Patrick M. Lank

Opioid Prescribing in a Cross Section of US Emergency Departments

STUDY OBJECTIVE Opioid pain reliever prescribing at emergency department (ED) discharge has increased in the past decade but specific prescription details are lacking. Previous ED opioid pain reliever prescribing estimates relied on national survey extrapolation or prescription databases. The main goal of this study is to use a research consortium to analyze the characteristics of patients and opioid prescriptions, using a national sample of ED patients. We also aim to examine the indications for opioid pain reliever prescribing, characteristics of opioids prescribed both in the ED and at discharge, and characteristics of patients who received opioid pain relievers compared with those who did not. METHODS This observational, multicenter, retrospective, cohort study assessed opioid pain reliever prescribing to consecutive patients presenting to the consortium EDs during 1 week in October 2012. The consortium study sites consisted of 19 EDs representing 1.4 million annual visits, varied geographically, and were predominantly academic centers. Medical records of all patients aged 18 to 90 years and discharged with an opioid pain reliever (excluding tramadol) were individually abstracted by standardized chart review by investigators for detailed analysis. Descriptive statistics were generated. RESULTS During the study week, 27,516 patient visits were evaluated in the consortium EDs; 19,321 patients (70.2%) were discharged and 3,284 (11.9% of all patients and 17.0% of discharged patients) received an opioid pain reliever prescription. For patients prescribed an opioid pain reliever, mean age was 41 years (SD 14 years) and 1,694 (51.6%) were women. Mean initial pain score was 7.7 (SD 2.4). The most common diagnoses associated with opioid pain reliever prescribing were back pain (10.2%), abdominal pain (10.1%), and extremity fracture (7.1%) or sprain (6.5%). The most common opioid pain relievers prescribed were oxycodone (52.3%), hydrocodone (40.9%), and codeine (4.8%). Greater than 99% of pain relievers were immediate release and 90.0% were combination preparations, and the mean and median number of pills was 16.6 (SD 7.6) and 15 (interquartile range 12 to 20), respectively. CONCLUSION In a study of ED patients treated during a single week across the country, 17% of discharged patients were prescribed opioid pain relievers. The majority of the prescriptions had small pill counts and almost exclusively immediate-release formulations.

Emergency Physicians’ Knowledge of Cannabinoid Designer Drugs

Introduction: The use of synthetic drugs of abuse in the United States has grown in the last few years, with little information available on how much physicians know about these drugs and how they are treating patients using them. The objective of this study was to assess emergency physician (EP) knowledge of synthetic cannabinoids (SC). Methods: A self-administered internet-based survey of resident and attending EPs at a large urban emergency department (ED) was administered to assess familiarity with the terms Spice or K2 and basic knowledge of SC, and to describe some practice patterns when managing SC intoxication in the ED. Results: Of the 83 physicians invited to participate, 73 (88%) completed surveys. The terms “Spice” and “K2” for SC were known to 25/73 (34%) and 36/73 (49%) of respondents. Knowledge of SC came most commonly (72%) from non-medical sources, with lay publications and the internet providing most respondents with information. Among those with previous knowledge of synthetic cannabinoids, 25% were not aware that SC are synthetic drugs, and 17% did not know they are chemically most similar to marijuana. Among all participants, 80% felt unprepared caring for a patient in the ED who had used synthetic cannabinoids. Conclusion: Clinically active EPs are unfamiliar with synthetic cannabinoids. Even those who stated they had heard of synthetic cannabinoids answered poorly on basic knowledge questions. More education is needed among EPs of all ages and levels of training on synthetic cannabinoids.

Emergency department alcohol and drug screening for Illinois pediatric trauma patients, 1999 to 2009

BACKGROUND Recent guidelines recommend universal substance abuse screening for all trauma patients aged 12 years and older because brief interventions can help prevent future trauma. However, little is known about actual rates of screening in this setting. METHODS The Illinois State Trauma Registry was queried for severely injured patients from 1999 to 2009. Multivariate logistic regression was used to characterize, according to demographic and physiologic parameters, which patients were screened with blood alcohol and urine toxicology and which screened positive. RESULTS Of the 12,264 pediatric patients, 40% were tested for alcohol and 37% for drugs. Nine percent of patients screened positive for alcohol and 8% for drugs. Age strongly predicted positive tests, as did male sex. Black and Hispanic patients were screened for alcohol most frequently, but only Hispanics were more likely to test positive. CONCLUSION Although current guidelines recommend screening all trauma patients 12 years and older, current practice falls far short of this goal.

