Editor's Choice US editor's choice

Explaining negative results

BMJ 2007; 334 doi: http://dx.doi.org/10.1136/bmj.39225.424479.3A (Published 24 May 2007) Cite this as: BMJ 2007;334:0-a
  1. Douglas Kamerow, US editor
  1. dkamerow{at}bmj.com

    This week's BMJ contains research that goes contrary to expectation, which raises all sorts of questions. Richard Holland and colleagues studied the effectiveness of sending pharmacists to visit heart failure patients in their homes (doi: 10.1136/bmj.39164.568183.AE). These pharmacists, called community pharmacists in the UK, had been specially trained to deal with patients about therapeutics and were knowledgeable about drug side effects and adherence problems. They made two visits to each of 293 heart failure patients shortly after hospital discharge and spent an average of almost six hours with each patient.

    Despite expectations that this intervention would reduce subsequent hospitalizations and improve patients' quality of life, it didn't. The reasons for this failure are not obvious. The authors state that it might be because the intervention wasn't delivered as intended, or was given to patients too late in the course of their disease to make a difference, or because the pharmacists were not specialists in heart failure. In an editorial commentary (doi: 10.1136/bmj.39213.660394.80), Peri Ballantyne wonders whether it was because the analysis emphasized global outcomes rather than focusing on what was really being delivered: instructions on how to take drugs, to both patients and their doctors.

    In the US, studies of home visits by specially trained nurses have improved outcomes in heart failure patients. Is there a difference between specialist nurses trained to work with heart failure patients and community pharmacists? Probably, related both to their specific preparation and their professional background. Another article in this issue sheds light on the subject. Charlotte Salter et al observed the research interventions of 11 of the pharmacists in a previous trial by the authors of the above study with 29 patients aged 80 or more (doi: 10.1136/bmj.39171.577106.55). They found that the pharmacists usually bombarded the patients with information and advice, even when the patients resisted it. Ballantyne, in her editorial (doi: 10.1136/bmj.39213.660394.80), says that the pharmacists might have been more successful if they had considered the patients' perspectives, how they make decisions, and what their agendas were for the sessions. It's a good example of how qualitative research can complement and inform randomized controlled trials.

    View Abstract