As some of you may know, ambulance 999 calls are triaged using a package called AMPDS — Advance Medical Priority Dispatch System. This is where we get all those stupid questions from. AMPDS is a very good thing in principle — without it, a patient with a broken leg would get an ambulance before a baby who had stopped breathing, if they called in five seconds earlier — but there are a couple of things that I (and others share this opinion) don’t like about it.

The biggest problem is the question which finds its way into almost every protocol: “Is he/she breathing normally?” This is a silly question, because when people are sick, in pain, or scared. It doesn’t mean that their breathing is dangerous. In my opinion, the question should be changed to “Is he/she having difficulty breathing?” but apparently that was judged as too medical/subjective for some callers. Because of this question, a lot of calls get a higher priority than they should — good news for those calls, not so good news for patients with more serious ailments who are breathing normally!

Another question which bothers me is “Is he/she completely awake?” This question used to be “Is he/she alert?” but again, this was rejected as “too medical”. The trouble here is that after midnight, very few people are “completely awake”, regardless of their state of health. I, for one, am rarely “completely awake” on nightshifts. Again, this question leads to inappropriately fast responses.

Finally, I would like to see a “Fevers” card added to AMPDS, which would effectively weed out the meningitises from the influenzas. Right now, fevers are handled on the “sick person” card which usually results in patients with flu getting a category A response because, guess what, they aren’t breathing normally. I think old people with flu should get an automatic amber response, because they can deteriorate rapidly, but young, healthy people should not be sent an ambulance for flu!

I would be very interested to see what happened if someone did a little experiment: as well as triaging each call by AMPDS, the call takers should be asked whether they think the call deserved a red, amber or green response. Then, when the crew assesses the patient they should be asked which response level the call really deserved. I have a very strong feeling that the call takers’ assessments would be a lot closer to the crews’ than AMPDS’s would be.

Of course, the reason we don’t do it that way is because this way, if a call gets the wrong category and the patient sues, AMPDS’s makers are responsible, not us. If call takers had the responsibility, we’d be liable in the event of a mistake. It’s a shame that fear of being sued makes us rely on a system that is less effective than our own intuition.

Published Feb 19, 2006 - 37 Comments and counting

37 Comments on “AMPDS”
  1. SWbod Says:

    AMPDS is a farce, seeing as i’ve been to a green2 that was actually suspended and a red1 ‘life status questionable’ to a pissed bloke on a train platform!!
    As for a ‘fevers’ card, can it also include d+v? im so totally fed up of going to healthy adults that think 24hrs of d+v is a medical emergency (and thankyou to the pt that gave the norovirus to me, 2days off sick and the possibility of an IAP) :-(

  2. quixote Says:

    How about a system in which call takers could note what their assessment would have been. After some period of evaluation, if, say, Mark Myers, polled better than AMPDS more than 60% of the time, or whatever the cut-off might be, then Mark would be given the option to override AMPDS at his discretion (and without losing the current legal umbrella). The system could be set up so that whenever a dispatcher acquired that level of experience, even if it took some people twenty years, they could get override abilities when it looked like they could use them.

    (I know. Totally utopian. It’ll never happen. But it’s fun to imagine solutions anyway.)

  3. Mark Myers Says:

    The “unknown problem” card is a load of rubbish as it gives a Red 1 if there is any evidence to suggest that the patient is dead — even if the caller has verified that the patient definitely isn’t dead. I try to avoid that one when I can. Yes, D+V should definitely go on the “Fevers and Other Common Illness That Really Aren’t Emergencies You Big Girls’ Blouses” card. I think some people think it is their divine right to never get ill and demand a nee naw on lights and sirens to put everything right instantly — they must be really shocked when you don’t turn up with a magic wand…

  4. Neil Says:

    Has anyone done a scientific study of how answers to questions correlate with the actual level of urgency a case requires?

    Or are the questions devised on an ad-hoc basis?

  5. M2KB Says:

    I fail to see what was wrong with our Ambulance Services’ old system. They recently changed from something quite reasonable, to AMPDS. The old system relied on the knowledge of their staff to ask appropriate questions given the call, and categorise it accordingly. Worked well. In fact, it was mostly the same as the system the police use. We have no set fields other than Phone, Location, Origin (caller), Front Page Text, Title and Grade. The rest is ‘free text’/comments.

    Depending on what the caller tells us and what we ask them in return, results in an appropriate grade for the call (most of the time). While it would be hard for the police to come up with a system like AMPDS, I never like systems that take the ‘control’ element away from the human operator. Despite what management, programmers and policy-makers may think, computers are NOT better at the job than the people who operate it. That goes for police, ambulance, and any other service that uses a computerised command and control system.

  6. Mark Myers Says:

    Well, that’s an interesting question. AMPDS is developed by a big and important organisation in the US — here is their website and used by ambulance services all over the world, which leads me to think that surely it must have been tested extensively and found to be reliable and valid… but then if this were the case, then surely they’d have found and fixed the problems I mentioned above.

    I’ve searched the internet and can’t find any studies which investigate its triage accuracy, either. There’s plenty about the life support instructions (which, incidentally, I can’t fault) but nothing about the initial triaging… if anyone knows of any, please let me know!

  7. Mark Myers Says:

    M2KB — that’s really interesting. I assumed that the police would be using an equivalent system. How do you come up with priority levels for your calls?

    I think the way the police call taking system works is often a lot more sensible than ours. If you make an inappropriate 999 call to the police, the call taker says “You shouldn’t be calling 999 for that, do this instead”. If someone calls 999 inappropriately for an ambulance, we have to begrudgingly send one, they wait ages, and then find that the ambulance can’t actually cure their flu/diarrhoea/spots and that their time was wasted. I doubt the police have ever caused anyone’s death by telling them their call was inappropriate, so I don’t know why we can’t do the same.

  8. M2KB Says:

    Mark – The police system is essentially based on common sense, interspersed with a National Policy. That link is a Sussex Police version of the policy (all I could find online and publically accessible!), but they’re all the same when it comes to the meat of it.

    Essentially, to cut a 10 page document down to one paragraph, our grading system is thus: If there is any immediate threat to life or property, it’s graded a 1 (and a few other bits and bobs). If the job is serious but has already happened, it’s graded 2. If it’s not particularly serious and has already happened, it’s graded 3. Non-attendance/advice is grade 4.

    Grade 1 = 15 minutes or less
    Grade 2 = 1 hour or less (that’s the target, anyway…)
    Grade 3 = Whenever
    Grade 4 = Never

    Our system does not force a grade on you. The danger of course is someone can misunderstand the situation (red mist/tunnel vision) and grade something that should really be a 1, a 2, or vice versa. But the dispatcher will sort that out when they receive the job from the calltaker. It’s all about supporting each other.

    As for sending an ambulance to anything, I disagree with that. As you say, to my knowledge we’ve never killed anyone by telling them that they should really speak to the care home staff to change the light bulb in their bedside lamp rather than us. If we’re feeling really nice and arn’t busy, we’ll even make the call for them…

  9. Mark Myers Says:

    That seems to make a lot of sense. My only worry about dispatchers grading the calls themselves would be that they would let the caller’s behaviour influence the gradings eg. nice little old ladies being upgraded, horrible rude fuckwits being downgraded, but then that happens to an extent with AMPDS (the way you intone a question, or rephrase it, can make all the difference as to how it is answered) and as you say, dispatch will sort out any mistakes in call taking.

    We are allowed to say that we can’t help if the caller is asking for something non-medical (like the call I had the other day from a woman whose kitchen pipes had sprung a leak… never heard of plumbers?!) but we have to accept all medical non-emergency calls like spots and splinters. These sit at the bottom of the piles and eventually get rung back by the TAS paramedic which is not what the patient wants anyway. It would be so much better for everyone if we could just say “sorry, no, try X instead…”

  10. M2KB Says:

    Perhaps you should have a “write to your MP” post? If enough people make a point that it’s a rediculous waste of tax-payers money to send an ambulance to a pimple outbreak, policy changes may flow down from The Powers That Be?