Outcomes for older trauma patients in the emergency department screening positive for alcohol, cocaine, or marijuana use

Abstract Background: Substance use among older adults is an increasing concern, with the prevalence of substance use in older populations expected to double in the next decade. Drug and alcohol use is associated with trauma risk and outcomes, but little is known about the specific risk for older trauma patients. Objectives: To evaluate the association between drug and alcohol use and trauma outcomes among adults aged 55 years and older. Methods: This retrospective observational study included older adults from the Illinois Trauma Registry between 1999 and 2009. Exclusion criteria were age younger than 55 years or absent date of birth, ethanol level, or urine drug screen (UDS). Alcohol intoxication was defined as ethanol level greater than 80 mg/dL. UDS was used to screen cocaine and marijuana use. Analyses, for both the alcohol and the marijuana/cocaine groups, compared outcomes for patients with negative vs. positive screens. Results: 21 320 patients were included in the alcohol analysis and 17 077 in the drug analysis. Compared to non-intoxicated patients, alcohol-intoxicated patients had significantly (p < 0.001) lower in-hospital mortality, decreased ICU admission, decreased intubation rate, and shorter hospital length of stay. Patients screening positive for cocaine or marijuana had significantly longer lengths of stay with increased ICU admission compared with those who screened negative. Conclusion: Among older trauma patients, this study shows significant associations with multiple trauma outcomes, including one between elevated ethanol concentrations and improved outcomes. Future research into the causes of these findings could inform the care of older trauma patients and aid in prevention of injuries.

Benzodiazepine-opioid co-prescribing in a national probability sample of ED encounters

Background: Benzodiazepine‐opioid combination therapy is potentially harmful due to the risk of synergistic respiratory depression, and the rate of death due to benzodiazepine‐opioid overdose is increasing. Little is known about the prevalence and characteristics of benzodiazepine‐opioid co‐prescribing from the ED setting. Methods: Secondary analysis of data from the National Hospital Ambulatory Medical Care Survey, using sample weights to generate population estimates. The primary objective was to describe the annual prevalence of benzodiazepine‐opioid co‐prescribing from 2006 to 2012, using 95% confidence intervals (95% CI) to compare adjacent years. The secondary objective was to compare characteristics of ED encounters receiving a benzodiazepine‐opioid co‐prescription versus those receiving an opioid prescription alone, using a multivariable logistic regression. Results: The prevalence of benzodiazepine‐opioid co‐prescribing did not significantly change from 2006 to 2012. During this period, 2.7% (95% CI: 2.5–2.8%) of ED encounters prescribed an opioid were also prescribed a benzodiazepine. Relative to encounters receiving an opioid prescription alone, encounters receiving a co‐prescription were more likely to represent a follow‐up rather than initial visit (Odds Ratio [OR] 1.52), receive more medications (OR 1.41) and fewer procedures (OR 0.48) while in the ED, and more likely to have a diagnosis related to mental disorder (OR 20.60) or musculoskeletal problem (OR 3.71). Conclusions: From 2006 to 2012, almost 3% of all ED encounters receiving an opioid prescription also received a benzodiazepine co‐prescription. The odds of benzodiazepine‐opioid co‐prescribing were significantly higher in ED encounters representing a follow‐up visit and in diagnoses relating to a mental disorder or musculoskeletal problem.

Emergency Medicine Faculty Are Poor at Predicting Burnout in Individual Trainees: An Exploratory Study

Burnout is common among emergency medicine (EM) physicians, and it is prevalent even among EM trainees. Recently proposed Accreditation Council for Graduate Medical Education requirements encourage faculty to alert residency leadership when trainees display signs of burnout. It remains uncertain how trainees experiencing burnout can be reliably identified. We examined if EM faculty advisers at one institution can accurately predict burnout in their EM resident advisees.

ED opioid prescribing is not associated with higher patient satisfaction scores

Despite the popular belief that providing an opioid prescription increases patient satisfaction [1], this association has not been well studied in the emergency department (ED) setting. Although a prior study found no association between patient satisfaction scores and opioid analgesics administered in the ED [2], providing an opioid prescription may reasonably differ. Prescription opioids may serve as physical tokens of the index ED visit at the time of survey response and may increase patient satisfaction similarly to the provision of an antibiotic prescription [3]. The potential association between opioid prescribing and patient satisfaction is worthy of investigation. Although not all opioid prescriptions result in misuse, the potential harm of indiscriminate prescribing practices is becoming increasingly evident, as unintentional opioid overdoses have now surpassed motor vehicle accidents as the leading cause of injury [4]. In a recent national survey of ED providers, 12% of respondents reported prescribing an opioid analgesic to improve patient satisfaction [5]. We conducted an institutional review board–approved retrospective cohort study of adult patients completing a Press Ganey survey relating to an ED visit in 2010 to a single urban academic ED in Chicago, IL (85 000 annual visits). Per institutional protocol, 30% of discharged patients are randomly selected to receive a survey bymail. ED providers at this institution do not receive financial incentives for patient satisfaction scores. We stratified patients into 1 of 3 exposure groups (no analgesic prescription, nonopioid analgesic prescription, or any opioid prescription) based on Multum coding of prescribed medications and investigated the association between prescription type and Press Ganey survey scores. The patient satisfaction survey is composed of 38 individual questions, each scored on a 5-point Likert scale (1 = very poor, 5= very good), and organized into 8 domains: Arrival, Nurses, Doctors, Tests, Family/Friends, Personal/Insurance Information, Personal

