Isndie the EPR Arms this week I was rding an intrsing atrilce in a sicintiifc juronal: Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.
‘Mmmm I thought. I need aonethr pnit’.
‘Whats that you’re reading?’ asked a young woman who had just bought herself a drink and sat down next to me.
‘An interesting piece of work done yonks ago by GE Rawlinson (1) about how we only read the words as a whole and not each individual letter and yet the reader can make sense of it by skimming it. Like a 6th sense. A feel. Don’t I know you – from across the road. Medical records or something?’
A loud groan came from across the room.’Gerrim Off. Useless person (2)’ shouted the crowd gathered around the massive TV screen on the wall. England were playing football in some competition or other in which they will not progress much beyond the initial stages. ‘England are rubbish!’ shouts one very aggressive grumpy person. Oh – it was me! I really must control myself.
‘Yes - My name is Jane and I am a clinical coder’ she said.
‘I summarise a patient’s hospital episode by selecting codes from my ‘Bible’ of clinical codes.’
‘Why? What for?’
‘Well, think about it. Where do you look to find out any information about any patient who has been treated in hospital?’
‘I can read their notes or letter to the GP from the hospital.’
‘Yes that is correct, but what if you want to know how many fractured necks of femur you’ve had this year compared to last. What if you want to ensure the Trust gets sufficient funding to cover the costs of these fractured necks of femur. What if you want to look at your clinical outcomes for patients with a particular operation for a particular condition?’
‘I would look it up on computer’.
‘Correct’ said Jane.
‘You couldn’t hit a barn door with a beach ball you useless Wombat (3)’ came the shout from the TV area of the pub.
‘As much use as a chocolate fireguard you are’ added another as the final whistle went. England were on their way out of the Euros 2016. Beaten by Iceland.
‘And how do you think that the computer knows who had what diagnosis and procedure done to them?’ she asked.
‘Someone inputs the information’
‘Yes! - with clinical codes: a clinical coder. That’s me. – Are you really listening?’
She noticed my eyes had wandered back to the screen like a magnet.
‘OK sorry’ I said making an effort to turn my back on the interview with heros from yesteryear - I remember him when he had hair. ‘So what do you record then and how?’
‘We are really at the mercy of what is recorded in the notes or in the discharge letter by the doctor’ she said.
‘Well that should be simple enough then?’ I optimistically suggested, hoping I could finish this conversation off quickly and treat myself to another pint.
‘Listen, why don’t I get us both a refresh and we can sit down over there and go through it’
‘OK – Vodka Tonic Ice and slice for me please’ she said shattering my hope that she was a half of lager girl but hey-ho.
I shouted the order to Trev who said he’d bring them over. I cleared someone else’s empty crisp packet from the table and using a beer mat, smoothed away some drips on the table as we sat down. They didn’t seem to mind and went to sit elsewhere!
‘So – I get it. You have to summarise a patient’s episode by selecting some codes from a book of wisdom. Pips!’
‘Well – not ‘Pips’ at all actually. It’s a real pain sometimes.’
‘In what way? Thanks Trev’ as the drinks arrived.
‘OK here we go. Firstly I can’t always find the casenotes. And we don’t have an electronic patient record yet. So nothing clinical is recorded on our computer. Problem Number 1- Not enough information to go on.’
‘Check!’ I said ticking an imaginery box as I licked the froth off my glass. ’Lets assume you always always get the casenotes on time. Then what?’
‘Lack of actual detail. For instance, almost every set of notes coming through the office on the General Medical side at the moment is for influenza, pneumonia & exacerbations of COPD... BUT.... is it due to infectious organism? If so is it influenza A with pneumonia (J10.0); is there consolidation present (J18.1); is it exacerbation COPD (J44.1); is it infective exacerbation of COPD (J44.0); is it emphysema (J43.9); is it chest infection (J22.x); is it a LRTI (J22.x).
‘What all that stuff in brackets?’ I said rather bizarely – as if I was reading her words!
‘Those are the ICD codes. J44.0. J43.9. J22’
‘Sorry – please continue’
At that point, a new group of customers came in and ambled straight over to where we were sat, obviously knowing Jane.
‘Hello you lot’ said Jane.
‘Are these your mates?’ I asked Jane. She nodded. ‘Fellow Coders’.
‘Hi – my name is James and I am a Clinical Coder’ offering me his hand.
‘My name is Laura and I am a Clinical Coder’ another handshake.
‘Barbara and I’ve been a Clinical Coder since Adam was a lad’ a grumpy nod - no hand.
‘I’m Adam and I am no longer a lad but I am now a qualified Clinical Coder’.
Well this is going to all get a bit intense I thought. It was like a weekly meeting of the Huddersfield Branch of ‘Coders Anonymous’.
‘So what do we call a group of Clinical Coders then? A gaggle?’ I asked.
‘How about a flock?’ suggests Adam.
‘A clutch?’ said Jane.
