EBME Forums Forums Management Medical Equipment Management Evidence-Based PM Strategies

Point 1) ... not under the "rules" of Evidence-Based PM you couldn't. It only applies to PM (surely all the clues are there)!

Point 2) ... see above! Fault codes (exit codes from Work Orders) are simply not valid as a basis for adjusting PM intervals. That's why George's points about NPF (No Problem Found) - although interesting - have no impact on Evidence-Based PM.

Taken to an extreme:- even if a biomed turns out a thousand times to a "no fault found" situation, that wouldn't indicate that a change in PM interval is on the cards, but rather (obviously, I would have thought) that staff need a bit of help in how to operate the kit! In short, it would indicate a training issue, rather than one of maintenance.

Meanwhile, the "evidence" held on the database should have originated from the Techie, anyway. Where else could it have come from?

What I was after (and still am) from this thread is:- what form does this "evidence" take, and how best should it be presented as a case to adjust PM intervals? My own thought (as I have said a few times already) is a simple interrogation of an agreed number of previous PM records to establish whether or not, in each case, a repair (or similar intervention) was found to be needed during the PM procedure. Up to the last three PM's, say.

I haven't just dreamed this stuff up, but have gleaned it from the progress made over the years by others (usually American biomeds) in developing innovative equipment maintenance models. The informed opinion of thoughtful people, based upon their experience of supporting large inventories of equipment, and a consensus of like minds. If (like them) you seek a better way, you shall invariably find it!

For more ... see here.

Geoff you referred us to the Evidence – the Biseng PPT and sometimes in life the evidence tells a difference story , doing a pm is one way of managing the risk , the evidence tells us what the real risks are ! In this case the earth is not the centre of the universe, from the evidence, manufacturing technology and the current pm program in those hospitals are working! So it not a case of don’t bother or remove the biomed department - it time to move up a notch and take on a new challenge

All of these devices connect to people - no matter how much you dislike managers , regulators and administrators that point will remain a grey area forever . Bunkering down in the basement will keep biomeds in the dark ages – not to mention the biomed department is the ONLY department which can collect evidence related to NPF and other associated device risk. It’s not about policing anybody .

Enthusiastic Amateurs should not be taking off the cover of a device and attempting to fix it – that was never implied. The User Troubleshooting Guide is for the User – When you get 3 to 5 works requests a day with a note – Broken – Fix ! on a pump and its only the tube release lock in the wrong position again ( and these pumps have been in the hospital for about 4 yrs ) it does make you wonder who are the amateurs are .

Or a neurosurgeon who doesn’t understand that in order to get the auto focus to function properly it needs to be within a focal range and the both aiming beams need to be on the same surface otherwise it hunts around and “takes too long”

That’s my take on the evidence

Evidence of poor operator skills, maybe. We're talking about two different things, George.

What I'm talking about is applying (the limited resources available to) maintenance activity where it is most needed through a system of assigning summed Risk factors (risk of failure, primarily from the patients' point of view) to each equipment type. From Risk we arrive at Priority, and PM then gets tackled in that order. In such a system it is acknowledged that some (many, perhaps) low-risk items will never get PM'd at all (as we shall run out of time available - and this is surely a "real world" scenario). Building on that approach, we have the option to adjust PM intervals based upon actual equipment reliability over time (as the years pass by) in the hope that we shall one day arise at the optimum interval in each case. "PM equilibrium", if you like. One measure of reliability is whether or not repairs are indicated at PM. Such "evidence" is then presented to whomever is in charge for them to sanction (authorize) a change of PM interval(s). What I am interested here is:- are other "evidence" metrics available (and valid) and how is such evidence best presented? That's it.

On the other hand you seem to be fixated on poor equipment skills on the part of clinical users. Fair enough, but that's an entirely different ball game, I would have thought, and one best addressed through equipment training of the user, surely. That is, nothing at all do do with PM, Risk-Based or otherwise.

You have mentioned requests for service resulting in (many) "no fault found" situations. We have all seen many of those, no doubt.

But if the Fault Code was NPF ... my own interest there is:- what happened next? What about the Action Code (the action carried out by the biomed)? To me this is the important point.

1) User Advised
2) User Manual Provided
3) User Trained
4) Etc. ...

In other words ... close the loop (and finish the job). That's how properly devised systems should work!

