Brief interventions in the inpatient setting

Yesterday, I summarized a review of randomized controlled trials (RTCs) of brief alcohol interventions in primary care and emergency care settings. The authors of that review concluded that brief interventions in those settings had proven efficacious. In contrast, according to a review by McQueen et al. (2010), efficacy is less evident among those admitted to hospital wards.

The authors reviewed eleven studies, five of which were RTCs with “adequate methods of randomisation and allocation concealment.” They concluded,

The evidence for brief interventions delivered to heavy alcohol users admitted to general hospital is still inconclusive. From data extracted from two studies it appears that alcohol consumption could be reduced at one year follow up though further research is recommended.

Why would brief interventions have a larger impact in primary and emergency settings than inpatient ones? First, it isn’t clear that statement is true. It could be that more studies are required before drawing that conclusion. In fact, McQueen at al. point to a number of problems with the available inpatient evidence, carefully reviewing the possible sources of bias.

However, if it is the case that brief interventions don’t work as well for admitted patients, what might explain that? Here I can only speculate:

  • Patients are more willing to accept the advice of primary care practitioners than those they encounter in hospital wards. Maybe they have stronger relationships with their primary care providers.
  • A visit to the ER* for an alcohol-related emergency makes a larger impression. (Of course, hospitalization can follow from an ER visit.)
  • The hospitalized population is somehow different from those not hospitalized in a way that makes them less responsive to brief interventions.

I think I’m most comfortable saying we just don’t know enough about brief interventions for hospitalized patients.

* Sorry, is it “ED” now, a per a recent exchange in the comments? Is “ER” not OK to use?

Hidden information below


Email Address*