Correspondence (letter to the editor): Inadequate Statistical Evaluation

The article „Patients Attending Emergency Departments“ is especially impressive for the fact that the authors grossly violate a viable study design in the selection of study participants: out of 6488 patients, 3396 patients are excluded in advance—in other words, almost half. Strikingly, the proportion of those who were very sick—that is, who were treated immediately or transferred directly to another department—is very high, at 1842. Of the remaining 3086 participants, 1043 refused to participate—which is almost every third patient, and 708 were treated before they could be interviewed (1).

In the end, only 1175 patients remained. Even in this group of less than 18% of the total patients, there were still cases that turned out to be urgent.

Nevertheless, a key message from PiNo Nord (“Patienten in der Notaufnahme von norddeutschen Kliniken“ [patients in the emergency departments of hospitals in Northern Germany]) is that more than half of the cases of emergency department patients do not meet the definition of an emergency.

However, this cannot be concluded from the data presented. Apart from the inadequate statistical evaluation, this examination does not pose the main question in emergency treatment: Can we use emergency departments to select those patients who are very sick, possibly with life-threatening illnesses?

Every doctor who handles emergencies knows that subjective distress and objective urgency are not always congruent. Patients presenting with subjective distress do sometimes have trivial conditions—and sometimes the opposite is true, for example, feeling unwell before a heart attack.

Distinguishing between these two possibilities is the job of a specialist, and this has a price. It is the task of the health insurance funds to finance this. If this is not possible from the paid fees, the health insurers and politicians must communicate and regulate this.