Tuesday, 29 May 2007

The protocol paradox

I felt an urge to follow up on the excellent Dr Grumble's musings on medical protocols, as throughout my practice as a doctor I have encountered numerous problems with these brain dead devices.

These days there are so many protocols available that one has to be very knowledgeable in order to decide which protocol to use, however if you are that clever that it is likely that one wouldn't need the protocols in the first place. Hence protocols are useless, QED. This my paradox of the protocol, and it means that if someone relies on protocols to manage patients; then are they really expert enough to be independently managing patients?

It can also be argued that if a protocol leaves room for manoeuvre then an individuals clinical acumen is the important factor in the decision making process, hence is the protocol of any use? However if the protocol is very rigid and allows no room for flexibility, then the protocol is likely to be doomed to failure as it lacks any common sense.

These generalisations are probably a little harsh on protocols as they do have their limited uses. However there is no denying that a over reliance on protocols can be a very dangerous way to practice. There is no substitute for good thorough medical training and a broad experience, and 'medicine by numbers' will never come close to replacing this.

NICE have produced their fair share of dodgy protocols and their head injury guidelines are but one example. According to their guidelines anyone who has a GCS of 14 or below at 2 hours after their head injury merits an urgent CT scan. If one obeys the protocol to the letter then anyone who is confused at two hours after their head injury needs an urgent scan. Obviously under certain circumstances this will be sensible, however should this apply to the demented elderly patient or the disoerientated drunk? Of course not. The NICE alogorith also reckons that anyone who is 65 years old or above who sustains any 'loss of consciousness' post head injury merits a CT scan, this is another ludicrously stupid rule.

The more clinical experience I have gained, the more I have learnt to take the dubious generalisations offered by protocols with a very large pinch of salt. These protocols are often hatched in secret by small committees who make no effort to engage with the majority of clinical opinion. Beware of health care professionals who rely on protocols.


Anonymous said...

So how do you know that the demented elderly patient or the disorientated drunk doesn't also have a serious problem from the head injury?

Marcin said...

You've got a bigger problem: each protocol is developed in isolation, so even if it could be applied sensibly, if it interacts with another protocol, you have the potential for an infinite loop, or other bad outcome.

Garth Marenghi said...


it's down to one's overall clinical judgement and skill which is a far better guide than a protocol- this is the point of the piece!

each case has to be managed according to its individual details and you can never 'know' for sure

this is the art of medicine, if you really want to understand then it may involve a few years at university, a few more learning your trade and even then even the best make mistakes

(it's always the doctor's fault though isn't it, when a mistake is made; it could never be that a doctor acted as any reasonable doctor would and that medicine is just a tricky business)

Garth Marenghi said...

ps marcin, good point, it's pretty common that more than one protocol applies and then even bigger problems start.

Anonymous said...

If it's down to one's overall clinical judgment and skill, how can "Of course not" (as per your post) possibly be a good answer?

Actually, I agree (on the whole) with your views about protocols; I was just carrying out a small experiment to test your devotion to the merits of polite and reasoned debate, as per your strictures on Spaghetti Harvest.

You seem to me, on the whole, to have reacted rather more defensively than she did. What can your last para in brackets possibly have to do with my question?

Leaving aside the defensiveness though, I also agree that doctors - indeed all professionals - make mistakes sometimes, and that mistakes are not in themselves any evidence of negligence or fault. At the age of 44, as a former professional myself, and with 11 years as a cancer patient, I do understand a fair amount about professional decision-making and even medical decision-making.

Garth Marenghi said...

unfortunately I cannot explain every statement, as if I explain that then that would would then require explaining and on and on.

I accpet your point, the use of 'of course not' is obvious to anyone with medical training.

I should have made it more clear for the non medical amongst us.

The NICE guidelines mean that an old lady for example who is normally confused merits an urgent CT scan if she comes in after the most trivial of injuries, even if she is exactly her normal self.

This is the point I assumed would be obvious, apologies.

Anonymous said...