    I’d sign it.

  11. Mark Myers Says:

    Well, I’d like to, but I guess I’m not brave enough. I don’t want to look like a troublemaker or like I am criticising the otherwise wonderful organisation that pays my wages. I wasn’t even sure if writing this blog entry was a good idea!

  12. PJ Says:

    Interesting. We’re all EMD trained here too in the States, and our primary question is “Is the patient breathing and concious?” This doesn’t seem to be too “medical” oriented for the Georgia rednecks we deal with. I’ve found protocols go out the window when there’s an emergency. We try to set aside protocols in favor of communication.

  13. Steven Says:

    ” If there is any immediate threat to life or property, it’s graded a 1″

    I love the way those two things are given equivalent priority.

  14. dispatch, eh Says:

    unfortunately, it is a shame that our dispatch priority systems are established equally toward “liability” as “patient condition”… though, you know as well as i, that every calltaker once and awhile is influenced more on split-second intuition than personal accountability;) …even if our respective governments frown upon it, patients can thank us. (oh wait, patients never thank us) nevermind.

  15. Olivia Says:

    Your points about grading are very valid, and you must get so frustrated!! But, regarding d&v calls – sometimes it IS an emergency!! One of my children had a laparoscopic fundoplication (stomach wrap) op when she was a toddler, and until this year any vomiting for her has merited a 999 call as she was at risk of splitting the site of the operation – technically patients who have had a fundoplication operation should be incapable of vomiting. The number of times I’ve had to call an ambulance for her are too numerous to mention, and I am only grateful to the wonderful despatchers who have listened to me over the years, and to the equally understanding ambulance crews to whom she became a regular! But it must have been very frustrating to take a call along the lines of “my daughter has just vomited and needs an ambulance!”

  16. Mark Myers Says:

    … and a call like yours would get a Green 2 lowest priority under AMPDS, which unsurprisingly doesn’t include any questions about laparoscopic fundoplications. Whereas if you explained the situation to me, I’d think it was definitely worthy of an amber response. Another reason why dispatchers’ brains are better than computers. All the examples I gave in the post were of calls being given too high a grading, but there are a few that come out too low at all.

    It also illustrates how careful I have to be about saying things like “NEVER call an ambulance for D+V… ALWAYS call one for chest pain”… there are exceptions to every rule!

  17. Anonymous Says:


    I have been reading your site with much interest for a while.

    I work for NHS re-Direct (no jokes or insults, please!) and we have a similar triaging system for our callers which grades them as 999 (call an ambulance), A&E (go to A&E by own transport), GP (only used when integrated with GP service out of hours), P1 (immediate transfer to nurse), P2 (call back from nurse within 2hrs), P3 (call back from nurse within 3hrs). If we are in any doubt over the priority our clinical assessment system has reached (via similar, sometimes apparently irrelevant questioning), we will call the call centre clinical supervisor (usually a v experienced nurse with time spent in A&E) to discuss what the patient has said, what answers the patient has given to priority triggering questions (i.e. answering yes to breathing difficulties when the caller has a blocked nose = 999), and what has made you think this is an incorrect outcome. The clinical supervisor is able to override the priority CAS has reached – everything discussed is recorded on tape and we document the advice from the supervisor in the call record. Overall clinical liability rests with the supervisor – not with the call taker or CAS. Generally, anything that has come out as 999 cannot be downgraded any lower than a P1 – so that the caller will be transferred directly to a nurse.

