Improvements in quality of clinical care in English general practice 1998-2003: longitudinal observational studyBMJ 2005; 331 doi: https://doi.org/10.1136/bmj.38632.611123.AE (Published 10 November 2005) Cite this as: BMJ 2005;331:1121
All rapid responses
This article touches on points made by the same group in 2003.(1) In
that publication improvements in care were reported, but these had largely
preceded the UK government’s publication of the National Service
Frameworks. This seems to contradict the inferences in the introduction of
this publication that the government’s ‘thrust to improve quality of care’
has anything to do with the observations in this study. Indeed in the
discussion the authors conclude ‘we cannot attribute these changes with
certainty to any intervention’. The authors have declared no competing
interests but one of the authors has advised on quality improvement
For a longitudinal study the absence of a control group and the
absence of blinding could have influenced the results in the study
Promoting the recording of care to become a measure of care is
convenient. It removes the need to consider whether any observed
improvement in recorded care merely represents an improvement in
administrative efficiency or whether this indeed represents improved
quality of care.
This shortcut was first described by the same group to circumvent the
problem of the gap between giving and recording of care. It is likely that
parallel improvements in IT with templates for care delivery are playing
an important part in the observed improvement of quality in this study,
this was indeed suggested by the same group previously.(3)
If there was no gap between actual care and recorded care, the UK
Quality and Outcomes Framework (QOF) scores would have been as projected.
The over-performance of UK general practice QOF in 2005 demonstrates the
previous discrepancy of recording and delivery of care. The authors refer
to previous research to justify inclusion of data recording as a quality
marker.(4) However this study found no association between documentation
and clinical outcomes and the association between documentation and
satisfaction disappeared after adjustment for patient variables. Indeed
the authors conclude ‘thoroughness of documentation should not be used as
a quality indicator.’
In the abstract the authors state ‘more improvements in coronary
heart disease occurred in large practices and practices in affluent
areas.’ It is a shame that large, small or affluent is not described or
discussed, nor the relation (if any) between practice size and affluence.
This might be a result of the space limitations of the Journal.
(1) Campbell S, Steiner A, Robinson J, Webb D, Raven A, Roland M. Is
the quality of care in General medical practice improving? Result of a
longitudinal observational study. BJGP 2003; 53: 298-304.
(2) Roland M. Linking physicians’ pay to the quality of care – a
major experiment in the United Kingdom. NEJM, 2004; 351: 1448.
http://content.nejm.org/cgi/reprint/351/14/1448.pdf Accessed 11-11-05
(3) Kirk S, Campbell S, Kennell-Webb S, Reeves D, Roland M, Marshall
M. Assessing the quality of care of multiple conditions in general
practice: Practical and methodological problems. Qual. Saf. Health Care
2003; 12: 421-427. http://qhc.bmjjournals.com/cgi/reprint/12/6/421
(4) Solomon D, Schaffer J, Katz J, Horsky J, Burdick E, Nadler E,
Bates D. Can history and physical examination be used as markers of
quality? An analysis of the initial visit note in musculoskeletal care.
Med. Care 2000; 38(4): 383-91. (Ovid full text accessed via Athens on 11-
We are researching primary care inputs and outputs, including the QOF. Our findings do not support a link between data recording and quality of care.
Competing interests: No competing interests