SEASON THREE – CAEP 2016 in Québec City
1. Ken Milne and Shannon Fernando: Contributing factors and time delays in management of difficult airways in the emergency department – a retrospective analysis (University of Ottawa).
2. Laurie Morrison and Nicole Beckett: Is there an association between the use of cardiac ultrasound and survival outcomes in patients arriving to the emergency department in cardiac arrest? The second Sonography in Hypotension and Cardiac Arrest in the Emergency Department (SHOC-ED 2) Study (Dalhousie Medicine New Brunswick).
3. Ken Milne and Eddy Lang: Does your patient really need intravenous therapy? A multicenter variation analysis of physician practice in low-acuity presentations (University of Calgary).
4. Brent Thoma and Sebastian Dewhirst: Do real-time Twitter metrics correlate with traditional Emergency Medicine post-conference speaker evaluations? (Univsersity of Ottawa).
5. Laurie Morrison and Carolyn Snider: Wraparound Care for Youth Injured By Violence: A Randomized Control Trial (University of Manitoba).
6. Ken Milne and Nadine Allain-Boulé (en français): Measuring frailty can help Emergency Departments identify seniors at risk of functional decline after minor injuries (Canadian Emergency Department Team Initiative, Québec, QC)
7. Laurie Morrison and Olivia Cook: Reasons for Referral and Hospitalization among Emergency Department Patients with Syncope (University of Ottawa)
8. Brent Thoma and Riley Hartmann: Using GRADE-based recommendations for analgesia and antiemetics in electronic order sets to influence physician behaviour towards best practice and cost-savings (Univsersity of Calgary).
9. Ken Milne and Grant Innes: Impact of physician payment mechanism on wait times and ED length of stay (University of Calgary).
10. Ken Milne and Nazanin Meshkat: Anticoagulation use in patients with atrial fibrillation/flutter in Canadian emergency departments since the introduction of the novel anticoagulants (Univsersity of Toronto)
11. Laurie Morrison and Allan Shefrin: What Ultrasonography Characteristics Predict Surgical Intervention for Children with Testicular Torsion? (Children’s Hospital of Eastern Ontario).
12. Brent Thoma and Rob Soegtrop: Exercise Prescription By Canadian Emergency Medicine Physicians (University of Western Ontario)
13. Brent Thoma and Miles Hunter: Systematic review of the management of lateral epicondylitis using transdermal nitroglycerin (University of Calgary).
Season Three: Reasons for Referral and Hospitalization among Emergency Department Patients with Syncope
MP023 Reasons for Referral and Hospitalization among Emergency Department Patients with Syncope
O. Cook, M.A. Mukarram, M. Rahman, S. Kim, K. Arcot, K. Thavorn, M. Taljaard, M. Sivilotti, B.H. Rowe, V. Thiruganasambandamoorthy (University of Ottawa)
Introduction: Syncope can be caused by serious life-threatening conditions not obvious during the initial ED assessment leading to wide variations in management. We aimed to identify the reasons for consultations and hospitalizations, outcomes, and the potential cost savings if an outpatient cardiac monitoring strategy were developed.
Methods: We conducted a prospective cohort study of adult syncope patients at 5 academic EDs over 41 months. We collected baseline characteristics, reasons for consultation and hospitalization, hospital length of stay and average total inpatient cost. Adjudicated 30-day serious adverse events (SAEs) included death, myocardial infarction, arrhythmia, structural heart disease, pulmonary embolism, significant hemorrhage and procedural intervention. We used descriptive statistics with 95% CI.
Results: Of the 4,064 patients enrolled (mean age 53.1 years, 55.9% female), 3,255 (80.1%) were discharged from the ED, 209 (5.2%) had a SAE identified in the ED, 600 (14.8%) with no SAE were referred for consultation in the ED and 299 (7.4%) were hospitalized: 55.5% of referrals and 55.2% of hospitalizations were for suspected cardiac syncope (46.5% admitted for cardiac monitoring of whom 71.2% had no cause identified). SAE among groups were 9.7% in total; 2.5% discharged by ED physician; 3.4% discharged by consultant from ED; 21.7% as inpatient and 4.8% following discharge from hospital. The mean hospital length of stay for cardiac syncope was 6.7 (95%CI 5.8, 7.7) days with total estimated costs of $7,925 per patient (95% CI: 7434, 8417).
Conclusion: Suspected cardiac syncope, particularly arrhythmia, was the major reason for ED referral and hospitalization. The majority of patients hospitalized for cardiac monitoring had no identified cause. An important number of patients suffered SAE, particularly arrhythmias outside the hospital. These findings highlight the need to develop a robust syncope prediction tool and a remote cardiac monitoring strategy to improve patient safety while saving substantial health care resources.
Season Three: Measuring frailty can help Emergency Departments identify seniors at risk of functional decline after minor injuries
MP016 Measuring frailty can help Emergency Departments identify seniors at risk of functional decline after minor injuries
N. Allain-Boulé, M. Sirois, L.E. Griffith, M. Émond and B. Batomen Kuimi (Canadian Emergency Department Team Initiative – Québec, QC)
Introduction: The CETI team has shown that around 18% of otherwise independent seniors remain in a state functional decline up to six months after a minor injury. In that context, frailty may be associated with increased likelihood of decline. As most seniors consult Emergency Departments (EDs) when injured, measuring frailty may help identify those at risk of functional decline. Objectives: This study aims to 1) describe frailty in the sub-group of independent community-dwelling seniors consulting Emergency Departments (ED) for minor injuries, 2) examine the association between frailty and functional decline three months post-injury, 3) ascertain the predictive accuracy of frailty measures and Emergency Physicians’ (EPs) for functional decline.
