From the Consulting Room on third party information records

This article first appeared on Microsoft’s NHS Resource Centre on 28 April 2010 (site now defunct; ).

Dr Neil Paul mulls over some problems with clinical coding, and the growing problem of references to third parties in medical records.

I recently attended a really interesting refresher course on child protection. During the discussion the whole concept of third party information came up, and this got me thinking again about how we record data in our notes.

Once again, I think our GP computer systems let us down and the way we record consultations needs rethinking and improving.

Coding – a work in progress

For a while, people have been suggesting that the way we record codes needs developing and that we should be able to modify codes by stating things like whether the issue they refer to is on the patient’s left or right. “The way we record consultations needs rethinking and improving.”

Many want the ability to code things as a provisional/working diagnosis. So rather than coding chest pain as a heart attack when that hasn’t been confirmed, it would be better to code it as chest pain (suspected heart attack) and only code heart attack once it was confirmed.

However, what is definite? Perhaps we should be able to state levels of certainty when entering a code and possibly who made the diagnosis - do you trust an F1 doctor as much as the Professor of Medicine?

Even experts get it wrong. Or things change. For instance, I recently had a patient that I was fairly certain had an essential tremor. The consultant neurologist agreed that was the most likely diagnosis - until the fancy scan came back showing it was Parkinson’s.

How do we code that? There are plenty of examples of new diagnoses being created once disease processes are better understood. I have always been a bit dubious about the theory that every consultation should have a Read Code as this just generates incorrect codes and spurious ones.

I would rather only enter a code when it was definite and use a much tighter formulary. We also need better tools to combine and merge these codes showing the progression of a condition, without losing the working diagnoses along the way. What we have at the moment is crude.

Whose note is it anyway?

As well as recording who told us the information and how certain they were, perhaps we also need a way of recording exactly who it is they were talking about?

Patient records can contain all sorts of information - and not just about them. In these days of information sharing, we all get worried about who is going to see our records. But are we concerned enough about what we can learn about third parties?

When looking at my records, you could work out that my dad has diabetes because it’s in my family history section - should you be able to know that about him without permission to see his notes, even if you have permission to see mine?

What about the person who is a carer for a wife with mental health issues? That might be in his notes; should that be kept secret? Perhaps it should be if he has a minor laceration and he attends a walk-in centre. But what if he’s unconscious after a road traffic accident?

I’m surprised by how much detail some of my junior colleagues put on the screen, I’ve always relied on memory and a few clues.

What about the patient who is getting a divorce who whinges about their partner. Perhaps she mentions medical things, such as snoring or - more seriously – depression. Or non-medical things with possible health impacts like affairs (which carry an STD risk) or alcoholism. Or non-medical things with wider ramifications, such as allegations of child abuse. “When records are in the cloud, are all these references to third parties going to be properly policed?”Should this be accessible from her notes? Should the information be put in her partner’s notes as possible diagnoses?

What about stuff that they mentioned with permission when they were together? Should it remain in their notes after their divorce? Has anyone ever reviewed anyone’s notes to remove references to ex partners?

When we get a request for copies of records for reports, we are meant to go through them removing references to third parties. Does anyone really do this properly?

I certainly have a quick flick through the notes on my computer screen and rack my brains to think if there is anything obvious – but if you are a GP or clinician, can you honestly tell me that for every patient you get out all the paper records and remove every entry? I think there is increasing risk here re future legal action.

Drawing lines

Luckily, most requests of this kind ask for info from a certain date. And in most cases we have a fairly good memory for which patients might have more sensitive records.

However, when records are in the cloud, are all these references to third parties going to be properly policed, even when they are accessed properly? It seems to me there should be a way of recording info about third parties so it is separate to the main record.

About the author: Dr Neil Paul is a full time GP working at the Ashfields primary care centre in Sandbach. He sits on his primary care trust’s professional executive committee and has a lead role for IM&T and Payment by Results.

Tags: clinical coding, cloud, GP, records