Obtaining information accurately and quickly: Are structured abstracts more efficient?

BACKGROUND: Structured abstracts - which, like the present one, contain several subheadings - have replaced traditional abstracts in most medical journals. Evaluation studies have shown that such abstracts usually provide more information, are of a higher quality, facilitate peer review and are generally welcomed. AIM: The aim of the studies reported here was to investigate an additional possible advantage for structured abstracts - namely, whether or not they are easier to search. METHOD: Two studies are reported. In Study 1, using an electronic database, 52 readers were asked to find the answers to two questions for each of eight abstracts in one format (say, traditional) followed by two questions for each of eight abstracts set in the other format. Time and error data were recorded automatically. In Study 2, using a printed database, 56 readers were asked to find five abstracts that reported a particular kind of study (e.g. studies with school-children and reading tests) and then to find five more that reported another kind of study. Again, the order and presentation of the format of the abstracts was counterbalanced. Time and error data were recorded manually. RESULTS: In Study 1, the participants performed significantly faster and made significantly fewer errors using the structured abstracts. There were, however, some unexplainable practice effects. In Study 2, the participants again performed significantly faster and made significantly fewer errors with the structured abstracts. In Study 2, however, there were asymmetrical transfer effects: participants who responded first to the structured abstracts responded more quickly to the following traditional ones than did those participants who responded first to the traditional abstracts. CONCLUSIONS: The overall findings, notwithstanding certain caveats, support the hypothesis that it is easier for readers to search structured abstracts than it is to search traditional ones.