Vergelijking van opvolgingspercentages bij twee communicatiemethoden en invloed van adviesfunctionaris bij de presentatie van adviezen uit een klinisch beslis singsondersteunend systeem op een afdeling geriatrie

OBJECTIVE: To determine whether two active alert communication methods-telephone intervention or live intervention-and communication by differently qualified persons, lead to differences in alert compliance on a geriatric ward. DESIGN: Prospective intervention study. METHODS: All patients admitted to the geriatric department of a Dutch hospital were included. The clinical decision support system (CDSS], Gaston, generated alerts using eighteen clinical rules. Relevant alerts were communicated with preseribers using two communication methods: telephone intervention, where the alerts were communicated by the hospital pharmacist by telephone, and live intervention, where the alerts were communicated by a medical research student on the ward. If the correct action occurred after alert communication, this was scored as alert compliance. A review session was used to evaluate the correctness of the alert classification during the study. RESULTS: The CDSS generated 148 unique alerts during both study periods. Alert compliance was 29% (n = 17/58) for telephone intervention and 31% (n = 27/90] for live intervention. Expressed as percentage of relevant alerts, telephone intervention [61%, n = 14/28) resulted in better alert compliance than live intervention (44%, n = 27/62, P = 0.131). The review session showed that of 20 reviewed alerts 8 (40%) were classified differently compared with the initial alert classification for both alert methods. CONCLUSION: This study showed no preference for either telephone intervention or live intervention as the better alert communication method. The results demonstrate that profession and knowledge of the person who judged the alerts affects the quality of alert classification.