The NHS Pathways assessment software is risk averse and sends ambulances needlessly.

The software is risk averse, but not needlessly so. Here's why:

While the vast majority of 111 calls are from individuals with minor ailments or other non-urgent needs, on occasion 111 callers are not always aware of - or prepared to accept - that they may be dealing with a life-threatening condition. As a result, it is not uncommon that people call 111 when they should be dialling 999 or attending an emergency treatment centre. Ideally, every member of public would be able to identify the onset of a stroke, heart attack, severe breathing problem or other potentially major problem, but ambulance staff of all people will know that is simply not the case. As such, Pathways is designed to rule out the presence of any 'red flag' symptoms as quickly as possible so the caller can move on to a more symptom-specific assessment.

It's fair to say that over the phone, this is a potential minefield. The inability to see the patient means that the call-taker is reliant on the information given by the caller. There are a multitude of reasons why this is sub-optimal, and that is the key limitation of telephone triage. While there is of course room for improvement, there is no way to make such a system fool-proof. No call-taker, clinical or not, is going to be able to guarantee they can be 100% accurate in separating indigestion from infarction, stroke from Bell's palsy, or hypoxia from hyperventilation. Not without a physical examination to rule things out.

So sometimes an ambulance gets sent when - even though its recognised that the worst case scenario is unlikely - it would be unprofessional, unethical and dangerous to do otherwise. We cannot not diagnose over the phone, even if every call was dealt with by a clinician.

2. Call handlers are prompted to ask ridiculous and irrelevant questions.

Some of the questions call handlers are prompted to ask during the assessment process can seem inappropriate or unrelated to the presenting problem, such as asking the caller who has been speaking freely during the initial conversation if they are fighting desperately for every breath, or having to ask the mother of a feverish baby if the child has been to a West African country affected by the Ebola outbreak in the last 4 weeks.

However, even though in the vast majority of cases the answer would seem to be an obvious no, imagine the outcry in the rare cases where those factors were in play but no attempt to identify them was made. For the record, both the above-mentioned breathing question and a further question regarding skin temperature are intended to catch signs of sepsis.