‘A conspiracy?’ added James.
‘An Enigma !’ offered Trev who had been hovvering by the table ready to pick up any empties. An Enigma of Coders was the winning entry with a hint of Bletchey Park.
‘So Jane, how are you today? I heard you were feeling a bit rough yesterday?’ asked James.
‘Not great, Woke up with a bit of a J02.9 so I worried it might be the start of J10.1’
‘Worried it might develop into G03.9 or G03.1?’ asked James.
‘I think it’s probably just a J02.8 + B97.8 but can’t rule out a J02.8 + B95.8 can I?’
‘No not at this stage’ replied Laura, ‘and you need to know if antibiotics will do you any good won’t you?’
It felt like I’d walked into a Star Wars convention and R2D2 or C3P0 would walk in any minute.
‘Wish I was retiring tomorrow!’ said Barbara which brought me back to earth.
‘Jane was just explaining to me what you lot do. I never realised so much depended on good timely accurate coding’.
‘Too right!’ said Laura ‘without clinical coding the Trust wouldn’t get its money and clinicians couldn’t do any research or clinical audit.
‘The Trust couldn’t do any planning’ said Adam.
‘….and I was hearing you have a few issues’ I suggested.
‘Issues! Issues! I’ll give you issues’ said Barbara, who seemed to be carrying all the problems of the coding world on her shoulders and not dealing with them with great humour.
‘Like?’ I prodded gently.
‘Well, the latest – I tell you – only just recently – today actually, this morning it were a consultant at our trust said "All pneumonia is bronchopneumonia".... Well, we coders laughed. Hysterically some of us. If only it was that clear cut…. We can't just code all pneumonia as bronchopneumonia unless that is what's documented in the patient’scasenotes. A bit like the Eskimos have over 50 words for snow (4), we have hundreds of different codes for different types of pneumonia (blocks J09-J18 to be precise). If you just search for ‘Pneumonia’ that takes us to a J18.9 (pneumonia, unspecified) which isnt detailed enough for clinical and analytical purposes.’
‘So - You are like detectives (5). Searching through the notes to identify the primary reason for that episode.’
‘Yes and our job is made no easier by weasel words.’ Said Laura.
‘In what way?’
‘Well’ she continued, ‘we get stuff from consultants which we just can't code – ‘likely’, ‘suggestive’, ‘impression of’, likely to be’, ‘could be’, ‘may be’. ‘has a hint of’. More like they are describing a nice crisp chilled Chardonnay ‘with a hint of Apricots’ rather than a clinical episode.
James joined in ’actually most of the diagnoses we get are preceeded by one of these little buggers!! And when they are, we can't code them -they’re practically useless to us. ‘
‘OK’ I say, ‘Check! Problem number 3. Vagueness. Next?’
‘OK well. Issue number 4 is the recording of co-morbidities’
‘‘Yes exactly that. According to our rule book, a co-morbidity is, and I quote, 'any condition which they have along with another disease that is currently being treated at the time the patient is in hospital or develops whilst they are an inpatient and affects the management of the patient's current admission.’ I was impressed that they had memorised the rule book!
‘OK I get that. So why don't you just record their chronic long term conditions then. Pips No. 2’ ‘No – we can’t.’
‘Course you can – go on’ I suggested with a hint of frustration.
‘No we are not allowed to!’ said Laura. ‘But you’ve got the notes. You can see they’ve got other issues. Just bung it in.’ I said ‘Simples’
Apparently not that simple, as James pointed out. ‘We are not supposed to code anything that hasn’t been recorded in the current episode. It is up to the doctors to record any significant clinical information. If they consider this current episode is likely to have been affected by an on-going condition, then they have to record it so. We are not supposed to. It’s in the rules. (See guidelines for coding and reporting here). It’s really frustrating. You can see the patient came in last week and you can see a list as long as your arm of co-morbidities on that episode which they have (diabetes, congestive heart failure, epilepsy, history of CVA....) and yet these underlying conditions have not been recorded as part of the current episode.’
‘And that can affect the Trust’s income’ added Jane. ‘It takes longer to treat a patient with underlying conditions than it does a fit rugby playing he-man adonis-like greek-god like creature with bronzed muscles on muscles…’
‘OK OK’ we get the picture Jane’ said Adam to Jane who had gone off into a strange place!
‘Oh I see.’ I said ‘But surely – the future’s bright with the electronic patient record we’ve been promised. Paperless by 2018 at the last count.’
‘Oh that! We’ve been promised that since – well, since Adam really was a lad!’ said Barbara. ‘What year were you born Adam?’
‘1993 actually! 23 next week’.
‘But surely with the electronic patient record, your job will be a lot easier’. I optimistically suggest (as I have been doing since Adam was starting Primary School in 1998!’)
‘Only if that EPR has some sensible structure and everything is brought together in one easy-to-navigate place’
‘I thought that was the plan?’ I added.