Methods: Prospective cohort in 2011-2013 among 1072 seniors aged ≥ 65, independent in basic daily activities, evaluated in Canadian EDs for minor injuries and discharged home. Frailty was assessed at EDs using the Canadian Study of Health and Aging-Clinical Frailty Scale (CSHA-CFS) and the Study of Osteoporotic Fracture index (SOF). Functional decline was defined as a loss ≥ 2/28 on the Older American Resources Services scale three months post-injury. Generalized mixed models were used to explore differences in functional decline across frailty levels. Areas Under the Receiver operating characteristic curve (AUC) were used to ascertain the predictive accuracy of frailty measures and EPs’ clinical judgement.
Results: The SOF and CSHA-CFS were available in 342 and 1058 participants, respectively. The SOF identified 55.6%, 32.7%, 11.7% patients as robust, prefrail and frail. These CSHA-CFS (n=1058) proportions were 51.9%, 38.3% and 9.9%. The 3-month incidence of functional decline was 12.1% (10.0%-14.6%). The AUCs of the CSHA-CFS and the EPs’ were similar (0.548 – 0.777), while the SOF was somewhat higher (0.704 – 0.859).
Conclusion: Measuring frailty in community-dwelling seniors with minor injuries in EDs may enhance current risk screening for functional decline. However, before implementation in usual care, feasibility issues such as inter-rater reliability and acceptability of frailty tools in the EDs have to be addressed.
MP010 Wraparound Care for Youth Injured By Violence: A Randomized Control Trial
C. Snider, W. Chernomas, K. Cook, D. Jiang, T. Klassen, S. Logsetty, J. Mahmood, E. Mordoch and T. Strome (University of Manitoba)
Introduction: Youth injured by violence is a major public health concern in Canada. It is the fourth leading cause of death in youth and the foremost reason youth visit an emergency department (ED). In Winnipeg, 20% of youth who visit an ED with an injury due to violence will have an ED visit for a subsequent violent injury within one year. Youth injured by violence are in a reflective and receptive state of mind, rendering the ED setting appropriate for intervention.
Methods: We completed a randomized control trial in November 2015 comparing wraparound care for youth age 14 – 24 who were injured by violence to standard ED care. Youth were excluded if their injury was due to child maltreatment, sexual assault or self-harm. An adapted pre-consent randomization methodology was used. The intervention was developed using a community based participatory research approach. Wraparound care was delivered by a support worker with lived experience with violence. Support workers were on call 24/7 in order to start the intervention in the ED and take advantage of the “teachable moment.” Care continued in the community for approximately one year.
Results: A total of 133 youth were randomized (68 intervention, 65 control) in one year. There was no difference in age, gender, or severity of injury between the two groups. Patients randomized to the intervention spent a median of 30 minutes less in the ED than those receiving standard care (p=0.22). Youth are safely housed, have enrolled in education opportunities, and are engaged in addictions care. Results of a chart review examining repeat visits to the ED for violent injury, substance use and mental health will be completed in Spring 2016 and will be presented.
Conclusion: There were no differences between standard care and intervention groups on baseline characteristics reflecting effective randomization. The introduction of an intervention at bedside in the ED did not have a negative impact on patient length of stay.
Season Three: Do real-time Twitter metrics correlate with traditional Emergency Medicine post-conference speaker evaluations?
MP020 Do real-time Twitter metrics correlate with traditional Emergency Medicine post-conference speaker evaluations?
S.H. Yiu, S. Dewhirst, C. Lee, A. Jalaili and J.R. Frank (University of Ottawa)
Introduction: Traditional post-conference speaker evaluations are inconsistently completed; meanwhile, real time social media tools such as Twitter are increasingly used in conferences. We sought to determine whether a correlation exists between traditional conference evaluation for a speaker and the number of real-time tweets it generated using data from a CAEP conference.
Methods: This study utilized a retrospective design. The hashtag #CAEP14 was prospectively registered with Symplur, an online Twitter management tool, so that all tweets related to CAEP conference 2014 were stored. A tweet was associated with a session if it mentioned the speaker name, or if the tweet content and timing closely matched that of the session in the schedule. A tweet classification system was developed to differentiate original tweets from retweets, and quotes from comments generating further discussion. Two authors assessed and coded the first 200 tweets together to ensure a uniform approach to coding, and then independently coded the remaining tweets. Discrepancies were resolved by consensus. One author reviewed post-conference speaker evaluation, and abstracted the value corresponding to the question “The speaker was an effective communicator”. We present descriptive statistics and correlation analyses.
Results: A total of 3,804 tweets were collected, with 2,218 (58.3%) associated with a session. Forty-eight (48%) (131 out of 274) of sessions receiving at least one tweet, with a mean of 11.7 tweets per session (95% CI of 0 to 57.5). In comparison, only 31% (85 out of 274) of sessions received a formal post conference speaker evaluation (p<0.005). For sessions that received at least one traditional post-conference evaluation, there was no significant correlation between the number of tweets and evaluation scores (R=0.087). This can be attributed to the fact that there was minimal variation between evaluation scores (median = 3.6 out of 5, IQR of 3.4 to 3.7).
Conclusion: There was no correlation between the number of real-time tweets and traditional post-conference speaker evaluation. However, many sessions which received no formal speaker evaluation generated tweets, and the number of tweets was highly variable between sessions. Thus, Twitter metrics might be useful for conference organizers to supplement formal speaker evaluations.
Season Three: Does your patient really need intravenous therapy? A multicenter variation analysis of physician practice in low-acuity presentations
MP025 Does your patient really need intravenous therapy? A multicenter variation analysis of physician practice in low-acuity presentations
N. Dil, D. Wang, K. Lonergan, G. Innes, A. McRae, S. Dowling, N. Zuzic and E. Lang (University of Calgary)
Introduction: The decision to treat with parenteral therapy may reflect a variable practice pattern among emergency physicians and represent an opportunity to standardize care. Our objective was to describe physician level practice variation for IV therapies in patients with low-acuity presentations and quantify the contribution of IV therapy to prolonging ED LOS.
Methods: Using administrative data merged with computerized physician order entry information we sampled 48 months of patient variables across four urban EDs (Jan 1, 2014 – Dec 22, 2015). Eligible patients: 1. presented with complaints of abdominal pain, nausea and vomiting or diarrhea or had a discharge diagnosis of cellulitis 2.were in a low acuity category (Canadian Triage and Acuity Scale – CTAS 3 or 4) 3.were triaged to non-stretcher zones of the ED and 4.were not admitted to hospital. The primary outcome was the physician-level variation in the decision to order IV therapies for this patient group; namely one or more of the following: IV fluids, opioid analgesia, anti-emetics and antibiotics. Secondary outcomes were a comparison of ED LOS, ED revisits at 7 days and ED revisits resulting in admission at 7 days for the IV and non-IV groups.
Results: Our analysis included 31 802 patient visits treated by 185 physicians. The average patient age was 37.8 years with 64.3% being female and the majority triaged as CTAS 3 (82.5%). On average 24% of these visits were treated with IV therapies; 90th percentile; 34%. For physicians seeing in excess of 100 cases, the variation in IV therapy use ranged from 1% to 47%. Patients receiving IV therapies demonstrated a 44% greater average LOS (6.2 hours vs 4.3 hours) and those receiving IV therapies had higher 7-day ED revisit rates (12.0% vs 8.8%) as well as 7-day ED revisits resulting in readmission (2.4% vs 1.0%). ‘mso-spacerun:yes’> Secondary outcomes were a comparison of ED LOS, ED revisits at 7 days and ED revisits resulting in admission at 7 days for the IV and non-IV groups.
Conclusion: This is the first study to examine physician preference for the use of IV therapies in a low-acuity population and has demonstrated in excess of a 47-fold variation between both extremes of use. Reducing practice variation in this area of ED care by standardizing indications for IV therapies could result in more rational resource utilization and improved throughput.
Season Three: Is there an association between the use of cardiac ultrasound and survival outcomes in patients arriving to the emergency department in cardiac arrest? The second Sonography in Hypotension and Cardiac Arrest in the Emergency Department (SHOC-ED 2) Study.
MP012 Is there an association between the use of cardiac ultrasound and survival outcomes in patients arriving to the emergency department in cardiac arrest? The second Sonography in Hypotension and Cardiac Arrest in the Emergency Department (SHOC-ED 2) Study.
N. Beckett, P.R. Atkinson, J. Fraser, J. French and D. Lewis (Dalhousie Medicine New Brunswick)
Introduction: The use of cardiac point of care ultrasound (PoCUS) to assess cardiac arrest patients is widespread, although not mandated by advanced cardiac life support (ACLS) guidelines. This study aims to examine if the use of ultrasound, along with the findings on ultrasound are associated with a difference in outcomes of cardiac arrest patients in the emergency department (ED).
Methods: A retrospective database and chart analysis was completed for patients arriving to a tertiary ED in asystole or PEA cardiac arrest, between 2010 and 2014. Patients were excluded if aged under 19, or with a previous DNR order. Patients were grouped based on whether PoCUS was used during ACLS (PoCUS group) and those without PoCUS (control group). Multiple data were abstracted from charts using a standardized form. Data was analyzed for the return of spontaneous circulation (ROSC), survival to hospital admission (SHA), and survival to hospital discharge (SHD), as well as initial cardiac activity findings on PoCUS.
Results: 230 patients met the study inclusion criteria, with 44 (19%) in the control group, and 186 (81%) in the PoCUS group. In the PoCUS group 20 (11%) had cardiac activity (Positive PoCUS) and 166 (89%) had no cardiac activity recorded. The control group had a higher rate of SHA than the PoCUS group (27%; 95% CI 15-43% vs. 10%: 6-15%, p = 0.0046), however there was no difference in frequency of ROSC (control: 37%; 24-55% vs. PoCUS 26%; 20-33%, p = 0.1373) or SHD (control: 7%, 95% CI 1-19%; PoCUS: 2%, 95% CI 0-5%, p = 0.0858). Positive PoCUS patients had a higher frequency of ROSC (75%; 50-91% vs. 20%; 15-27%, p < 0.001) and SHA (25%; 9-49% vs. 8%; 4-13%, p = 0.0294) than patients with no PoCUS cardiac activity, however there was no difference in the rate of SHD between the positive PoCUS patients (0%; 0-17%) and patients with no PoCUS cardiac activity (2%; 0-5%, p = 1.0000).
Conclusion: Our results suggest that there is no difference in survival between cardiac arrest patients receiving PoCUS and those who do not. Although finding positive cardiac activity on PoCUS is associated with greater ROSC and survival to hospital admission, it does not identify patients with a final outcome of survival to hospital discharge.
Season Three: Contributing factors and time delays in management of difficult airways in the emergency department – a retrospective analysis
SEASON TWO – CAEP 2015 in Edmonton
- Ken Milne speaks with Bjug Borgundvaag about why he wants us to rise to the challenge for MRSA (University of Toronto).
- Ken Milne hears from Kyla Caners about the timely availability and correlates of code status in oncology patients presenting to the ED (McMaster University).
