Patient coding and the ratings game
BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c2153 (Published 25 April 2010) Cite this as: BMJ 2010;340:c2153
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Many of the criticisms of HSMRs made by authors such as Black [2],
and Lilford and Pronovost [3], relate in practise to the failure of Trusts
to code in depth, and to variations in results depending on which HSMR was
used. That different techniques give different results should of course
surprise no-one.
Nigel Hawkes [1] in his informative scrutiny of HSMRs examines a
still more serious concern, which is that HSMRs can and are being ‘gamed’
by labelling high proportions of admissions (over 80% in one DGH) as
needing palliative care. Apart from the sheer improbability of such high
proportions of unselected "takes" comprising such cases, there is the
concern that this approach could conceal seriously deficient practice.
Relevant to these concerns are the arguably lax entry criteria for the
Liverpool Care Pathway (LCP) [5], and the poorly explained but
significantly higher mortality rates across a range of common conditions
in the UK compared to other developed countries. The suggestion that these
deaths are "expected" begs the question of how this should be defined, and
whether there is not in reality unacceptably wide variation between
individual clinicians in what consists an expected death. The weakness of
UK survival data across a range of conditions might suggest that patients
who would survive, for example, their cardiac disease, or cancer, or CVA
if managed elsewhere in the developed world may be being prematurely
written off in the UK. The "palliative care" option in the guise of the
LCP [4] [6] [7] [8] may make this kind of non-care or option for death
easy and all but impossible to prevent. This whole rather murky area is in
need of urgent examination.
Yours etc
Ronald J Clearkin
[1] Hawkes N. Patient coding and the ratings game. BMJ 2010;340:c2153
[2] Black N. Assessing the quality of hospitals. BMJ 2010;340:c2066
[3] Lilford R, Pronovost P. Using hospital mortality rates to judge
hospital performance: a bad idea that just won’t go away. BMJ
2010:340:c2016
[4] Dr Peter Hargreaves. Letter under “Heal our Hospitals”. The
Sunday Telegraph: 5 April 2009
www.telegraph.co.uk/comment/letters.5108982/Heal-our-hospitals-Doctors-now
-treat-traumas-caused -by-hospital-stays.htm
[5] Dr Adrian Treloar. Dutch research reflects problems with the
Liverpool Care Pathway. BMJ;336:p905 (26 April 2008)
[6] Mary Knowles. Continuous Deep Sedation and LCP concerns. BMJ;
336:p905 (26 April 2008)
[7] G.Craig. Palliative care and sedation: the Liverpool Care
Pathway. http://www.bmj.com/cgi/eletters/336/7648/781 (19 May 2008).
[8] Kate Devlin. Sentenced to death on the NHS. Daily Telegraph: 2
September 2009
http://www.telegraph.co.uk/health/healthnews/6127514/Sentenced-to-death-on
-the-NHS.html
Competing interests:
None declared
Competing interests: No competing interests
Variation in capturing primary diagnosis influences mortality rates
Dear Editor
Nigel Hawkes’ article on Dr Foster’s mortality data highlights the
difficulties that hospital trusts face when dealing with variation in
coding practice between hospitals.[1] It is also important to note that
the way primary diagnosis is captured at coding has an important influence
on standardized mortality rates (SMRs). The table (source: Dr Foster) below
shows the 2009 mortality rates for cerebrovascular accidents (CVAs)of
various trusts in the Eastern region . It is clear that some trusts code
their CVAs more specifically than others. Most patients admitted with a
CVA will have brain imaging within 24 hours of admission and it is
therefore now possible to categorize the CVA to infarction or haemorrhage
accurately soon after admission. Whilst some trusts use this definitive
diagnosis as the primary diagnosis, others may just use the provisional
diagnosis made at admission (that is, before any brain imaging) as the
primary diagnosis. The provisional diagnosis is often a CVA of unspecified
cause. This difference in coding practice is important as trusts with more
accurate coding of their CVAs will be compared with trusts whose coding is
not so specific, resulting in a skewed analysis of mortality rates. Dr
Foster’s data will be very valuable if coding practice is uniform across
the trusts. We urgently require a set of stringent national guidelines on
coding aiming to reduce coding variation.
1. Nigel Hawkes. Patient coding and the ratings game. BMJ 2010;340:c2153,
doi: 10.1136/bmj.c2153
Competing interests:
None declared
Competing interests: No competing interests