I knew what you meant: it's just not what you said.

I just replied to you over on Spaghetti Harvest, but her spam filter ate it, and I kind of lost interest. I was just rather sorry for her for being ticked off on her own blog for a crime that, IMHO, she hadn't committed. I didn't entirely understand what you meant about insults and politically correct (has anyone insulted me?) but anyhow, if you want to insult me, please feel entirely free. I'm quite aware that I often come across as a didactic cow.

I think you write a good blog. I pretty often read it, but if you like, I will revert to my previous habit of not commenting!

Garth Marenghi said...

All in all I think a lot of it comes down to the typed word not coming across as it was intended to by the author.

I am guily of sounding like an obnoxious git without intending to on a frequent basis, every comeback ends up sounding so unintentionally rude.

The insults were not coming from you, they were coming from other individuals. Anyway its all water off a duck's back anyway.

Please keep commenting, I will try not to be as defensive in the future, it is quite instinctive with the typed word though, it's hard to get across one's argument without sounding like an aggressive tosspot.


Dr Grumble said...

So how do you know that the demented elderly patient or the disorientated drunk doesn't also have a serious problem from the head injury?
The drunk who is dismissed as such is, of course, a classic mistake. But you are more likely to make a mistake if you just blindly follow the protocol instead of thinking - the sort of thinking that Potentilla did. Just firing off tests according to a head injury protocol is not the answer; the problem could be nothing to do with the head injury or the apparent drunkenness.

Anonymous said...


It occurred to me today that, although protocols ("procedures") are widely used in my own industry, they are used in two circumstances. One is where a junior person is doing something which could cause really spectacular damage if it were done wrong (eg move millions of pounds from one place to another). The other is in the more-or-less mechanical implementation of a professional decision which has already been taken (eg a fund manager decides that all funds of x type should go y% overweight in Taiwanese equities). But we don't have protocols for professionals making professional decisions.

Anyhow, what I really stopped by to say was, Garth re your last comment above, you might consider using emoticons. :-) sometimes. This doubtless sounds like didactic-cowism too, but in fact it's the enthusiam of a convert; I was doing an online philosophy course a year or so ago, and found smileys and so on actually did help to avert a few misunderstandings about whether (for instance) a sentence was intended sarcastically or not.

Garth Marenghi said...

I am an emotionless sociopath :(


dreamingspire said...

A personal experience: when recently I received a phone call to say that my ancient mother had collapsed and was just about to be carted off to hospital from the excellent care home, I took the trouble to phone the hospital. They in turn quickly put me through to the right nurse station even though Mum had not yet arrived (the nurse knew about the pending arrival), and therefore I was able to explain her dementia and thus avoid her suffering from protocol blight. When she arrived she was quickly and correctly diagnosed (it was a new problem) – and shipped off back to the care home within 10 hours, because they can cope now that they know what the problem is. The care home and the paramedics did the right things from the symptoms – their worst fears sent Mum to the right place, the doctors listened to my report to the nurse and used their brains to quickly show that it was serious but not what was first suspected. Let’s keep on supporting excellent standards of care.

Anonymous said...

Now are we talking about protocols, guidelines or pathways ? because they are all different.

The head injury algorithm by NICE is a guideline, s you say, not a protocol.
NICE state categorically that 'guidance DOES NOT override the individual responsibilty of health professionals to make decisions appropriate to the circumstances of the individual patient'.

My own experience since the NICE guidelines for head injuries was introduced is that A&E clinicians have far fewer arguments with the neuro-radiologist when it comes to getting a patient into the C/T scanner at 3 o'clock in the morning.

What about significant another group who are subject to a pathway, those suffering an MI ?
Well, the MI 'pathway' has clearly highlighed the yawning chasm between what some clinicians think happens and what actually does happen - the net effect since this pathway was introduced has been a huge reduction in pain to thrombolysis or angio-time.

Another example of a pathway improving care is the so called 'fractured neck of femur pathway'.
This proforma soon exposed avoidable delays in obtaining patients adequate analgesia, x/ray, and ortho referral.