- Meghan Garnett (University of Western Ontario) and Ken Milne remind us about the dangers of drunk driving.
- Michele Foster (University of Calgary) speaks to Ken Milne about her shifty research.
- Ken Milne and Natalie Le Sage (Université Laval) discuss mild traumatic brain injuries – en français!
SEASON ONE – CAEP 2014 in Ottawa
- Ken Milne hears from Sally Carver about a new app that helps to assess alcohol withdrawal (Mount Sinai Hospital)
- Ken Milne learns more about the McMAP from Teresa Chan (McMaster University)
- Ken Milne learns from Kelly van Diepen how to meet them, greet them, treat them, and street them (McMaster University)
- Ken Milne is in the zone with James Gould (EHS Halifax)
- Ken Milne gets the chills from Chris Fabian (University of Ottawa)
MP31 Predicting the risk of major cardiac adverse event for emergency department patients with suspected ACS using the modified HEART score
Introduction: Most validated clinical decision rules for risk stratification of patients presenting to the emergency department (ED) with suspected acute coronary syndrome (ACS) include subjective elements or require repeat laboratory investigations, limiting widespread adoption. The HEART score is potentially useful; but also relies on the subjective evaluation of clinical history. The objective of this study was to modify the HEART score using objective criteria and evaluate its performance.
Methods: This was a secondary analysis of prospectively collected data for adult patients presenting to an academic ED with suspected ACS over a four-year study period. Eligible patients included those with a troponin ordered in the ED according to a medical directive. Clinical history was quantified using points (1 point for each) assigned for specific elements; central chest pain, chest pain radiating to the neck, jaw, shoulder or arms; chest pain aggravated by exertion, chest pain alleviated by nitroglycerin, chest pain squeezing in nature, chest pain similar to previous angina, or chest pain associated with shortness of breath, nausea/vomiting, diaphoresis, or (pre)syncope. A modified HEART score was used to stratify patients into low, intermediate and high risk categories. The primary endpoints were major cardiac adverse cardiac events (MACE) of myocardial infarction, cardiac arrest, death or coronary re-vascularization at 30 days and 6 months.
Results: 1569 patients were included in this study. Mean (SD) age was 61.2 (16.4) years and 791 (50.4%) were male. Of the 501 (31.9%) patients identified as low risk; 7 (1.4%) had a MACE within 30 days and 3 (0.6%) additional patients went onto have MACE within 6 months. Of the 948 (60.4%) identified as intermediate risk; 115 (12.1%) had MACE within 30 days and an additional 27 (2.8%) patients had MACE within 6 months. The remaining 120 (7.6%) patients were identified as high risk; 56 (46.7%) had a major event within 30 days and 1 (3.2%) additional patient went onto have MACE within 6 months.
Conclusion: A modified HEART score using objective criteria is capable of stratifying undifferentiated ED patients into low, intermediate and high risk for
MACE at 30 days and 6 months. This tool compares favourably to others and may be useful in identifying patients who clearly require admission, as well those who can be safely discharged home.
Keywords: troponin, chest pain, risk score
Season Two: Quality assessment of resident charting in emergency departments
P001 Quality assessment of resident charting in emergency departments
A. Adatia, MD, D.H. Grushka, BSc, MSc, MD, M. Bhimani, MSc, MD; University of Western Ontario, London, ON
Introduction: There is a paucity of literature regarding quality improvement in emergency department (ED) charting, specifically among residents working in the ED. Completeness of ED documentation is essential for continuity of patient care and flow of information to care providers. It is during residency that good charting practices can be imbibed. Our study seeks to detect deficiencies in documentation of ED encounters by residents and suggest recommendations for improvement.
Methods: Charts were examined from both an academic and a community Emergency Department. Data was extracted from charts using a Microsoft Excel™ spreadsheet with established fields, based on areas of documentation considered important. The data was qualitatively analyzed to assess adequacy of charting and identify areas of deficiency. Ethics approval was obtained for the study.
Results: A total of 500 charts were reviewed. Most charts were deemed legible (96%). Time of assessment and adequate history and physical examination were documented 100% of the time and a review of vital signs was charted 82% of the time. Reassessment time and note was only documented in 47% of charts where reassessment was expected. Also, 20% of charts lacked a record of past medical history and 34% of charts lacked documentation of the patient’s medications being reviewed. Allergies were documented as being reviewed 38% of the time. While discharge instructions were provided 91% of the time, a clear impression and plan was only documented 59.6% of the time. A communique with the primary care provider was documented 41% of the time. Signoff and review by the attending physician was documented 40% and 27.6% of the time respectively. Discharge diagnosis was present in 94.6% of charts reviewed.
Conclusion: Clear improvements are necessary in ED resident documentation of reassessments, patients’ medications and allergies, primary care follow-up and review by the attending physician. Based on the deficiency in charting practices identified here, we can now implement an approach to improve documentation skills for residents rotating in the ED. Future endeavours will involve creation of an audit tool that emergency departments can use to audit their rotating residents’ charting practices. Our exercise is an initial step towards efforts in improving charting in emergency department encounters.
Keywords: quality improvement, documentation, emergency department
Season Two: Detection of Protein S100B in plasma and urine after a mild traumatic brain injury (en français)
N. Le Sage, MD, MSc, J. Frenette, PhD, M. Emond, MD, MSc, J. Chauny, MD, MSc, P.M. Archambault, MSc, MD, J.J. Perry, MD,MSc, L.Moore, PhD, M. Roy, PhD, P. Bouchard, MSc; Université Laval, Quebec, QC
Introduction: Mild traumatic brain injuries (mTBI) are frequent and even if they sometimes may seem minor, their consequences can be worrisome. Among potential new tools to help a better identification of patients at risk of persistent symptoms after a mTBI, a number of biomarkers have been described to be elevated after a mTBI of which the S100B protein is probably the most promising. Current literature describe that this particular protein can be detected in blood, cerebrospinal fluid or even urine. Urinalysis is a minimally invasive method, but the concentration and stability of proteins could possibly interfere with its use. The main objective of this study was to assess whether S100B concentrations can be measured in urine when detectable in plasma after a mTBI.
