Alerts versus on-demand CDS

From Clinfowiki
Jump to: navigation, search

Electronic Alerts versus On-Demand Decision Support to Improve Dyslipidemia Treatment: A Cluster Randomized Controlled Trial (Article review) Jacobus T. van Wyk, Marc A.M. van Wijk, Miriam C.J.M. Sturkenboom, Mees Mosseveld, Peter W. Moorman and Johan van der Lei 2008;117;371-378; originally published online Jan 2, 2008; Circulation DOI: 10.1161/CIRCULATIONAHA.107.697201

Question: Should the hospital (or clinical organization) management enforce the physicians to use CDS system by obligating usage of reminders ad alerts or make it on demand? Background: Clinical decision support systems (CDSS) have obviously increased the quality of healthcare , but , making alerting enabled all the time makes a great change in users (clinicians) behavior but there's no clear evidence for that.

Methods: Randomized controlled trial that compare three groups of general practices. First group (13 practices) is assigned to receive alerts, the second group (12 practices)is assigned to make CDSS on-demand . The last group is the control group (11 practices) doesn't have a CDSS. Rules to choose the participated practices: All practices should completely depend on ELIAS electronic health record (iSOFT B.V., Leiden, the Netherlands) and used it for one year as minimum. The CDSS used based on Dutch college for general practitioners (DCGB) guidelines. Eligible patients between 18 and 70 years old Outcomes and measurements: The primary outcome measured by to main percentages:

Percentage of screened patients=(the number of patients needed to be screened and screened)/( The number of patient needed to be screened according to DCGB)

Percentage of treated patients=(the number of patients needed to be treated and treated)/(The number of patients needed to be treated according to DCGB)


Results: incase of alerted group Percentage of screened patients was 65% and percentage of treated patients was 66%,while in on-demand group Percentage of screened patients was 35% and percentage of treated patients was 40% in compare with the control group which has a percentage of 25% for screened patients and 36% of treated patients.

The likelihood of patient to be screened increased by 76% in the alerted group than the control group while in the on-demand group the percentage of increase was 28%. And the likelihood of being treated increased in the alerted group by 40% while in the on-demand group it was only 19% increase in compare with the control group.

Conclusion: The usage of CDSS significantly increased the percentage of patient screened and treated from dyslipidemia, in addition using an alerting system increased the screening rate significantly than the on-demand group.

article review by: Mohamed Abd-Elgawad