Table 1

Evidence for the effect of CPOE+CDSS on adverse drug event rates

ReferenceStudy typeSettingNumber of studiesIncluded study designsIntervention(s)Measured outcomes (level 1, 2, 3, 4)*Conclusions
Systematic reviews
 Wolfstadt et al15CPOE+CDSSHospital (9, 4 ICU) and ambulatory (1)10CCTs and observational studiesHomegrown 7 studies*
Commercially sold 3 studies
1Evaluated ADE as outcome measure;
5 out of 10 studies found significant reductions in ADEs (no RCTs; p≤0.05)
 Ammenwerth et al17Electronic prescribing including CPOE+CDSSHospital and ambulatory27 (7 CPOE+CDSS)CCTs and observational studiesHeterogeneous systems, including homegrown and commercial systems1, 2, 34 of 7 studies found that CPOE+CDSS reduced ADE rates; relative risk reduction 30–84%
 Schedlbauer et al 200916CPOE+CDSSHospital (15) and ambulatory (5)20Pre-post studies, time series and RCTs (4)Heterogeneous studies including 27 alert systems; identified basic, advanced and complex CDSS1, 2, 3Majority of CDSS demonstrated improved prescribing; only 4 studies evaluated clinical outcomes
 Van Rosse et al18CPOE+CDSSHospital (including ICU, adult and paediatric)12Observational studies onlyHomegrown and commercially sold systems1, 2, 3, 4Decreased risk of medication prescribing errors, no effect on ADEs or mortality
 McKibbon et al19MMIT including CPOE and CDSSHospital and ambulatory87 (10 CPOE+CDSS)RCTsHomegrown and commercially sold systems1, 2, 3, 4Noted improvement in process measures; few studies included patient outcomes
Narrative reviews
 Stultz and Nahata20CPOE+CDSS (medication related)Hospital and ambulatory (paediatric)44Observational studies, CCTs and RCTs includedHeterogeneous; not specified1, 2, 3No clear effect on ADE rates
Reference System studied SettingPatient population Study design CDSS featuresMeasured outcomes Conclusions
Original studies
 Leung et al22Commercial CPOE+CDSS5 community hospitalsAdult inpatientsControlled before–after1, 2, 3Preventable ADEs decreased post-implementation (p=0.007)

Non-preventable ADEs and potential ADEs increased (p<0.001 for both)
Overall rate of ADEs increased (14.6/100 vs 18.7/100 admissions; p=0.03)
 Westbrook et al21Two commercial CPOE+CDSS2 hospitalsAdult inpatientsControlled before–after2, 3Significant reduction in prescribing errors at both hospitals
  • *Level 1, clinical outcome such as ADEs, morbidity or mortality; level 2, surrogate outcome such as observed errors or intermediate outcomes associated with clinical outcomes (eg, abnormal laboratories); level 3, other variables with an possible link to outcomes (eg, prescribing practices); level 4, no reported measured outcomes associated with clinical outcomes (eg, detection).

  • ADE, adverse drug event; CCT, controlled clinical trial; CDSS, clinical decision support systems; CPOE, computerised provider order entry; ICU, intensive care unit; MMIT, medication management information technology; RCT, randomised controlled trial.