Can context explain heterogeneity in complex interventions?

Complex interventions are usually defined as containing several interacting components. These components usually include people (researchers), who are  trying to influence other people (healthcare professionals) to do, or not do something (adhere to clinical guideline recommendations). Some common complex interventions that have been shown to be effective in changing clinical practice include providing educational materials, attending educational meetings, and face to face educational outreach visits. While all these interventions have shown modest but important effects on clinical practice, it is often difficult to understand the actions of the healthcare professionals being influenced to change.

On the one hand we know that there can be wide variation in how the intervention is designed and implemented; printed brochures, posters and journal articles may be read, discussed with colleagues or distributed by senior and respected staff. In fact we know that journal articles alone do not always describe these complex interventions fully. Ideally we would want to know what the intervention is, who can use it and what training they need, where it should be used, for how long and why.

On the other hand, it is often the way in which the people act that either enhances or limits the effect of an intervention; brochures may be ignored, colleagues may criticise journal articles, but managers may decide to reinforce or enforce certain behaviours. The way people act will inevitably, be influenced by their environment. The term context is widely used to describe the environment, or situation surrounding an event, and these circumstances may help in understanding the event. There are often subtle differences in the environments in which healthcare professionals work. Even when people do what is asked of them, they will adapt their behaviour based on positive or negative feedback they receive. These simple adaptations, which may be a consequence of their environment can ultimately influence the way interventions are adopted.

If many non-drug healthcare interventions are complex, and depend on people to do something different than their usual everyday habits, then could it be that the way people behave determines how well they implement the intervention? If this behaviour depends on their context, could it be the context that actually limits or enhances the measured effect?

If context can explain some of the variation in actual practice, could this clinical diversity explain a large part of the statistical heterogeneity that is commonly reported but unexplained in the usual forms of sensitivity and subgroup analyses in high quality systematic reviews of complex interventions?

There is an inherent irony here, in that clinical trials are usually designed to exclude the potential confounding factors of context. We recognise that randomisation is designed to do this within a particular study. But it seems that when many studies of the same intervention are synthesised, the different contexts in which people work influences their behaviour in various ways, which together limit the expected positive effect.

There is a burgeoning interest in defining and evaluating context in healthcare research. Sadly, it seems that the inherent complexity of context can be best understood in retrospect. But perhaps there is a middle ground, to look for aspects of context that have been reported by  study authors to have contributed to the intervention’s effectiveness. Perhaps it might be most easily found in those (unpublished) articles where the intervention did not achieve the effect desired!!


I conduct and supervise research that promotes the translation of quality research evidence for use in clinical practice.

Tagged with: , , , , , , ,
Posted in complex interventions, implementation

Please make a comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: