TY - JOUR T1 - Pitfalls of administrative database analysis are evident when assessing the ‘weekend effect’ in stroke JF - Evidence Based Medicine JO - Evid Based Med SP - 35 LP - 35 DO - 10.1136/ebmed-2016-110557 VL - 22 IS - 1 AU - Frank J Attenello AU - William J Mack Y1 - 2017/03/01 UR - http://ebm.bmj.com/content/22/1/35.abstract N2 - Commentary on: Li L, Rothwell PM. Biases in detection of apparent “weekend effect” on outcome with administrative coding data: population based study of stroke. BMJ 2016;353:i2648OpenUrlAbstract/FREE Full Text.Multiple studies attempt to clarify the role of weekend admissions in stroke.1 ,2 Recent studies have leveraged large populations in administrative databases to retrospectively evaluate hypotheses. While these studies disagree as to whether weekend admissions are associated with increased stroke mortality, concern exists regarding inherent limitations of administrative coding databases, including accuracy of patient and disease variables.3 Concern is justified, as coding inaccuracies unduly influence the study results. The Oxford Vascular Study (OXVASC), in contrast, contains detailed, comprehensive data prospectively recorded for all patients registered in Oxfordshire, the UK. The reviewed study admirably addresses administrative database accuracy, comparing data extracted from hospital coding with prospective OXVASC data, focusing on the accuracy of weekend versus weekday cohorts. … ER -