Methods: This study is a planned sub-analysis of a larger prospective cohort study involving the follow-up of mTBI patients. Eligible all patients who met the following inclusion criteria: patient sustained a mTBI, aged 14 years or more, assessed in the emergency department (ED) of a tertiary trauma center within 24 hours of injury, and who did not require hospitalization. For each patient, relevant clinical data and blood and urine sample were collected at the initial ED visit. The exact time of trauma and samplings were recorded for all patients. After adequate manipulation, serum and urine were be stored at -80C. The concentrations of S100B, were analyzed by ELISA. Results: One hundred and thirty-six patients with mTBI were included in this study. Forty-seven of them had detectable S100B protein in plasma, however the protein was not detectable in urine, even when plasmatic concentration was very high.
Results: One hundred and thirty-six patients with mTBI were included in this study. Forty-seven of them had detectable S100B protein in plasma, however the protein was not detectable in urine, even when plasmatic concentration was very high.
Conclusion: S100B protein elevation in plasma does not cause a detectable concentration in urine. Otherwise, further research will be necessary to better identify the exact significance of detectable plasmatic S100B protein after a mTBI.
Keywords: mild traumatic brain injuries, biomarkers, concussion
Season Two: Optimal shift duration for emergency physician efficiency, effectiveness and safety: a comparison of 6, 7, and 8-hour shifts
M. C. Foster, MBA, Z. Sun, PhD,D.Wang, MSc, G. Innes,MD, L. Baker, MD, A. McRae, MD, E. Lang, MD; University of Calgary, Calgary, AB
Introduction: Emergency departments require 24-hour physician coverage, and must develop a schedule that balances patient safety with physician efficiency and preferences. There are no studies comparing shift length in regards to safety, metrics and productivity. Our objective is to determine if there is a difference in the efficiency of patients seen per hour between six, seven, or eight-hour shifts in the ED. Secondary outcomes include the number of patient handovers, and unscheduled ED revisit rates associated with different shift lengths.
Methods: This retrospective study is based on one urban ED, where 81 physicians provided care for 79,941 visits during the study period (September 1, 2013 to August 31, 2014). 41 physicians met our inclusion criteria, working a minimum of 30 shifts of varying lengths over this period. Minor treatment shifts were excluded from analysis. Administrative data and an online scheduling system was used to compare scheduled shift length and number of patients seen per hour as well as percentage of patients handed over to the next physician from the total seen that shift, and return visits within 72 hours. One way analysis of variance and t-tests were used to compare the means between 3 different shift types.
Results: A total of 3,214 shifts of varying start times (1,467 6-hour shifts, 531 seven-hour shifts and 1,216 8-hour shifts) were included. Mean start times for six, seven and eight-hour shifts were 12:00, 17:00 and 10:00 respectively. The average number of patients seen per hour for six, seven and eight-hour shifts was 2.56 (95%CI 2.53-2.59), 2.75 (95%CI 2.68-2.82) and 2.50 (95%CI 2.47-2.53), respectively; p <0.001 for comparison of seven against six and eight-hour shifts. The average handover rate for six, seven, and eight-hour shifts was 22.14 (95%CI 21.50-22.78), 27.45 (95%CI 26.37-28.53) and 17.36 (95%CI 16.86-17.87) percent, respectively; p <0.001 for the comparison of eight against six and seven. All night shifts in this ED are 7 hours; a limitation of this result is the inclusion of night shifts which have higher handover rates. There was no significant difference between shifts for 72-hour return rates.
Conclusion: In this comparison, a seven-hour shift duration offers optimal physician efficiency, while eight-hour shifts may be more desirable from a safety perspective, as fewer handovers mean less opportunity for error.
Keywords: efficiency, safety, ED
Season Two: How changes to the Ontario Highway Traffic Act have affected the proportion of alcohol-impaired motor vehicle collisions seen at a Level I Trauma Centre over a 10-year period
M. Garnett, BHSc, MD, T. Charyk-Stewart, BSc, MSc, M. Miller, PhD, R. Lim, MD, K. Van Aarsen, BSc, MSc, W. Millard, MD; Western, London, ON
Introduction: Despite widespread public education on the dangers of drinking and driving over the last 30 years, impaired driving is still the leading cause of criminal death in Canada. Since May 1, 2009, all drivers in Ontario with a blood alcohol concentration (BAC) between 0.05-0.08% receive a three-day roadside license suspension. Since August 1, 2010, all novice drivers and drivers under 22 years must maintain a BAC of zero at all times.
Methods: This was a retrospective review of the trauma registry at a Level I Trauma Centre in Southern Ontario from 2004-2013. Descriptive statistics (total, percentage, mean, medium, interquartile range, range) were calculated on demographic, crash, injury, and hospital variables for all impaired (ethanol >0 mmol/L) drivers involved in a motor vehicle collision (MVC) with Injury Severity Score (ISS) ≥12 or Trauma Team Activation.
