Effect of Electronic Prescribing Strategies on Medication Error and Harm in Hospital: a Systematic Review and Meta-analysis

Abstract

Background: Computerized physician order entry and clinical decision support systems are electronic prescribing strategies that are increasingly used to improve patient safety. Previous reviews show limited effect on patient outcomes. Our objective was to assess the impact of electronic prescribing strategies on medication errors and patient harm in hospitalized patients.

Methods: MEDLINE, EMBASE, CENTRAL, and CINAHL were searched from January 2007 to January 2018. We included prospective studies that compared hospital-based electronic prescribing strategies with control, and reported on medication error or patient harm. Data were abstracted by two reviewers and pooled using random effects model. Study quality was assessed using the Effective Practice and Organisation of Care and evidence quality was assessed using Grading of Recommendations Assessment, Development, and Evaluation.

Results: Thirty-eight studies were included; comprised of 11 randomized control trials and 27 non-randomized interventional studies. Electronic prescribing strategies reduced medication errors (RR 0.24 (95% CI 0.13, 0.46), I² 98%, n = 11) and dosing errors (RR 0.17 (95% CI 0.08, 0.38), I² 96%, n = 9), with both risk ratios significantly affected by advancing year of publication. There was a significant effect of electronic prescribing strategies on adverse drug events (ADEs) (RR 0.52 (95% CI 0.40, 0.68), I² 0%, n = 2), but not on preventable ADEs (RR 0.55 (95% CI 0.30, 1.01), I² 78%, n = 3), hypoglycemia (RR 1.03 (95% CI 0.62-1.70), I² 28%, n = 7), length of stay (MD - 0.18 (95% - 1.42, 1.05), I² 94%, n = 7), or mortality (RR 0.97 (95% CI 0.79, 1.19), I² 74%, n = 9). The quality of evidence was rated very low.

Discussion: Electronic prescribing strategies decrease medication errors and adverse drug events, but had no effect on other patient outcomes. Conservative interpretations of these findings are supported by significant heterogeneity and the preponderance of low-quality studies.

Keywords: CDSS; CPOE; electronic prescribing; medication error; preventable adverse drug events.

Figure 1

PRISMA study flow diagram. CPOE, computerized physician order entry; CDSS, clinical decision support system. *Ten of 14 authors contacted to confirm study eligibility. If author was not reached, and study eligibility remained unclear, study was excluded.^†Two of 5 authors contacted clarified data for quantitative analysis.

Figure 2

Quantitative analysis using forest plot for the effect of electronic prescribing strategies on risk of a overall medication errors and b dosing errors. a Overall medication errors. b Medication dosing errors. RCT, randomized controlled trial; NRIS, non-randomized interventional study; M-H, Mantel-Haenszel random effects model. Counts are expressed as events (errors) per total number of prescriptions, except Terrell which is events (errors) per total number of renal dosing alerts. Studies are ordered by calendar year. a Medication error definitions: prescription incomplete (Ali); unintended discrepancies (Zoni); any error in drug ordering, transcribing, dispensing, administration or monitoring (Aziz, Walsh, Van Doormal); proportion of variance between ordered and administered meds (Taylor); any error including drug name, pharmacologic form, dosing, allergy, or interaction (Armada); any pharmacy intervention for wrong dose, drug, patient, drug interaction, allergy, missing medication, or wrong dosage form (Davis); incomplete, insufficient information, illegible, error of prescribing decision or other (Warrick); dosing error (Holdsworth); dosing within 30% above or below appropriate drug dose from gentamycin, vancomycin, and enoxaparin (Roberts). Garner et al. (NRIS) not included in meta-analysis as number of errors exceeded number of prescriptions (1.1 errors/prescription in control phase). Han et al. (NRIS) not included as medication errors expressed as number of errors per 1000 patient-days. Definition unspecified. b Dosing error definitions: incomplete or wrong dose (Ali); > 10% over- or underdosing for age and weight (Garner, Taylor); gentamycin/enoxaparin/vancomycin dosing conformity with 30% of dose (Roberts); error in dosage of dosing figures (Armada); error in strength, frequency, dosage (Aziz, Davis), or length (Van Doormal).

Figure 3

Meta-analysis using forest plot for effect of electronic prescribing strategies on a adverse drug events (ADE), b mortality, c length of hospital stay, and d hypoglycemic events. a Adverse drug events (ADE) and preventable ADE. b Mortality. c Length of stay (in days). d Hypoglycemic events (all RCTs). RCT, randomized controlled trial; NRIS, non-randomized interventional trial; M-H, Mantel-Haenszel random effects model. a Counts are expressed as events (ADE) per total number of patients. Subtotals not pooled due to duplication of studies in each subgroup. b Counts are presented as deaths per number of patients in each group. Mortality is presented as follows: 30-day mortality (Dean), 180-day mortality (Leibocivi), ICU mortality (Haddad), hospital mortality (Micek, Haddad, Newton, Han), overall mortality (Bertsche, Nachtigall at 2 time points), and 12-month mortality (LeMeur). c Counts are mean (SD) length of stay in days in each group, analyzed with mean difference in each group. Studies reported the following: hospital length of stay (Van Doormal, Haddad, Newton, Micek, Dean, Han) and ICU length of stay (Nachtigall). Dean results unadjusted and originally reported as median (95%CI) due to skewness. d Counts are patients with hypoglycemic events in each group. All studies of hypoglycemic events are RCTs. Subtotals are not pooled due to study duplication in reporting mild and severe hypoglycemic events. Dumont et al. not included in meta-analysis of hypoglycemic events as data were presented in hypoglycemic events per total glycemic measurements.