Why patient safety is such a tough nut to crackBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d3447 (Published 21 June 2011) Cite this as: BMJ 2011;342:d3447
All rapid responses
As a junior doctor, your refreshing view on patient safety is one
that I shall include in my clinical practice. I am passionate about
patient safety and the Criteria for Success part of your article is
representative of the reasons behind the many battles I have fought to
bring my ideas to patient safety issues in my trust.
Inpatients are frequently subjected to blood tests for the duration
of their hospital stay as a means to assess any improvement to treatments.
Many hospitals have a phlebotomy service that take blood from patients
once requested to do so by a physician. Usually, blood samples are taken
in the morning, processed by the hospital laboratories, and then results
are available in the early afternoon. Once available, doctors often alter
their management of a patient.
Blood tests and their results are usually requested and checked by
the most junior member of the medical team. Good medical practice dictates
that if a doctor requests a particular investigation, they should follow
up the results. Unfortunately, there are cases when patients may be
discharged before the results of investigations are known. In this
instance, the care of a patient may be compromised and this could
potentially result in a preventable adverse outcome. With this in mind, I
set out to quantify the amount of patients who have blood tests taken on
the day of their discharge, if these results were abnormal, and if they
were checked by a doctor. I then looked what proportion of patients that
were readmitted had blood tests taken on the day of discharge that were
abnormal and not checked by a doctor.
24.6% of all patients had blood tests taken on the day of discharge,
and 61.9% of these were abnormal. Just 18% of these results were checked
by a physician before discharge. 57% of re-admissions, in the 14-day
period that followed, included patients discharged with abnormal blood
results that were not checked.
The results clearly show that a significant number of patients had
abnormal test results on the day of discharge. Three patients were re-
admitted the following day, suggesting this can be a potential source of
harm. The rate of re-admissions could potentially be reduced if patients
were unable to be discharged before their test results were available.
As a junior doctor, I have presented my data at local clinical
governance meetings, to the medical director as well as at a national
patient safety conference. I have provided many ideas for change, most of
which are related to checks pre-discharge. My desire to change the system
has meant that I am still trying to use my evidence to improve patient
My hope is that with articles such as this, patient safety will
become the focus of quality improvement projects and that junior doctors
will see that we have a crucial role in changing the patient safety
culture of medicine.
Competing interests: No competing interests
The analysis by Leistikow on patient safety provides a novel and
useful way of looking at the problem of implementing patient safety
initiatives relating to visibility, ambiguity, complexity and autonomy. I
would like to challenge (clarify) the complexity section.
describes hospitals as complex and tightly coupled. This terminology was
used by Charles Perrow in his book "Normal Accidents" first published in
1984 (updated in 1999, Princeton University Press). Perrow describes two
continuums for organisations; firstly complexity (from linear to complex)
and secondly coupling (loose to tight). There is no doubt that according
to his definitions hospitals are complex organisations where it is not
easy to clearly define the "production process". In terms of coupling
however, hospitals fall more into the loosely coupled category where
delays in processing are possible, sequences can be changed and many
redundancies are available due to the flexibility of humans within the
system. Perrow did not refer directly to healthcare in his initial
description of tight and loose coupling but Erik Hollnagel placed us in
the complex & loosely coupled box in his account of performance
variability in "The ETTO principle" (Ashgate publishing, 2009).
Although Leisitkow does not lead with the specific term, his article
encompasses many strands of human factors science at individual and system
level. An understanding of human factors at every level in healthcare from
frontline clinical staff through non-clinical hospital managers and chief
executives right up to the Department of Health will help us overcome some
of the barriers to delivering safer healthcare.
Competing interests: Chris Frerk is a doctor at Northampton General Hospital. He is a trustee of the charity the Clinical Human Factors Group www.chfg.org
I do not have access to the full article but, as a patient, would
like to comment on three of the suggested four key challenges in
implementing patient safety interventions.
Visibility: We need systems that allow healthcare staff to whistle-
blow in complete safety. But, importantly, all levels of staff also need
to be able to speak out to prevent harm when they consider a proposed
action to be inappropriate or harmful. A forthcoming 'Whistleblowers'
Special' edition of Private Eye (July) will examine these issues.
Complexity: We have to look at the deeper issues affecting patient
safety. Healthcare staff themselves are 'damaged' when they do not speak
out to prevent or highlight unnecessary patient harm. Doctor/patient
communication issues (especially paternalism) can lead to physical and
psychological patient harm. Lack of communication skills training means
doctors and others 'distance' themselves for self protection, leading to
lack of compassion.
Patients' stories which graphically illustrate the ill effects of
clinicians' attitudes and other patient safety issues, can help students
and healthcare professionals acknowledge harm - the first step towards
change - so are great learning tools. Not only books but websites such as
www.patientstories.org.uk. Could these play a part in revalidation?
Websites which focus on respect and compassion
(www.collegeofmedicine.org.uk - 'provoking debate and discussion to
generate change' and www.compassioninhealthcare.org) bring patients and
healthcare professionals together to enhance learning and change
Autonomy: Changing attitudes can be helped along by the use of more
patient speakers in medical education, at all levels and as keynote
speakers in conferences, while interaction on social networking sites can
enhance understanding of patients' perspectives, for example use of
'concordance', rather than 'compliance' and foster mutual understanding
and respect. By sharing web links and reading matter such as 'Towards the
Emancipation of Patients, patients' experiences and the patient movement',
Charlotte Williamson, Policy Press, 2010, patient autonomy becomes a
natural progression, rather than being perceived as a threat to medical
Competing interests: Author, Nothing Personal, disturbing undercurrents in cancer care, Radcliffe Publishing 2008.