Results: 377 alcohol-impaired drivers were treated at our trauma centre over the 10-year period, representing 20.9% of all MVC drivers. The majority (330; 87.5%) were male. The median age was 31 (IQR = 23) years, and median ethanol was 35.3 mmol/L. Only 32.3% (n = 142) of drivers were wearing their seatbelts, and 112 (25.5%) had no protective devices in use at all. Over three-quarters (292; 77.5%) were single vehicle MVCs, with 41.9% (n = 158) of vehicles impacting a fixed object, and 35.5% (n = 134) were vehicle rollovers. The median ISS for impaired drivers was 20 (IQR = 20). A total of 29 patients (7.7%) succumbed to their injuries. The difference between the proportion of drivers who were impaired before and after the legislative change on May 1, 2009, was not significant (Χ2 = 3.31, p = 0.069), but there was a decline after stricter legislation was introduced for novice and young drivers on August 1, 2010 (Χ2 = 8.37, p = 0.004). A time series analysis of the data will be presented at CAEP 2015.
Conclusion: Alcohol-impaired driving is still an important cause of severe trauma, resulting in significant injuries and a substantial impact on the health care system. Changes to the Highway Traffic Act affecting novice and young drivers made a difference in the proportion of alcohol-related MVCs. The role of tougher laws, along with public health education and awareness campaigns, need to continue to be explored to reduce the number of alcohol-related crashes and injury recidivism.
Keywords: trauma, motor vehicle collision, ethanol
Season Two: The timely availability and correlates of code status in oncology patients presenting to the ED
K. Caners, BSc, MD, A. Pardhan, BSc, MD, MBA, V. Sandu, BSc, X. Chen, BHSc, W. Bhanich Supapol, PhD, C. Langmann, MD, PhD, P. Miller, MD; McMaster University, Hamilton, ON
Introduction: Oncology patients frequently present to the emergency department (ED) in the last six months of their life. These visits are associated with a high admission rate and 13-20% of these visits lead to death. There are no specific studies on advance directives in oncology patients. The primary objective of this study was to determine the percentage of oncology patients presenting to the ED who have a code status identifiable within their electronic medical record.
Methods: First, a convenience sample physician cased-based survey was conducted. Second, a retrospective chart review was performed by two independent reviewers who were blinded to the study objectives. All adult patients with an ED presentation leading to admission under an oncology service over a four month period were included. Backward stepwise logistic regression was used to determine which variables affected code status documentation.
Results: 41 ED physicians responded to the survey. Physicians indicated they would not search for a code status in 36.9% of cases. For those cases in which physicians would search for a code status, the median acceptable time frame was 5 minutes (IQR 3,10). Code status was documented in 34 (16.2%) of the 209 charts reviewed. The longest documented search time was 4.3 minutes. Search time was less than two minutes for the majority (79.4%) of patients with a documented code status. Age, gender, and presence of metastases did not influence likelihood of code status documentation. Previous admission to an oncology service through the ED was significantly associated with having a code status documented (OR 9.30, 95% CI 3.97, 21.70).
Conclusion: Rates of code status documentation among oncology patients presenting to the ED are quite low. Time required to locate a code status within the patient’s electronic medical record is surprisingly brief, with the majority obtained within two minutes.
Keywords: code status, oncology, medical record
Season Two: Regional changes in Methicillin-resistant Staphylococcus aureus in purulent skin and soft tissue infections among patients presenting to Canadian emergency departments
MP34 Regional changes in Methicillin-resistant Staphylococcus aureus in purulent skin and soft tissue infections among patients presenting to Canadian emergency departments
B. Borgundvaag, PhD, MD, W. Ng, MHSc, I. Cajee, MD, M. Clory, MD, P. Ellis, MD, M. Emond, MD, MSc, J.S. Lee, MD, MSc, A. McGeer, MD, MSc, A. McRae, MD, V. Porter, MD, B.H. Rowe, MD, MSc, D. Scolnik, MD, A. Simor, MD, B. Willey, MSc, K. Katz, MD; Schwartz/Reisman Emergency Medicine Institute, Toronto, ON
Introduction: Community-associated methicillin-resistant Staphylococcus aureus (MRSA) has emerged as a leading cause of purulent skin and soft tissue infections (SSTI). The evolving prevalence of MRSA in SSTIs across Canada is not well described. The objective of this study was to characterize the changing prevalence and microbiology of MRSA in patients presenting to emergency departments (EDs) across Canada over a five-year study period.
Methods: Using a prospective, observational design, patients presenting to 26 hospital EDs in seven provinces with acute purulent SSTIs were enrolled over three phases: P1-7/1/2008 to 4/30/2009; P2-1/16/2012 to 11/30/2012; and P3-4/28/2013 to 3/31/2014. Participating EDs collected wound swabs on all patients with purulent SSTIs. Eligible patients were those whose wound cultures grew S. aureus. Standard antimicrobial susceptibility testing was undertaken on all isolates. Pulsed-field gel electrophoresis, PVL gene (PCR), SCCmec work is ongoing on MRSA isolates.
Results: There were 4,680 (P1: 1340; P2: 1623; P3: 1717) S. aureus positive encounters during the study period. Accounting for all sites, the overall MRSA prevalence decreased significantly between P1 (31%) and P2 (27%, p=0.005), and remained unchanged in P3 (27%, p=0.42). A similar trend was observed among the 12 sites that participated in all 3 phases. Among the 18 sites participating in at least two study phases, most (56%) experienced a declining trend in MRSA prevalence, while 28% observed an increase (3 Ontario and 2 Alberta sites). City-level analyses revealed variability in the MRSA prevalence. Most cities experienced a decrease in the prevalence, with Calgary, AB observing the greatest decrease (P1: 91%, P3: 29%, p<0.01). The highest prevalence was seen in British Columbia (P1: 44%, P2: 66%, P3: 53%), Saskatchewan (P2: 47%, P3: 44%), and Alberta (P1: 48%, P2: 28%, P3: 30%), while the lowest prevalence was observed in Quebec (P1: 20%, P2: 19%, P3: 11%).
Conclusion: MRSA prevalence continues to evolve across Canada. While the overall Canadian prevalence of MRSA in SSTIs remains substantial, it is variable across the country and may be decreasing regionally.
Keywords: MRSA, skin and soft tissue infection, emergency medicine
Season One: Ken Milne hears from Sally Carver about a new app that helps to assess alcohol withdrawal
S.M. Carver, BSc, N. Norouzi, BEng., S. Bromberg, BASc, S.H. Gray, MD, M. Kahan, MD, P. Aarabi, PhD, B. Borgundvaag, PhD, MD, Mount Sinai Hospital, Toronto, ON
Introduction: Alcohol withdrawal syndrome is potentially life-threatening. It is commonly encountered and poorly managed in the Emergency Department (ED). A symptom-guided approach to the treatment of alcohol withdrawal has the potential to improve care of these patients in the ED by reducing the total dose of benzodiazepine administered, reducing complication rates such as seizures while in the ED and after discharge, and potentially reducing the length of stay for these patients. Symptom-guided tools, such as the CIWA-Ar, are not routinely used because they are time-consuming, subjective, and amenable to manipulation by benzodiazepine-seeking patients. The most universally accepted clinical measure of alcohol withdrawal severity is tremor, the assessment of which is subjective and variable. We developed an iOS CIWA-Ar mobile application (‘app’) which standardizes tremor assessment using the phone’s accelerometer. A secondary question was whether it is possible to discriminate a real alcohol withdrawal tremor from an intentional attempt to deceive the assessor. Methods: Data were collected from 10 patients in alcohol withdrawal and 12 nurses mimicking an alcohol withdrawal tremor. Assessments were videotaped and compared against “gold-standard”, subjective assessments of three expert clinicians according to the CIWA-Ar scale. We used time-frequency analysis to characterize the tremors, specifically to differentiate real from factitious tremors, and to calibrate tremor intensity.
Results: Real alcohol withdrawal tremors had a mean frequency of 8.1 Hz and a standard deviation of 1.2 Hz. By rejecting tremors with an average frequency below 7 Hz, real tremors could be detected with 16% false positives and 13% false negatives. Additionally, there was a general linear relationship between the CIWA-Ar tremor score and the energy of the signal such that the higher the CIWA-Ar tremor rating, the higher the energy in the 7-16 Hz range.
Conclusion: It is possible to both quantify tremor severity and differentiate most real and fake alcohol withdrawal tremors. Ongoing work will further characterize the tremor of alcohol withdrawal and refine the calibration of the app.
Keywords: alcohol withdrawal, tremor, emergency department
Season One: Ken Milne learns more about the McMAP from Teresa Chan
T.M. Chan, MD, J. Sherbino, BSc, MD, MEd, McMaster University, Hamilton, ON
Introduction: The McMAP system contains a wide variety of workplace based assessment (WBA) tools that are assembled into Can-MEDS-based units, while longitudinally assessing global functioning. At the end of block, the assessments are compiled into a score report for supervising faculty members, who then complete a guided descriptive, narrative overview of the resident’s performance (the McMAP intraining evaluation record, ITER).
Methods: 41 McMAP ITERs written by 25 faculty members (out of a possible 64 McMAP ITERs) were available. We rated the quality of our McMAP ITERs to a historical cohort of random ITERs prior to our pilot launch. A random cohort of 25 McMAP ITERs and a matched historical cohort of 25 ITERs from PGY1 & 2 residents from the pre-McMAP period were randomly selected. A program administrator redacted all ITERs to anonymize the documents. All 50 ITERs (both pre- and post-McMAP) were scored by two investigators (TC, JS) using the Completed Clinical Evaluation Record Rating (CCERR).
Results: The level of agreement between our two raters on the CCERR scale was alpha=0.916 (df=49, p<0.001). There was a significant increase (p<0.001) in the median CCERR scores after the McMAP was introduced, resulting in a 15.7 point jump (out of the 45 point CCERR scale). The median CCERR score of our McMAP ITERs was 27.5 (IQR 20.5-23.5), whereas our historical cohort was 13.8 (IQR 11.3-15.8).
Conclusion: Our McMAP ITER system generated significantly improved quality assessments, as judged by the CCERR scale. Summarized data and scaffolded questions can greatly improve ITER quality.
Keywords: residency education, assessment, competency-based medical education
Season One: Ken Milne learns from Kelly Van Diepen how to meet them, greet them, treat them, and street them
K. van Diepen, MD, D.M. Frost, PhD, J. Sherbino, BSc, MD, MEd, C.Langmann, MD, PhD, W. Bhanich Supapol, PhD, I. Preyra, MD, McMaster University, Hamilton, ON
Introduction: Emergency Department (ED) overcrowding is a significant international problem. Placing a physician in triage to rapidly
assess patients and initiate diagnostic testing is cited as a method of improving ED throughput and reducing overcrowding. The goal of this study was to assess the impact of an initial assessment physician (IAP) on measures of ED throughput by time of day, patient acuity and admission and consultation status.
Methods: This was a retrospective pre-post intervention study that examined the impact of an IAP on measures of ED overcrowding in one academic teaching hospital. Pre and post IAP time periods were matched by time of year (April 2nd to June 2nd, 2012 and April 2nd to June 2nd, 2013). Patients were also grouped and compared by acuity and time of day based on IAP hours (11:00 to 21:00 and 21:00 to 11:00). Data was retrieved from an electronic ED tracking board. The primary outcome measures included time to physician initial assessment (PIA), emergency department length of stay (EDLOS) for non-admitted patients with and without consultation, and EDLOS for admitted patients. The percentage of patients who left without being seen (LWBS) by a physician was a secondary outcome measure.
