Patients with cardiovascular diseases are common in the emergency department (ED), and continuity of care following that visit is needed to ensure that they receive evidence-based diagnostic tests and therapy. We examined the frequency of follow-up care after discharge from an ED with a new diagnosis of one of three cardiovascular diseases.

We performed a retrospective cohort study of patients with a new diagnosis of heart failure, atrial fibrillation, or hypertension, who were discharged from 157 non-pediatric EDs in Ontario, Canada, between April 2007 and March 2014. We determined the frequency of follow-up care with a family physician, cardiologist, or internist within seven and 30 days, and assessed the association of patient, emergency physician, and family physician characteristics with obtaining follow-up care using cause-specific hazard modeling.

There were 41,485 qualifying ED visits. Just under half (47.0%) had follow-up care within seven days, with 78.7% seen by 30 days. Patients with serious comorbidities (renal failure, dementia, COPD, stroke, coronary artery disease, and cancer) had a lower adjusted hazard of obtaining 7-day follow-up care (HRs 0.77-0.95) and 30-day follow-up care (HR 0.76-0.95). The only emergency physician characteristic associated with follow-up care was 5-year emergency medicine specialty training (HR 1.11). Compared to those whose family physician was remunerated via a primarily fee-for-service model, patients were less likely to obtain 7-day follow-up care if their family physician was remunerated via three types of capitation models (HR 0.72, 0.81, 0.85) or via traditional fee-for-service (HR 0.91). Findings were similar for 30-day follow-up care.

Only half of patients discharged from an ED with a new diagnosis of atrial fibrillation, heart failure, and hypertension were seen within a week of being discharged. Patients with significant comorbidities were less likely to obtain follow-up care, as were those with a family physician who was remunerated via primarily capitation methods.

Vertigo is common in the emergency department (ED). Most aetiologies are peripheral and do not require hospitalization, but many patients still fear falling. Some patients may be taking opioid analgesic medications (for other reasons); the risk of falls leading to fractures among patients with vertigo could be potentiated by the simultaneous use of opioids.

To examine the risk of fractures in discharged ED patients with peripheral vertigo who were being prescribed opioids during the same time period.

Linked administrative databases from Ontario were used to compare discharged ED patients aged ≥65 with peripheral vertigo to patients with urinary tract infection (UTI) from 2006 to 2011. We used Cox regression analysis with an interaction term to estimate the modifying effect of an opioid prescription on the hazard of fracture within 90 days.

There were 13,012 patients with a peripheral vertigo syndrome and 76,885 with a UTI. Thirteen percent of the vertigo cohort and 25% of the UTI cohort had access to a filled opioid prescription. Compared to vertigo patients who did not fill an opioid prescription, the adjusted hazard of fracture among vertigo patients who did fill a prescription was 3.59 (95% CI 1.97–6.13). Among UTI patients who filled an opioid prescription the hazard ratio was 1.68 (95% CI 1.43–1.97) compared to UTI patients who did not.

Patients discharged from the ED with peripheral vertigo who were also being prescribed opioids had a higher hazard of subsequent fracture compared to those who were not, and the effect was much greater than among UTI patients. These results suggest that in the acutely vertiginous older patient, opioid analgesic medications should be modified, where possible.

Many patients are seen in the emergency department (ED) for hypertension, and the numbers will likely increase in the future. Given limited evidence to guide the management of such patients, the practice of one’s peers provides a de facto standard.

A survey was distributed to emergency physicians during academic rounds at three community and four tertiary EDs. The primary outcome measure was the proportion of participants who had a blood pressure (BP) threshold at which they would offer a new antihypertensive prescription to patients they were sending home from the ED. Secondary outcomes included patient- and provider-level factors associated with initiating an antihypertensive based on clinical vignettes of a 69-year-old man with two levels of hypertension (160/100 vs 200/110 mm Hg), as well as the recommended number of days after which to follow up with a primary care provider following ED discharge.

All 81 surveys were completed (100%). Half (51.9%; 95% CI 40.5-63.1) of participants indicated that they had a systolic BP threshold for initiating an antihypertensive, and 55.6% (95% CI 44.1-66.6) had a diastolic threshold: mean systolic threshold was 199 mm Hg (SD 19) while diastolic was 111 mm Hg (SD 8). A higher BP (OR 12.9; 95% CI 7.5-22.2) and more patient comorbidities (OR 3.0; 95% CI 2.1-4.3) were associated with offering an antihypertensive prescription, while physician years of practice, certification type, and hospital type were not. Participants recommended follow-up care within a median 7.0 and 3.0 days for the patient with lower and higher BP levels, respectively.

Half of surveyed emergency physicians report having a BP threshold to start an antihypertensive; BP levels and number of patient comorbidities were associated with a modification of the decision, while physician characteristics were not. Most physicians recommended follow-up care within seven days of ED discharge.

The American Heart Association (AHA) recommends a benchmark door-to-electrocardiogram (ECG) time of 10 minutes for acute myocardial infarction patients, but this is based on expert opinion (level of evidence C). We sought to establish an evidence-based benchmark door-to-ECG time.

This retrospective cohort study used a population-based sample of patients who suffered an ST elevation myocardial infarction (STEMI) in Ontario between 1999 and 2001. Using cubic smoothing splines, we described (1) the relationship between door-to-ECG time and ECG-to-needle time and (2) the proportion of STEMI patients who met the benchmark door-to-needle time of 30 minutes based on their door-to-ECG time. We hypothesized nonlinear relationships and sought to identify an inflection point in the latter curve that would define the most efficient (benefit the greatest number of patients) door-to-ECG time.

In 2,961 STEMI patients, the median door-to-ECG and ECG-to-needle times were 8.0 and 27.0 minutes, respectively. There was a linear increase in ECG-to-needle time as the door-to-ECG time increased, up to approximately 30 minutes, after which the ECG-to-needle time remained constant at 53 minutes. The inflection point in the probability of achieving the benchmark door-to-needle time occurred at 4 minutes, after which it decreased linearly, with every minute of door-to-ECG time decreasing the average probability of achievement by 2.2%.

Hospitals that are not meeting benchmark reperfusion times may improve performance by decreasing door-to-ECG times, even if they are meeting the current AHA benchmark door-to-ECG time. The highest probability of meeting the reperfusion target time for fibrinolytic administration is associated with a door-to-ECG time of 4 minutes or less.

Current guidelines suggest that most patients who present to an emergency department (ED) with chest pain should be placed on a continuous electrocardiographic monitoring (CEM) device. We surveyed emergency physicians to determine their perception of current occupancy rates of CEM and to assess their attitudes toward prescribing monitors for low-risk chest pain patients in the ED.

We conducted a cross-sectional, self-administered Internet and mail survey of a random sample of 300 members of the Canadian Association of Emergency Physicians. Main outcome measures included the perceived frequency of fully occupied monitors in the ED and physicians' willingness to forgo CEM in certain chest pain patients.

The response rate was 66% (199 respondents). The largest group of respondents (43%; 95% confidence interval [CI] 36%–50%) indicated that monitors were fully occupied 90%–100% of the time during their most recent ED shift. When asked how often they were forced to choose a patient for monitor removal because of the limited number of monitors, 52% (95% CI 45%–60%) of respondents selected 1–3 times per shift. Ninety percent (95% CI 84%–93%) of respondents indicated that they would forgo CEM in certain cardiac chest pain patients if there was good evidence that the risk of a monitor-detected adverse event was very low.

Emergency physicians report that monitors are often fully occupied in Canadian EDs, and most are willing to forgo CEM in certain chest pain patients. A large prospective study of CEM in low-risk chest pain patients is warranted.