Risk assessment of methanol poisoning in outbreaks not applicable to isolated cases

We read with great interest the recent study by Paasma et. al. in the November issue of Clinical Toxicology , regarding prognostic indicators in methanol poisoning. 1 This valuable multicenter retrospective study demonstrated the importance of pH and coma as risk factors for mortality and discussed the emerging signifi cance of respiratory compensation of acidosis. While we applaud the authors ’ generation of a prognostic chart for their patient population (Paasma et al., Fig. 2), we are concerned that there are factors that make the chart non-generalizable. The study population consists largely of clusters of people poisoned during methanol outbreaks in various countries and times. We worry that people involved in methanol outbreaks are fundamentally different from those who ingest methanol in other settings – either knowingly as a means of suicide or as an ethanol substitute, or unintentionally. For example, previous publications on the Norwegian 2002 – 2004 cluster from this study found that these subjects tended to be older (median age 53), with a high prevalence of alcoholism, 2,3 possibly infl uencing their outcomes. It is also conceivable that since victims of a methanol outbreak are unaware of the methanol ingestion, they may not provide historical clues on initial presentation, increasing the possibility of delays to diagnosis and treatment. Alternatively, people involved in a methanol outbreak from tainted “ alcohol ” may have improved prognosis because of the presence of antidotal ethanol in the liquor. With the multiple possible confounding factors related to outbreaks, we are apprehensive about applying their results to non-outbreak methanol-poisoned patients. We would specifi cally like to caution against the use of Fig. 2 in making treatment decisions, particularly, as the authors note, “ when triaging patients. ” It is widely held that there is a role for hemodialysis in severe methanol poisonings, but its use has been conspicuously omitted from this study and the development of the risk-assessment chart. Interestingly, in a paper on the cohort from the 2001 Estonia outbreak, Paasma et al. note that patients received shorter duration of hemodialysis due to limited resources in the face of the high number of patients involved in the outbreak. 4 Given the widespread use of hemodialysis for severe methanol poisoning, as well as its potentially insuffi cient use in resource-limited settings, its absence from any paper on severe methanol poisoning prognosis is unfortunate. Clearly the authors have demonstrated that low pH and coma are indicators of poor prognosis. Although the risk assessment chart presented by the authors makes logical sense, we believe caution should be used when using it to determine mortality risk and should not be used in making triage or treatment decisions, especially in nonoutbreak poisonings.

Development of an Emergency Medicine Wellness Curriculum

Burnout, the triad of emotional exhaustion, depersonalization, and low personal accomplishment, begins early in medical education and the prevalence continues to increase over time among U.S. physicians. The Accreditation Council for Graduate Medical Education (ACGME) now requires that programs and sponsoring institutions have the same responsibility to address well‐being as they do other aspects of resident competence. Yet, there are no studies published in the emergency medicine (EM) literature that discuss the development and institution of a formal wellness curriculum. The authors conducted a needs analysis among EM residents with the aim of creating a multifaceted 12‐month wellness curriculum. The needs analysis determined that residents are not comfortable with their knowledge of wellness principles. In response, the authors developed a curriculum by integrating components of published non‐EM wellness curricula and online academic wellness programs with commonly accepted domains of wellness. The curriculum was subsequently introduced at five EM residencies. This curriculum represents an example of successful multi‐institution collaboration to meet an ACGME Common Program Requirement.

When good times go bad: managing ‘legal high’ complications in the emergency department

Patients can use numerous drugs that exist outside of existing regulatory statutes in order to get “legal highs.” Legal psychoactive substances represent a challenge to the emergency medicine physician due to the sheer number of available agents, their multiple toxidromes and presentations, their escaping traditional methods of analysis, and the reluctance of patients to divulge their use of these agents. This paper endeavors to cover a wide variety of “legal highs,” or uncontrolled psychoactive substances that may have abuse potential and may result in serious toxicity. These agents include not only some novel psychoactive substances aka “designer drugs,” but also a wide variety of over-the-counter medications, herbal supplements, and even a household culinary spice. The care of patients in the emergency department who have used “legal high” substances is challenging. Patients may misunderstand the substance they have been exposed to, there are rarely any readily available laboratory confirmatory tests for these substances, and the exact substances being abused may change on a near-daily basis. This review will attempt to group legal agents into expected toxidromes and discuss associated common clinical manifestations and management. A focus on aggressive symptom-based supportive care as well as management of end-organ dysfunction is the mainstay of treatment for these patients in the emergency department.