Results: In total, 18 371 patient were included in the 4 month study period. Despite an increase in ED volume and admission rates in 2013, the median reduction in time to PIA for all patients seen by the IAP was 86 minutes (110 to 24 minutes; p<0.001). The median reduction in EDLOS for all non-admitted patients who did not receive a consultation was 27 minutes (211 to 184 minutes; p<0.001) while the EDLOS for non-admitted patients who received a consultation was reduced by a median of 60 minutes (388 to 328 minutes; p<0.001). The IAP did not reduce EDLOS for admitted patients. Though low at baseline (1.2%), LWBS rates decreased by 50% during IAP hours.
Conclusion: Results of this study suggest that an IAP improves ED throughput and measures of ED overcrowding.
Keywords: triage physician, emergency overcrowding, throughput
Season One: Ken Milne is in the zone with James Gould
A. Carter, MD, J. Gould, BSc, P. Vanberkel, PhD PEng, J.L. Jensen, BSc, J. Cook, MD, S. Carrigan, MHA, M. Wheatley, A. Travers, MD, MSc, EHS, Halifax, NS
Introduction: Offload delay is a prolongation of the interval between ambulance arrival in emergency department (ED) and transfer of patient care. This reduces the availability of ambulances for emergency response in the community. Offload Zones (OZ), which can receive multiple ambulance patients waiting for an ED bed and release paramedics from the ED, have been implemented as a possible mitigation. The objectives of this study were to process map the functioning of the OZ and conduct a hazard analysis to identify steps that could compromise patient safety or process efficiency.
Methods: A Health Care Failure Mode and Effect Analysis was conducted. All major processes and steps were identified by observation and expert opinion. Hazard analysis was conducted by focus group. Failure modes (FM) were identified for each step. For each FM a probability to occur (1-4) and severity of impact on patient safety and process efficiency (1-4) was determined and a hazard score (probability X severity: 1 -16) was calculated. Any hazard score ≥8 was considered ‘high risk’. Root causes were identified for all ‘high risk’ FM. A hazard score was calculated for those causes. Mitigations were sought for ‘high risk’ causes with no control measure.
Results: The OZ consists of six major processes: (1) Patient transported by ambulance, (2) Arrival in ED, (3) Transfer of patient Care, (4) Patient assessment in OZ, (5) Patient care in OZ and (6) Patient transfer out of OZ to ED. There were 110 steps within these major processes categorized as decisions (n=13, 11.8%), skills (n=29, 26.4%) and tasks (n=68, 61.8%). 78 FM were identified, of which 28 (35.9%) were ‘high risk’: patient safety (n=7/28, 25.0%), process efficiency (n=10/28, 35.7%), and both (n=11/28, 39.3%). High scoring FM were lack of equipment (safety) and failure to move patient to ED bed (process). 57 causes for ‘high risk’ FM were identified of which 37 were ‘high risk’ with no control measure. 17 mitigations were suggested for those ‘high risk’ causes with no control measure.
Conclusion: This process map and hazard analysis has prospectively identified a number of potential failures of the OZ. The results from this study will inform current policy and practice, and future work to understand the use of the OZ to reduce offload delay.
Keywords: emergency medical services, emergency overcrowding
Season One: Ken Milne gets the chills from Chris Fabian (University of Ottawa)
J.C. Fabian, MD, J.J. Perry, MD, MSc, M. Taljaard, PhD, M. Tran, S. Agaybi, BSc, University of Ottawa, Ottawa, ON
Introduction: Frostbite treatment guidelines are based on observational data from the 1960s and rational approaches based on physiology. Their effectiveness in preventing surgical intervention has yet to be examined. Our objectives were to characterized frostbite injuries in Eastern Ontario, their associated risk factors, and assess the efficacy of current treatment guidelines in preventing operative outcomes.
Methods: A historical cohort study examined patients treated for frostbite over 10 years at the two emergency departments of The Ottawa Hospital. Patient demographics and risk factors were recorded on standardized data extraction forms. Frostbite severity was categorized into superficial (1st and 2nd degree) or deep (3rd and 4th degree). Treatments provided were recorded, including adherence to guidelines (twelve), and surgical interventions. Results: Of the 265 frostbite patients identified, mean age was 37.5 years old, 74.0% were male, 93.7% of injuries occurred between November to March and 21.9% were admitted to hospital. Deep frostbite accounted for 17.7% of patients, of which 14.9% had an amputation, and 10.6% had debridement. Superficial frostbite accounted for 82.3% of patients, of which 4.1% had an amputation, and 2.3% had debridement. Risk factors included alcohol abuse (14.7%), smoking (10.6%), illicit drug use (6.4%), psychiatric illness of any kind (20.8%), schizophrenia (9.1%), mood disorders (6.0%), drug induced psychosis (1.9%), suicide attempts (1.1%), motor vehicle collisions (1.5%), skidoo use (0.8%),
winter sports (8.3%) and inadequate footwear (32.1%). Of the twelve treatment guidelines, adherence of each ranged from 0.8% to 32.8% per guideline. Prevalence of amputation or debridement was 7.7% among patients not adherent to guidelines and 42.9% in patients adherent to guidelines.
Conclusion: This study found that alcohol abuse, smoking, psychiatric illnesses, winter sports and poor footwear were
prevalent frostbite risk factors. Frostbite treatment guidelines were not regularly followed. Amputation and debridement rates did not appear to improve when patients followed treatment guidelines.
Keywords: frostbite, treatment