    Now obviously, re-Direct mostly get calls that require little more advice than “yes, it usually is that colour” or “don’t scratch it then”, so it is not on a par with the calls you would be dealing with every day. If there is any thought that there might be a problem with safety, the supervisor would keep the call at a higher priority or upgrade. This system weeds out the potential 999 calls for a broken nail or a cold. Would a similar set-up be useful at ambo control?? Just wondering what your opinions are??

  18. Mark Myers Says:

    I think a similar set up would be very useful, yes. The only trouble is that we’d only be able to speak to the supervisor at the end of the call (999 calls do not take kindly to be putting on hold!) and by then the ambulance would already be sent. By the time we’d discussed it, then the ambulance could well have turned up on the patient’s doorstep. If we held the call until we’d finished discussing it, then that would stop up meeting those 8 minute targets the government is so keen on.

    But it’s a good idea in principle. I’m just not sure how it would work for us!

  19. Big Al Says:

    I’m a Dispatcher for Westcountry Ambulance, and also a trainer for our triage system – Criteria Based Dispatch. CBD enables us to triage our calls by asking appropriate questions and selecting A, B or C (same as your red, amber and green) dispatch codes manually. This obviously enables us to use skill and judgement, and often gut feeling, rather than having computer software decide what level of response is appropriate. Regarding the wording of your questioning, our two key initial questions are “is the patient conscious and able to talk to you” and “is the question breathing normally”. This assumes that a patient with a reduced conscious level who is unable to communicate is potentially unable to maintain his/her airway, therefore at significantly higher risk of respiratory arrest (yes even those saturday night drunks – classic rock star inhaling vomit death). We ask if the patient is breathing normally because if the patient is breathing in any way abnormally, we can ask the caller to describe the breathing, either to identify agonal resps, or to establish if the patient’s breathing is severely impaired and therefore again at risk of resp arrest. We do this by just asking if the patient is able to speak in a full sentence (keep things simple). Ultimately, the primary aim of CBD questioning is early identification of real/potential cardiac arrest. We’re all judged on our A-Cat 8 minute responses, and this questioning enables us to identify very quickly those A-Cat calls. Of course, there are plenty of calls given A-codes that one crews’ arrival turn out to be less serious, but telephone triage is a one dimensional world and you can never get it 100% right. At least with CBD we are not tied in to a structured inflexible line of questioning. So what’s the future for us? Word from above is that we are going to be forced to adopt AMPDS because Tony and his cronies what a single national triage system!!! What happened to “if it ain’t broke, don’t fix it”? Oh, and the cost for using CBD is significantly cheaper than for AMPDS. Flexible, user-friendly AND cheap. It’s no wonder the DoH want us to drop it!!!

  20. M2KB Says:

    Thank you! CBD is what my area control USED to use before they went to AMPDS.

  21. Big Al Says:

    Thanks M2KB, just as a matter if interest, which Trust are you with, and were you given any explanation for changing from CBD to AMPDS?
    The more I think about the prospect of changing triage systems, the more frustrated I get. I suppose a system that relies on “yes/no” answers and no degree of inteligence could be used by less skilled people, therefore could be banded lower in the dreaded A4C and save a significant amount of money!!!

  22. Mark Myers Says:

    CBD actually sounds much better than AMPDS to me. More room for user error, but far less system errors. I suppose the ambulance services would rather have the system making lots of mistakes than people making just one.

    And don’t mention stinking rotten Agenda For Change! It is all people have been talking about in the control room for at least a month and I am sick to death of it and quite sure that I won’t make a penny!

  23. M2KB Says:

    Big Al – I’m not with any Trust (in a control capacity anyway, although I am a trainee first responder). I work for the rozzers.

  24. BigAl Says:

    Only talking A4C for a month!!!!
    We’ve had nothing but A4C for 18 months. We’ve just had our appeals.
    Call Takers now Band 4, Dispatchers Band 5.
    Not bad money for a bunch of pasty munchers!!!

