Improving question formulation for use in evidence appraisal in a tertiary care setting: a randomised controlled trial [ISRCTN66375463]

• Main Authors: Rajendran Meera, Fennessy Paul A, Burrows Elizabeth A, Villanueva Elmer V, Anderson Jeremy N
• Format: Article
• Language: English
• Published: BMC 2001-11-01
• Series: BMC Medical Informatics and Decision Making
• Online Access: http://www.biomedcentral.com/1472-6947/1/4

<p>Abstract</p> <p>Background</p> <p>The specificity of clinical questions is gauged by explicit descriptions of four dimensions: subjects, interventions, comparators and outcomes of interest. This study determined whether adding simple instructions and examples on clinical question formulation would increase the specificity of the submitted question compared to using a standard form without instructions and examples.</p> <p>Methods</p> <p>A randomised controlled trial was conducted in an evidence-search and appraisal service. New participants were invited to reformulate clinical queries. The Control Group was given no instructions. The Intervention Group was given a brief explanation of proper formulation, written instructions, and diagrammatic examples. The primary outcome was the change in the proportion of reformulated questions that described each the dimensions of specificity.</p> <p>Results</p> <p>Fifty-two subjects agreed to participate in the trial of which 13 were lost to follow-up. The remaining 17 Intervention Group and 22 Control Group participants were analysed. Baseline characteristics were comparable. Overall, 20% of initially submitted questions from both groups were properly specified (defined as an explicit statement describing all dimensions of specificity). On follow-up, 7/14 questions previously rated as mis-specified in the Intervention Group had all dimensions described at follow-up (<it>p</it> = 0.008) while the Control Group did not show any changes from baseline. Participants in the Intervention Group were also more likely to explicitly describe patients (<it>p</it> = 0.028), comparisons (<it>p</it> = 0.014), and outcomes (<it>p</it> = 0.008).</p> <p>Conclusions</p> <p>This trial demonstrated the positive impact of specific instructions on the proportion of properly-specified clinical queries. The evaluation of the long-term impact of such changes is an area of continued research.</p>