new psychiatric symptoms have an organic etiology. We recommend performing a medical history, physical examination, SMA-7, calcium, CPK if there is possible myoglobinuria, alcohol and drug screens, computed tomography scan, and lumbar puncture as part of the medical clearance of these patients. [Henneman PL, Mendoza R, Lewis RJ: Prospective evaluation of emergency department medical clearance. Ann Emerg Med October 1994;24:672-677.]

Using exit interviews, we determined parental recall of their child's diagnosis, treatment, and follow-up instructions after a visit to the emergency department (ED). Over 2 weeks, 159 parents were interviewed. Exclusion criteria were: parental language other than Spanish or English, admission, trauma, child abuse, or a primary psychiatric diagnosis. Neither language nor parental satisfaction with communication (reported in 93%) was associated with ability to state the diagnosis correctly ( P = NS). Seventy-five percent (88 of 117) of parents of children given a single diagnosis stated it correctly; 55% (23 of 42) of parents of children given multiple diagnoses were able to correctly state them all ( P = 0.013). If a single medication was prescribed, 30% (20 of 67) of parents knew its name, while only 13% (four of 31) knew all names of multiple medications (P= 0.070). Of those parents given a single medication, 51% (34 of 67) knew how to administer it, while only 10% (three of 31) knew how to administer multiple medications ( P