  25. Mark Myers Says:

    We have been talking about it since I joined, but for the last month since the bandings were announced no-one has talked about anything else at all. We are still having appeals but at the moment grade 1 and 2s (mostly call taking) are band 3, and grades 3 and 4 (mostly dispatch) are band 4. Let’s hope they are moved up in line with yours!

    I think call takers should get more than dispatchers anyway — it’s much more like hard work. The only difficult thing about dispatch is talking to grumpy midwives.

  26. Louise Says:

    Ahh trying working for Queensland Ambulance in Australia. The place where we can’t say no. If your toenail falls off and you ring for an ambulance (its 000 here) you will get one!!!
    Is it just here or are there a lot of “drink spikings” on your side of the world? I wonder.. which drink was spiked the 3rd or 33rd???
    But my favourites are the ones that come out as code 1′s (lights and sirens fastest response) and we end up at a apparantly dead and dying patient who is sitting on his front steps having a smoke. Gotta love it.
    Anyway, thanks for reading my rant. On a different note, I am planning on moving over in July. What ambulance (that uses AMPDS) would be good to work for? Cheers Louise

  27. Clare Says:

    I work for Two Shires Ambulance (Northants & Bucks) where we are all on band 2 regardless of call taker or dispatcher as all staff do both on a shift rotation (after approx 6mths call taking and dispatch training) I think Dispatching is alot more stressfull as you are constantly working your a**s off, where as call taking can be stressfull when you get an arrest or a drunken idiot but your not constantly stressed for the whole 12hour shift like dispatching (with minimal breaks) i think the guys who decided which band each control room fits into should actually spend a day in the room seeing how hard we all work and what we have to put up with to do the job.would be nice to be able to do an exchange with a fellow EMD to see how another service works.

  28. Darren Says:

    Incase any other EMD’s with AMPDS stumble upon this, for some interesting reading:

    I did a journal search and found only about 6 studies dealing with MPDS. The abstract and conclusion for each is above. I’ll let you make your own opinion.

    If anyone using AMPDS is interesting in trying to inject some sanity into the whole process, drop me an email. We can compare compliance and responses.

    It is time for the end user (the EMD) to be heard.

  29. Diagnosis? N.F.I. » Blog Archive » The Last Post (Revisited) Says:

    [...] So, for the last two weeks I’ve been running around as an extra ‘resource’ (I am not a resource, I’m a free man!!) while we all get used to the new AMPDS – Advanced Medical Priority Dispatch System (blimey an acronym with more than three letters – must be good. Actually it’s sh*t according to everyone I’ve spoken to in Control). They use it in London – Nee Naw posted an article about it. [...]

  30. Cllr Graham Smith Says:

    Mark Myers wrote:
    It’s a shame that fear of being sued makes us rely on a system that is less effective than our own intuition.

    Sorry to comment so late, but I have just been reading “The Wheel Reinvented – an introduction to MPDS version eleven” by Brett Patterson (NAOED Council of Standards Reader) and Curriculum Board Editor Robert Martin (Executive Director), which seems to say that the EMD should NOT rely entirely upon the MPDS:when assigning the determinent code:

    The MPDS is not meant to replace a thinking calltaker and cannot reasonably be expected to predict outcomes or assure EMD compliance in all cases. EMDs always have the option of overriding a recommended choice for patient safety, or of reconfiguring a determinant code based on new information. There are also times when the EMD may have to choose the most appropriate telephone treatment options from several possibilities. These are issues for consideration on a case-by-case basis, through local continuing dispatch education, quality improvement, and medical control.

