Evidence based implementation of complex interventions

Dear Sir / Madam,

Thompson’s editorial reflections (Thompson, 2009) on the attempt by
Besson and colleagues (Besson et al., 2009) to encourage uptake of the
Ottowa ankle rules perpetuate the ubiquitous misuse of the term ‘complex’
in discussion of health interventions and their implementation and impact.
Referring to the clinical decision tool in this way detracts attention
from the real challenges involved in achieving change in adaptive systems
which demonstrate “dynamic conservatism” (Argyris & Schön, 1978).
The delivery of health care involves people and procedures operating in
systems designed to achieve a range of goals - sometimes these goals and
evidence conflict.

In the context described by Besson et al, (2009) we contend, it is
not the intervention which explains the problems of implementation of
evidence based practice. The "Ottawa ankle rules" were informed by
evidence that suggests that at a population level lots of x-rays are
unnecessary. However, clinicians in A&E deal with individual human beings.
Six year old Tarquin has fallen off his swing and is taken to A&E because
he is complaining of a painful ankle. The nurse is sure it is not broken
but Tarquin’s Mum and Dad are not happy about him being sent home without
an x-ray. Tarquin is sent for X-ray, receives the all clear and he and Mum
and Dad go home happy. Although she attended a recent in-service tutorial
and has the new form, the referring nurse is unlikely to have agonised
over the detail of the Ottawa ankle rules. She is concurrently expected to
get people in an out of the A&E department within target times, avoid re-
presentation of patients and to provide patient-centred care. Furthermore
she would rather be safe than sorry – when she trained, evidence based
medicine was firmly in favour of radiography to promote patient benefit
rather than its minimisation to promote efficiency savings.

We contend that misapplication of the term ‘complex’ to relatively simple interventions, even those which comprise multiple components, obscures a fundamental flaw in the capacity of the evidence-based quest for predictability. By calling simple interventions complex we deflect necessary attention from the need for methodologies that help us better understand the inherent unpredictability of outcome generation in the real world.

Argyris C & Schön D Organizational learning: A theory of action
perspective. Reading, Mass: Addison Wesley. 1978.

Bessen T, Clark R, Shakib S, Hughes G. A multifaceted strategy for
implementation of the Ottawa ankle rules in two emergency departments. BMJ
2009;339:b3056.

Thompson R Evidence based implementation of complex interventions BMJ
2009;339:b3124

Competing interests:
None declared