Framework for design and evaluation of complex interventions to improve health

Dear Sir - Recognising the limits of the randomised controlled trial
in addressing all questions in the complex enviroment of health care,
Campbell et al (1) succeed not only in missing the point but demonstrating
the dissonance between academic theory and end user reality.

Firstly, while focusing on "optimum interventions" and "optimal
effectiveness" they give no clear idea of what we are seeking to optimise
or how it should be achieved. The dictates of effectiveness, efficiency
and patient/public acceptability are often mutually exclusive. In complex
systems solutions are satisfactory, never optimal.
Secondly, the primary aim of research in complex systems is to facilitate
the allocation of limited resources. Cost effectiveness is acknowledged as
an after thought whereas the converse may be more appropriate - starting
with the way in which resources are allocated, identifying possibilities
for change and constructing research activities around these margins.

But there are more fundamental concerns arising from their limited
understanding of what complex systems really are. Although acknowledging
the contribution of qualitative research, the authors remain entrapped
within the modern, Newtonian paradigm and confuse complicated (linear,
predictable, understandable by a break down into component parts) with
complex. Complexity is the result of a rich interaction of simple elements
that only respond to the limited intervention each of them are presented
with. (2) The key point is that complex systems are non-linear,
unpredictable and can not be understood by a disaggregation of their
component parts.

Engaging with complexity infers that simplistic solutions are rarely
obtainable and that we can only ever make general remarks about the
behaviour and dynamics of a system. Concentrating on being vaguely right
rather than precisely wrong but thinking always of context and opportunity
cost.

This paper misses an important opportunity to move the research agenda
forward into the real world. Its importance is to illuminate the gap
between the academic
research community and the consumers of their activity who struggle with
uncertainty, ambiguity and paradox.

1 Campbell M, Fitzpatrick R, Haines A, et al.. Framework for
design and evaluation of complex interventions to improve health. BMJ
2000;321:694-6.

2 Cilliers P. Complexity and post modernism. Ch 1, P5 Routledge
Press 1998.

David Kernick

General Practitioner

St Thomas Health Centre,
Cowick Street, Exeter EX6 7SW