‘May well be the plan but the reality is not quite so slick. I’ve tried coding from an EPR up the road at the Uni hospital. They’ve scanned in their casenotes, so now instead of flicking through the paper, you have to scroll through 50 odd pages of scanned documents on an Electronic Document Management System (EDMS) and half of them are blank and all just to code an episode of just 2 days.’
Laura joined in. ’And it’s not all in one place. You have to log onto three or four different systems: The PAS, the scanned documents and the lab results. I ended up juggling all these different systems on two screens.... Tabbing between the Lab results system, then electronic patient discharge letters, then scanned letters from the GP and then opening up our coding system to add the codes to the patient’s current episode. I also have Google for researching the conditions I am coding. I need 5 screens and four pairs of hands. It actually took longer on the EPR because it is not structured.’
‘And then there’s SNOMED CT’ added James, as if we didn’t have enough to worry about.
‘SNOMED CT?’ I asked. (6)
‘Yes, we currently use the International Classification of Disease Version 10 and the intention, as I understand it,’ continued James, ‘is that clinical terms will be added to the emerging electronic record by the clinicians using the Snomed CT terminology. These will then be linked/mapped across to the ICD codes (currently version 10 moving to version 11 no doubt in a couple of years), these codes are for analytical use, financial and epidemiological uses and the coders’ role will change. Snomed CT is a very different structure with attributes and concepts and values of attributes and and ……..
‘….and I’m glad I’m retiring soon I tell you,’ said Barbara. ‘I’ve done my time. I’m not changing now for anyone! ’
‘Well for us it IS happening,’ Jane joined in. ‘EPR’s have to be in place by 2018 ready for SNOMED CT in 2020.
‘’………and I guess when I go, they won’t replace me. We’re all doomed’ added Barbara. This month’s two page column was rapidly becoming four and more like a Dads Army script!
‘Doomed I say! Drinks everbody?’ she ended with a flourish turning to the bar.
Jane took me to one side. ‘Most coders are scared they will lose their job when this long-awaited electronic patient record and the new terminology comes. Some of us think it will just change the role to more of a review/audit type of role rather than actually making them redundant.’
‘Yes - that was my understanding of this EPR,’ I added, ‘that the clincians would capture clinical words/terms electronically as part of the delivery of clinical care using live clinical systems in real-time using Snomed CT. But then my understanding is that someone (a qualified coder) still then has to go through this (now) electronic record and either assign codes using the rules of clinical coding or validate those codes that ‘the system’ has automatically applied’. ‘That may well be right but it doesn’t stop them from being worried’.said Jane.
‘So you Enigma lot, what’s the answer?’ I asked, turning round to them at the bar, and trying to bring the conversation to an end before the football programme did’.
Laura: ‘At present, we can only assign codes using the clinical data we have in front of us. If it is incomplete or incomprehensible, we can’t code accurately. If there are too many ‘ifs’ and ‘maybes’, we can’t code accurately. We aren’t allowed to go back and delve deeply into the records. If the clinician does not think an underlying clinical condition is relevant to this current episode or simply forgets to document it, then we are not ‘allowed’ to code it.’
One of them added ‘To be honest, I’m not convinced that all clinicians appreciate these implications. Clinical coding is not ‘just something for the managers’, it is essential to the efficient running of any Trust. It is essential for the funding of work done. But more importantly it is critical to get it right for the delivery of clinical care once the clinical terminology Snomed is introduced. In short, it is key.’
James:’ The new electronic patient record isn’t just a pulling together of lots of different computer systems into one view. It will contain coded information, free text; letters created on the PAS and other computer systems and not forgetting scanned documents. This 'record' has to have a structure that enables the users, be they clinicians or other secondary users, to find their way around it easily.’
'Otherwise it will take a lot longer doing any coding' added Barbara who looked like she'd just sucked a lemon!
Jane:’ Snomed is long-awaited but its introduction along with the new OPCS for coding operative procedures and then, in a couple of years, ICD-11 will have huge implications. I mean we’ve actually only just implemented ICD-10 ! These changes should not be underestimated – they have the potential to be a major upheavel for everyone’
Adam: ‘I wish I were a lad again!’
Barbara: ‘Glad I’m leaving soon!’’
The television inevitably drew my attention back like a moth to a light. Gary Linekar was interviewing Bobby Charlton. ‘How do you think your England team of 66 would have coped against Iceland’?
‘Oh wed’ve won 1-0’
‘Only one nil? Surely you would have scored more than that?’
‘Well, we are all over seventy now’ he said.
2 Cleaned up a bit!
3 Cleaned up a bit more!
ICD 10 Coding used:
Sore throat (unspecified) - J02.9
Influenza A – B J10.1
Meningitis - G03.9
Viral throat infection - J02.8 + B97.8
Staphyloccocal throat infection - J02.8 + B95.8
Pain in the neck - M54.2