Article Text


Primary survey

Statistics from

Thinking, not doing (for once)

It’s good, once in a while, to take a moment to explore the philosophical issues that often lurk unseen in our rather frenetic specialty. Rick Body and Bernard Föex provide the basis for many such moments in their thoughtful piece on diagnosis and utilitarianism. Starting with a description of our positivist approach to diagnosis they argue that it is an untenable proposition. Passing by Oslers’ ageless wisdom they touch on Baysean concepts and discuss whether “doing no harm” is possible in a world of false positives and false negatives. Then they describe utilitarianism and explore its place in modern medicine. If you didn’t know that the Greeks had three different terms for knowledge and if you want to indulge in a bit of thinking and reflection then read this excellent article (see page 238).

3 minute predictor

In an interesting preliminary paper Andy Pan and colleagues, from Ottawa, report on the predictive value of a simple 3 minute walk test as a predictor for the need for admission in patients with dyspnoea or stable chest pain. In the small numbers of patients studied it certainly seems that this simple test had some predictive value and that it deserves more investigation. A multicentre study is planned (see page 278).


Prediction is a bit of a theme this month with two other papers seeking to see if the future can be predicted. In the first Susanna Au-Yeung and colleagues from the Departments of Computing and Mathematics at Imperial College London use historical data to build a predictive model for walk-in and ambulance attenders. They find that their model has more success in the former than the latter and they discuss possible reasons for this. They conclude that their predictions for walk-in patients can have value for emergency department managers seeking the holy grail of matching resources with demand in an efficient way. Read the paper and decide for yourself (see page 241). Our second predictive paper is clinical rather than operational. Christine Vorwerk and co-workers from Leicester explore the efficacy of a number of physiological scores and measures at predicting 28-day mortality in patients with sepsis. They find that the abbreviated Mortality in Emergency Department Sepsis score is the best performer and go on to suggest that it may aid clinical decision making in sepsis—see whether you agree (see page 254).


Staying with a mathematical theme we also have a couple of papers this month that look at calculation. Marianne Toftegaard and colleagues from Denmark seek to explore whether a mathematical treatment of venous blood gas results in better correlation with arterial sample values. They find that both pH and PCO2 can be calculated precisely but PO2 cannot (see page 268). In another paper Andrew Lindford and co-workers look at the accuracy of healthcare professionals at calculating fluid resuscitation requirements in burnt patients. It seems that the simpler you make it for people the more likely they are to get the calculation right (see page 245).

Patients from care homes

In a paper from Scotland, Lisa Carter and others report the results of their study into the old chestnut of patients brought to emergency departments from care homes (a generic term covering both nursing and social care environments). They looked at consecutive attendees at the Royal Infirmary in Edinburgh arriving over one month. They asked three practicing general practitioners to review the case records and to say whether they thought the patient could have been cared for out of hospital. The wide disparity of opinion among these doctors is, to me, the most interesting finding in this study. These decisions really aren’t either easy or clear cut (see page 259).

Improvised rockets and mass casualties

In a topical paper from Israel, Dagan Schwartz and colleagues give a structured report into a mass casualty incident caused by an improvised rocket. This paper is interesting not only for the description of the incident and the response (and the lessons that can be learnt by us all from that), but also for the structured reporting methodology itself. Major incidents are rare and structured reports, such as this, help planners and responders to maximise the learning (see page 293).

…and finally

Just to remind us all that conditions that we worry about actually do exist, Hoon Lim and co-workers from South Korea report the case of a middle-aged woman who suffered from ricin poisoning. This patient had eaten five castor beans for constipation, not been poisoned by terrorists, and fortunately did well with general supportive therapy (see page 301).

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles