Three Ways in Which a Treatment Effect may be Masked in Policy / Service Delivery Studies

Contamination may mask the effectiveness of an intervention in an otherwise perfectly conducted trial. But what is contamination? The CLAHRC WM Director proposes that we do away with this moniker and instead use the term ‘masking’ to describe circumstances where improvement in the control groups vitiates a positive result. He further proposes that ‘masking’ be categorised into three classes:

  1. Mimetic masking, where the control group becomes aware of the study intervention and implements it.
  2. Neighbourhood masking, where the control group benefits from any benefit in the intervention group through the ‘neighbourhood effect’. Herd immunity is an obvious example. Another recent example relates to the effects of deworming, which reduces the load of eggs that may infect others.[1] In both cases the controls, rather than being ‘contaminated’ are ‘de-contaminated’!
  3. Temporal masking, where the intervention, or another effective intervention, comes into vogue quite independently of the study and hence where headroom for further gains in the intervention group are diminished. This has also been referred to as a ‘rising tide phenomenon’.[2]

Trials may be conceptualised as having pragmatic or aetiological motivations.[3] Categories 1 and 2 undermine both the pragmatic and aetiological motivations for a trial. Category 3 undermines only aetiological. Category 2, and to a variable extent 1, may be avoided by cluster studies, but this does not mitigate Category 3 masking.

052 DC - Three Ways Fig1-1

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Miguel E, & Kremer M. Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities. Econometrica. 2004; 72(1): 159-217.
  2. Chen YF, Hemming K, Stevens AJ, Lilford RJ. Secular trends and evaluation of complex interventions: the rising tide phenomenon. BMJ Qual Saf. 2015. [ePub].
  3. Schwartz D, & Lellouch J. Explanatory and pragmatic attitudes in therapeutical trials. J Chronic Dis. 1967; 20: 637–48.
Advertisements

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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