Nonadherence to prescribed medication is associated with increased morbidity and mortality as well as the increased use of health services. The main objective of our study was to assess the incidence of prescription-filling and medication adherence in patients discharged from the emergency department (ED).

This was a prospective, observational study carried out at a Canadian tertiary care ED with an annual census of 69 000. We enrolled a convenience sample of patients being discharged with a prescription. We queried a provincial prescription-dispensing database 2 weeks later to determine whether prescriptions had been filled. We used a standardized follow-up interview to assess adherence and whether or not the patient experienced an adverse drug-related event (ADRE) or an unplanned revisit to an ED or clinic.

Of the 301 patients who agreed to participate, follow-up was successful for 258 (85.7%). Fifty-one patients (19.8%, 95% confidence interval [CI] 15.4%–25.1%) failed to fill their discharge prescriptions and 104 (40.3%, 95% CI 34.5%–46.4%) did not adhere to 1 or more medications. Antibiotics were associated with a lower odds ratio (OR) of nonadherence (OR 0.21, 95% CI 0.08–0.52). There was a trend toward increasing nonadherence in patients who reported an ADRE (OR 1.84, 95% CI 0.98–3.48) or had 2 or more medications coprescribed (OR 1.71, 95% CI 0.95–3.09). There was also a trend toward a higher risk of a revisit to an ED or clinic in nonadherent patients (OR 1.75, 95% CI 0.94–3.25).

Approximately 4 in 10 patients discharged from the ED did not adhere to his or her prescribed medication. Our results suggest that patients who are prescribed antibiotics are more likely to be adherent, and that further evaluation of the associations between nonadherence, ADREs, the coprescription of 2 or more medications and the use of health services is warranted.

We compared the appropriateness of visits to a pediatric emergency department (ED) by provincial telephone health line–referral, by self- or parent-referral, and by physician-referral.

A cohort of patients younger than 18 years of age who presented to a pediatric ED during any of four 1-week study periods were prospectively enrolled. The cohort consisted of all patients who were referred to the ED by a provincial telephone health line or by a physician. For each patient referred by the health line, the next patient who was self- or parent-referred was also enrolled. The primary outcome was visit appropriateness, which was determined using previously published explicit criteria. Secondary outcomes included the treating physician's view of appropriateness, disposition (hospital admission or discharge), treatment, investigations and the length of stay in the ED.

Of the 578 patients who were enrolled, 129 were referred from the health line, 102 were either self- or parent-referred, and 347 were physician-referred. Groups were similar at baseline for sex, but health line–referred patients were significantly younger. Using explicitly set criteria, there was no significant difference in visit appropriateness among the health line–referrals (66%), the self- or parent-referrals (77%) and the physician-referrals (73%) (p = 0.11). However, when the examining physician determined visit appropriateness, physician-referred patients (80%) were deemed appropriate significantly more often than those referred by the health line (56%, p < 0.001) or by self- or parent-referral (63%, p = 0.002). There was no significant difference between these latter 2 referral routes (p = 0.50). In keeping with their greater acuity, physician-referred patients were significantly more likely to have investigations, receive some treatment, be admitted to hospital and have longer lengths of stay. Patients who were self- or parent-referred, and those who were health line–referred were similar to each other in these outcomes.

There was no significant difference in visit appropriateness based on the route of referral when we used set criteria; however, there was when we used treating physician opinion, triage category and resource use.

We sought to determine the degree and possible causes of variability in admission practices among individual emergency physicians (EPs) at 1 emergency department (ED) using a Canadian Emergency Department Triage Acuity Scale (CTAS)–matched ED patient population.

We distributed a survey measuring attitudes and demographics to all EPs (n = 30) at a large regional hospital. Hospital admissions data from 1 calendar year were matched to individual EP survey results. Emergency physicians were ranked as “lower,” “average” or “higher” admitters and, using these categorical variables, the data set was analyzed for correlations and trends.

Overall, 97.0% of the EPs responded to the survey. Admissions by EPs ranged from 8.7% to 17.0%, (mean 12.52, standard deviation [SD] 2.21) of all patients seen. CTAS category–specific admission data demonstrated variability in the admission ranking of individual EPs. No EPs consistently performed at any 1 admission ranking across all CTAS categories. More years of emergency medicine experience was significantly correlated with higher admissions in the CTAS-2 ranking (r = 0.4, p < 0.05). Whether a physician worked full-time, part-time or as a locum was not associated with patterns of admission, nor was any particular postgraduate certification (e.g., CCFP, CCFP EM, FRCPC) or any of the surveyed attitudinal traits.

Individual EPs' overall and CTAS-specific admissions varied substantially, and followed an approximately normal distribution curve. Emergency physicians with more years of experience had a statistically higher CTAS-2 admission rate; however, other variables, including postgraduate certification status, decision-related attitudes toward admission, and reported practices were not associated with admission proportions. Emergency physicians tend to have uniquely individual admission ranking profiles across all the CTAS categories.

We sought to determine the incidence of acute myocardial infarction (AMI) in emergency department (ED) patients with syncope, the characteristics of these AMIs and how helpful the initial electrocardiogram (ECG) was in identifying these cases.

In a prospective cohort of consecutive patients with syncope, the initial ECG was found to be abnormal using a prespecified definition (any nonsinus rhythm or any new or age-indeterminate abnormalities). Patients were then followed up to identify an AMI diagnosed within 30 days of presentation.

There were 1474 consecutive patient visits for syncope or near-syncope over a 45-month period spanning from Jul. 1, 2000, to Feb. 28, 2002, and Jul. 15, 2002, to Aug. 31, 2004, of which 46 (3.1%) were diagnosed with AMI. The majority of the AMI patients (42) had no ST segment elevation. The initial ECG was abnormal in 37 out of 46 cases. The diagnostic performance of the initial ECG was sensitivity 80% (95% confidence interval [CI] 67%–89%), specificity 64% (95% CI 61%–67%), negative predictive value 99% (95% CI 98%–100%), positive predictive value 7% (95% CI 6%–8%), positive likelihood ratio 2.2 (95% CI 1.6–2.5) and negative likelihood ratio 0.3 (95% CI 0.2–0.5).

The incidence of AMI in patients presenting with syncope is low. A normal ECG has a high negative predictive value, although its sensitivity is limited.

Youth violence continues to trouble Canadians. Emergency department (ED) visits by youth after a violent injury may represent a “teachable moment,” and thus secondary violence prevention interventions may be effective. We conducted a systematic review to identify the success rates of any interventions, the populations likely to benefit and the outcome measures used.

We searched 8 databases (i.e., MEDLINE, EMBASE, PubMed, CINAHL, the Cochrane Database of Systematic Reviews, the ACP Journal Club, DARE and CENTRAL).

Studies were included if they described and evaluated an intervention, were health care–based and targeted youth who were injured by violence. Two blinded investigators selected 15 articles from 181 abstracts. After full-text review, 8 articles were excluded, leaving 7 articles from 4 intervention programs.

All interventions used ED case management of the violently injured patient. One randomized control trial (RCT) demonstrated a significant reduction in reinjury rates (treatment group 8.1% v. control group 20.3%, p = 0.05). Another small RCT found no statistically significant reductions in repeat violence or service use. One retrospective cohort study demonstrated a lower relative risk (RR) in future criminal justice involvement (RR = 0.67, 95% confidence interval 0.45–0.99). A retrospective study of pediatric patients with violent injuries found only 1% of these youth returned with injuries as a result of repeat violence.

Although all 4 case management interventions that we reviewed showed promise in the United States, small sample sizes and incomplete follow-up limited their ability to demonstrate significant decreases in reinjury.

Future research is necessary to help EDs capitalize on the opportunity to effectively reduce youth violence.