PT  - JOURNAL ARTICLE
AU  - De Wit, HAJM
AU  - Mestres Gonzalvo, C
AU  - Cárdenas Ávila, JC
AU  - Derijks, HJ
AU  - Janknegt, R
AU  - Van der Kuy, PHM
AU  - Schols, JMGA
TI  - PS-011 Evaluation of clinical rules in a clinical decision support system for hospitalised and nursing home patients
AID  - 10.1136/ejhpharm-2013-000436.362
DP  - 2014 Mar 01
TA  - European Journal of Hospital Pharmacy: Science and Practice
PG  - A147--A148
VI  - 21
IP  - Suppl 1
4099  - http://ejhp.bmj.com/content/21/Suppl_1/A147.3.short
4100  - http://ejhp.bmj.com/content/21/Suppl_1/A147.3.full
SO  - Eur J Hosp Pharm2014 Mar 01; 21
AB  - Background Computerised clinical decision support systems can be defined as aiding tools that provide clinicians or patients with clinical knowledge and patient-related information, intelligently filtered or pre-set at appropriate times, to enhance patient care. Purpose To improve the currently used clinical decision support system (CDSS) by identifying and quantifying the benefits and limitations of the system. Materials and methods Alerts and handling of the clinical rules acted upon were extracted from the CDSS in the period September 2011 to December 2011. The data was analysed for the number of clinical rule alerts acted upon, percentage of relevant alerts and the reason why alerts were classified as non-relevant. Results The 4065 alerts were differentiated into: 1137 (28.0%) new alerts, 2797 (68.8%) repeating alerts and 131 (3.2%) double alerts. Of all these alerts, only 3.6% were considered relevant, i.e. when the pharmacist needed to contact the physician. The reasons why alerts were considered as non-relevant were; the dosage was correct or already adjusted, the drug had been (temporarily) stopped, the monitored laboratory value or drug dosage had already improved to within the reference range. The low efficiency of the current system can be related to three subjects; the algorithm construction, the CDSS executing the clinical rules and the data delivery to the CDSS. Conclusions The results of this study clearly show many points of improvement for the CDSS since only 3.6% of the alerts were considered relevant. We have defined three categories of importance for the efficiency when improving or developing a CDSS: algorithm differentiation, CDSS optimisation and data delivery. No conflict of interest.