Association between article citation rate and level of evidence in the companion animal literature.

Abstract

BACKGROUND: Level of evidence (LOE) hierarchies rank scientific articles according to the use of study design features intended to limit bias. Citation analysis of medical articles has shown that studies with high LOE ranking are preferentially cited. OBJECTIVES: To determine whether clinical companion animal articles reporting study designs classified as high LOE are more frequently cited than those with designs classified as low LOE and to characterize other factors associated with 5-year citation rate. METHODS: Literature survey of all original clinical articles published in 2004 in 5 peer-reviewed clinical veterinary journals. For each eligible article, details of scientific and nonscientific characteristics were collected, an LOE classification was assigned, and the 5-year citation rate following publication was determined. Linear regression was used to identify factors associated with citation rate. RESULTS: Overall LOE was low with 188 of 209 eligible articles describing a study design classified as low LOE. An association was not identified between 5-year citation frequency and LOE classification or any specific feature of study methodology. Articles pertaining to infectious disease or published in the Journal of Veterinary Internal Medicine were associated with significantly greater subsequent citations. CONCLUSIONS AND CLINICAL IMPORTANCE: Reports of veterinary studies designed to limit the influence of bias are not more widely referenced than articles reporting data obtained through less stringent methodologies. Medical subspecialty and publishing journal prestige can influence an article's subsequent citation rate.