As I see it, the only way that the "multitude of NPF's scenario" (that is, many "no faults found" at requests for service follow-ups) can have impact on our beloved Risk-Based (perhaps more accurately thought of as "priority driven") et al maintenance schemes is:-

If the ham-fisted (or otherwise inept) users persist, through their frustrated (nay, ignorant) efforts to get equipment to do their bidding results in a whole series of trashed equipment ... well, this would mean that a repair would indeed be necessary when we "found" the kit in pieces at the next PM visit. No extension of PM interval there, then.

And also that we might feel the need to add the following line (task) - or something similar - to the corresponding PM procedure(s):-

"User trained" (again)

And lastly, George ... any "real risks" that you deem important can simply be incorporated as one of the Risk Factors which form the basis of the Risk-Based PM Model. We are engineers and technicians ... and I see nothing wrong in applying engineering methods (that is, ones based on objectivity) to overcomes the problems we are presented with. Subjectivity (and appeals to emotion) are best left to the Caring Professions, in my opinion. We all have a part to play. It's a team effort, after all.

Geoff - Limited Resources is the common ground I guess – Yes I am fixated on User Errors , that is largest risk with respect to patient safety in my opinion.

The Action Code : get NPF accepted as a reportable metric just like “falls” are for example , get educators to focus repeat in-service training on those issues ( most of which can be addressed by referring to the troubleshooting guide) , remeasure NPF ( see the NAMDET link for closed loop diagram )

Yes , this will have no impact the pm frequency of a device . ( but it might make biomeds more assertive )

Thanks for the link, George. There's some thoughtful stuff in there.

Yes, the CMS proclamation may have muddied the waters somewhat (but, if nothing else, it has stimulated debate), but if Evidence-Based Maintenance is a thing of the past (which, of course, it isn't - that was just a catchy headline), it can only be because it's been renamed as we move on to Higher Ground.

It's an evolutionary process, after all. And made available to us, in large measure, by that great tool - the Personal Computer (with databases, and spreadsheets, getting honourable mentions)!

PM* -> Reliability-Centred -> Risk-Based -> Evidence-Based ... etc., etc ... and (in reality) a combination of all of them (and others).

* Calendar-Based, "Sweep", "hours (miles) run" ... and all the rest!

But - and more to the point (and as I may have mentioned before) - I know of places in the UK that haven't actually got around to even the most basic PM programme yet!

A long time ago my trust subscribed to ECRI and in return were provided with a CD, which apparently had large amounts of data, regarding reliability of medical devices. We are talking nearly ten years ago, not long in our terms, but long enough for technology to change, do you know if this data is now obtainable on their website? I have checked what I can, but a login is required to delve deeper, and rather annoyingly they haven't been very forthcoming with info when I have e-mailed them querying the kind of data they may or may not hold.

Once again you will have to subscribe, decide at what level you require and what information you desire.

Having just re-read this thread (you know, the way you do), it was nice to recall some of the thought-provoking stuff (to my mind, at least) found in there.

But (as so often is the case) I also found myself wishing that more folk had joined in the discussion.

So if anyone would like to wade in, here are a couple of "primers":-

1) Evidence-Based PM ... what "evidence" are we talking about? What "evidence" would hold up as a valid reason for stretching out (or reducing) PM intervals?

2) "No Fault Found" ... hopefully this should always be the case at PM. But for non-PM call-outs (Work Orders, Post-It Notes etc.):- how do we deal with all those NFF's? Should we record them? And then what action should we take? George has suggested that it's not just a user training issue - I say that it is; if not, what else is it, and how can it be dealt with?

But we need to be careful about "No Fault Found". Whilst it may be the desired outcome at PM (and could even be a trigger - the "evidence" - for adjusting schedules), surely it should prompt a reaction if continually the case at non-PM "visits" (call-outs, Requests for Service, CM, U/PM etc., etc.). I reckon we need two different terminologies (definitions) here:- NFF at PM, and NFF when responding. Let's hear some suggestions!

Meanwhile, for what it's worth, my system still looks back at previous PM's (defaulting to three) and asks the question "was a repair needed in all cases"? And (depending upon the answer) either reduces the PM interval (PM more often) or increases it (less PM). So that was my "evidence" (was a repair required) - but what I was fishing for when I opened this thread were alternative examples of "evidence" (NFF at PM could well be one).

And lastly, if anyone has a New (or Better) way of organising PM and its Mate (CM), let's hear about that. Open a new thread, if you like.

Following the manufacturers resting and regs sounds sensible as long as they are enforced through documentation