Learning from our mistakes

Finding a way to safely report and learn from medical errors is always a challenge. In this issue, Hohenstein and colleagues describe a web-based critical incident reporting system that allows anonymous submission for German and Austrian out-of-hospital providers. Fifteen per cent of the reported incidents clearly affected patient outcomes (and another quarter may have). Among the incidents causing harm, almost 80% were related to staff issues, suggesting that the system may have a lot to tell us about training gaps.

Adults only

The elbow is nobody's favourite body part, but it finally gets its day in the sun. Based on results from 348 adult patients with elbow injury, Arundel et al derive a 100% sensitive (95% CI: 97% to 100%), 3-element decision rule for elbow fracture. The elements include: inability to fully extend; tenderness over the radial head, olecranon or medial epicondyle; and the presence of bruising. Validation if this adult rule is of course needed.

But it doesn't work for little ones; the performance of the adult rule for children under 16 was inadequate (sensitivity 78%), and the authors were unable to derive a dedicated paediatric rule with adequate diagnostic performance.

Ill or not ill, in Botswana

In a perfect world, resources would match need, and triage would have no role. But we live here, where triage is essential and its importance is increased where resource-need mismatch is greater. Mullan and colleagues in Botswana look at the change in over- and under-triage rates associated with implementation of a modified version of the South African Triage Scale. A pre- and post- analysis suggests that implementation of the new system reduced the overtriage rate from 53% to 38%, and the undertriage rate from 47% to 16%.

Keep in touch

Scott, et al, offer a snapshot of the cohort of frequent callers to the Yorkshire Ambulance Service. An astonishing 7808 calls were made in one year by the top 100 callers— on average, that's 20 calls a day from this group alone. International readers may be struck by the expanded services offered by the call centre, including, in addition to emergency (999) response, non-emergency patient transport to hospital appointments, and some out-of-hours GP service coverage.

What's in your pocket?

Although the paperless hospital remains a pipe dream for most of us, the white coat with bulging pockets has become a rarity, as sleek handheld devices have replaced chubby handbooks. But choosing from among the thousands of medical applications now available can be as much of a challenge as sifting through an old-fashioned textbook index. Launching our new Top Ten feature, Michelle Lin, Editor-in-chief of Academic Life in EM, and colleagues, round up the best mobile apps for use in the clinical setting, providing a brief description of what each app does, a quick explanation of why it matters and what the app offers over competitors.

Weekend warriors

In a 10-year registry review from a major trauma centre in Sydney, Dinh, et al analyse the records of 2380 patients admitted after assault. Twelve per cent of these patients were clinically intoxicated, and this cohort was 3 times more likely than sober patients to have major trauma. Over half of patients admitted for assault presented between 20:00 Friday and 08:00 Monday.

Bedside echocardiography in Malaysia

Portable ultrasound is rapidly becoming a standard part of emergency practice, but there is very limited research on what kind of training is needed to use it well. Dr Bustam and colleagues look at the ability of nine first and second-year post-graduate emergency medicine trainees in Malaysia to perform point-of-care echocardiography after a web-based training module and 3 hours of supervised scanning. Among 100 scans, agreement between the trainees and a cardiologist was high for both visual estimation of LV function and ejection fraction, and for the detection of pericardial effusion, but the sensitivity of trainee exam for effusion was only 60%.

What makes a complex patient complex?

We all know that older, sicker patients tend to take more of our time, but exactly which factors impact length of stay is less clear. In a multisite study of three hospitals in the Paris area, Casolino et al collected data on age, gender, triage acuity, testing, interventions, and final disposition, to construct an aggregate clinical complexity score comprised of acuity and resource consumption. An innovative aspect of the study is its use of a combined metric of ED plus observation unit (EDOU) length of stay, which is impacted here by factors that have no effect on simple ED length of stay.

Troponin elevation in marathoners

Baker and colleagues report on cardiac biomarkers in a cohort of runners completing the 2009 London marathon, including runners with and without structural heart disease (SHD) and those who collapsed during the race. Post-race troponin levels were increased in all groups, but there was no significant difference between the SHD and non-SHD groups. While there was a trend towards an increase in high sensitivity troponin T levels in collapsed athletes, no participants had clinical sequelae on follow up, supporting the view, the authors argue, that troponin elevation is physiologic in the setting of strenuous exercise.

Reader’s choice

In this issue, we introduce the Reader's choice, a paper that generated the most hits on our website among those appearing in this month's issue. This month’s reader’s selection is “The Impact of Social Media on a Major International Emergency Medicine Conference.” See what's got everyone "a-twitter".