We sometime hear that medics need a little bit of artistic lisence, so to speak, to contemplate the less common differential - and that absolutely fine as long as they don't lose sight of the red flags.

Certainly from an A&E perspective I can think of numerous occassions when a bread and butter diagnosis has been missed, or an avoidable complication overlooked - to a certain extent this why guidelines were introduced in the first place.

Our hospital is about to introduce a thrombolysis for CVAs - do you think patients will receive better care if we leave it to each individual doc to decide what to do, or by providing a pathway stipluating, exclusion criteria [which is extensive], how soon the stroke SpR should attend, how soon a C/T should be done, tPA regime, etc, etc ?

Garth Marenghi said...

indeed, when they are used as 'guides' not as absolutes, that's what I'm trying to get across.

a lot of the credit you give to 'guidelines' is probably attributable to other improvements that have come with the guidelines, such as raised awareness of door to needle time etc and education in key areas.

the 'fractured neck of femur' pathway doesn't stop people hanging around waiting a week for their operation due to lack of theatre space; hence there are more efective ways of improving the service

likewise with PCI for MI I know of some large tertiary referral units that provide a woeful service in this regard

for example again with the CVA thrombolysis, the best way is the multiprongeg approach that involves making everyone aware of when it is appropriate to thrombolyse, it's not the guidelines on its own that does this

in these days of less hours and experience guidelines can be of some use, however good training with a good level of overall clinical exposure is far more important

as this seems to be the case with NICE, they are producing endless amounts of paperwork, so much that no one can hope to keep up to date with it all, it would be better to produce small amounts of stuff that people actually read

what next a NICE guidline for management of the acutely unwell patient? FFS

Anonymous said...


I'm no lover of paper work but the fact is A&E has relied on protocols for many years, any advanced life support provider will tell you that.
Imagine trying to run a cardiac arrest without the universal VF/EMD-asystole Resus council protocol.

I think your comment 'no one can hope to keep up to date with it all' is an almost inevitable consequence of specialisation, hence the old addage, never ask an orthopod to interpret an ECG.

Doctors who are very inexperienced, or, who find themselves managing a clinical problem outside their day to day sphere of expertise can have a quick glance at the guidelines to make sure they have not overlooked a potentially important detail.

Guidelines are merely a tool but they can also be an important safeguard especially if a doctor is tired, stressed, or pissed off - and after MTAS who can blame them.

Don't forget we have a new set of FY2s every 4 months, not to mention the specialty house officers, SHOs and SpRs for surgery, medicine, plastics, orthopaedics, paediatrics........well you get the point - what are the chances of that lot all singing from the same hymn sheet ?

You may well be proved right about the proliferation issue - but I still maintain that ALS and MI protocols have undoubtably saved lives.

While the likes of the CVA, DVT and asthma proforma [the last one based on British Thoracic guidelines, of course] are at least useful aide memoirs.

The Shrink said...

My induction at my current hospital was scant. Well, being honest, there wasn't one.

This, I feel, is fantastic.

With plausable deniablity I can now raise my hand and confess that I have not been instructed on any protocol.

Of course, in practice I do work in a fashion consistant with many protocols much of the time, my annual appraisal shows good/better outcomes than peers, so everyone is happy.

Protocols apply to one problem. they do not reflect comorbid pathology, concommitant psychosocial morbidity, contextual factors and specific medical factors ("But doc, last time I had drug X it cured me in 3 days, drug Y makes me puke and takes weeks!") and support the patient has.

I really value the latitude to do what seems right for the patient in front of me, with the complex business of being a person rather than a pathology to treat.

But then, I work in the messy world of mental health when physical health, social adversity, medical intervention, carer stress, practical support available and multi agency morking all influence care planning, so protocols are tricky animals to grapple with.

Garth Marenghi said...

indeed, there are places for protocols, however the key is knowing when to apply certain bits of a certain protocol

without good training, experience and good old common sense the protocols are only as good as their master....