  31. Mark Myers Says:

    That’s interesting – I didn’t know it said that and I’m going to have to read that article in full when I have a moment. At the LAS we’re not allowed to override just because we think we ought to, although we do have protocols which let us choose a higher determinant than the one AMPDS comes out with for broken bones and sickle cell crisises. I might show that article to QA and see what they say. Thanks :)

  32. Howard Says:

    I am sorry to but in but——– read the sequence of dialogue above and see the build up of selfish aggression and so-called expertise, Professionalism is paramount and solutions the priority. As a teacher I professionally address my responsibilities despite the rhetoric and paperwork. For this I expose myself to the real world and the consequences of failing ( no matter how minor). Is the solution in the paperwork or experience. Where ….. is your head in place… apologies!. As a Christian I seem to have a dilema too, but as one of God’s Saints I have no problem…… get it right and get the paperwork to document your plight. Priorities discuss the issues at the top of this page not the bottom. The makers of the Law, made the law for their image not the the truth, nor anything like God desire. Pebbles move stone and not mountains. Mountains need earthquakes but generally last for ever as we see them but erosion is slow and effective. Stand out as erosive forces and be patient! I had a classic non-Hodgkins lymphoma and it took 4 months to diagnose, instead of weeks and eventually a dentsit pointed me to a doctor who saved my life 30 years ago. I knew I was unwell and not the experts. They asked the wrong questions and read the wrong signs. Ativan is not the cure ( that’s what my doctor prescribed). The solution for you all is dialogue, where someone listens. I listened! I just happen to be the wrong guy.

  33. Karl Says:

    Thank God Howard was not my Teacher, no wonder we have so many screwed up kids in society, well i like the ampds system and if any of you guys have ever heard of POCSAG its a pager protocol and i spend many hours decodeing my local ambulance despatches, to see whats going on, and linked with our wondeful gps systems can usually get right to the action before the ambulance in most cases, what a wonderful system, Happy New Year to you all, prehaps the security of such issues should be looked into especially with the introduction of airwave imminant, totaly encrypted comms, would like to here about your comms experiances as i here many times i will call you on your mobile as you are unreadable, and as we all no cell phone comms does not work with 100% coverage,

  34. JonG Says:

    Has anyone noticed that AMPDS tells call takers to lay unconscious patients on their back? I am currently noting all these cases and in less than a week have attended one patient who I believe died from airway obstruction as a consequence of bad advice. Watch this space…it is allegedly being “looked into”

  35. bullaz Says:

    well im a call taker for north east amb and we used to use cbd until october, we now use all new *pathways* – which was sold to us by the company as “much better than ampds” and “its not robotic so you wont lose your brain power”. well thats a load of rubbish!! my brains are now fried from the stress and by what i know of ampds it can be miles worse.

    the system has good advantages as it can triage people to pass them to gp’s or advise them to make their own way to a&e(if they are willing to accept those dispositions!). it is basically a similar form of ampds but less structured. the idea is you dont have to ask every question if u have been told the information that answers the question.although now we are a few months in, it turns out officially we never really know the answer unless we ask the question! so if someone rings with a stubbed toe, no blood, talking and breathing fine, we would still have to ask the question to rule out A response(8min) shock…just incase. the other problem i find is the questions are too vauge or too technical for joe bloggs which means the call takers are interpreting the questions themselves to explain them and to get an acurate answer, which in its self means the system has user faults.even the fact that ongoing conditions are not weeded out, and i mean simple things such as asthma, copd etc. these are gp dispositions which we are are meant to upgrade, even though there are no guidelines to what should be upgraded and what shouldn’t.

    although its a new system i dont think anyone feels it is being adapted fast enough compared to the problems we are finding.

    we are on a 6 month trial of the system for the goverment, this is possibly the system that will roll out to the whole country. well all i can say is i wish u all the best of luck!

  36. Burns Says:

    Gotta say i agree exactly on these points. yes i do.

  37. Malcolm Alexander Says:

    Why is it that in 2009, 4500 patients who were classified at Cat A by the LAS got a response after 19 minutes. Can anyone explain this to me?

    Nee Naw
    Nee Naw was a blog about life in the London Ambulance Service control room. It was written by Suzi Brent from 2005 to 2010. The blog is no longer being updated, but the archives will remain here.
    Buy My Book!

    Latest Tweets

    Recent Comments
    August 2018
    M T W T F S S
    « Aug