15 September 2010

Assessing Equity in Systematic Reviews: Anything new?

There's a discussion piece in this week's British Medical Journal on Assessing equity in systematic reviews: realising the recommendations of the Commission on Social Determinants of Health. Systematic reviews seek to identify and synthesise all high quality studies on a particular topic. The studies that are included in systematic reviews vary, though often they're limited to randomised controlled trials.

The article sets out seven guidance points for assessing equity in systematic reviews
  1. developing a logic model,
  2. defining disadvantage and for whom interventions are intended,
  3. deciding on appropriate study design(s),
  4. identifying outcomes of interest,
  5. process evaluation and understanding context,
  6. analysing and presenting data, and
  7. judging applicability of result.
This guidance seems sound but my question is does this represent anything new?

I feel like I've heard or read variations of these "should" points for over a decade with very few concrete examples of it being done. It's probably not an accident that few analyses of equity are done within systematic reviews. It adds several levels of nuance and complexity to what is already a time-consuming and demanding task.

We desperately need high quality evidence on differential impacts, that evaluates process and describes context (a need the authors did well to identify) to inform our HIAs. I hope it's forthcoming.

1 comment:

  1. I agree, I think we have to do these ourselves somehow. By HIA practitioners getting into research and forming research consortia at national and international levels, by lobbying the public health community to undertake more evaluations of policies, plans and projects; and by lobbying for more money to flow to this kind of research.

    I also think there is a lack of skills in the public health research community to undertake the kinds of evaluations you are talking about becaus ethey complex to design and difficult to undertake.

    ReplyDelete