Synopsis

Papers on electronic prescribing (computerised prescriber order entry) published in 2011 are summarised below under a range of headings. Key findings are as follows:

Effect on outcomes: An observational before and after study in the ICU of hospital in Saudi Arabia found no difference in key indicators (mortality, length of stay, use of mechanical ventilation) following the introduction of e-prescribing in the unit.

Effect on medication errors and patient safety: Numerous studies were published in 2011 in a range of healthcare settings. The general conclusion seems to be that the introduction of e-prescribing is likely to reduce some types of prescribing (and probably dispensing) errors, particularly those associated with handwriting, legibility, transcription and use of abbreviations. Evidence regarding more significant types of error is equivocal, and some studies have found increased rates of certain types of error associated with e-prescribing systems, particularly in the period immediately following implementation. This is perhaps not surprising, as non-procedural errors will depend not just on the use of an e-prescribing system per se, but on the specific system used, the clinical decision support provided (if any) and the arrangements for prioritising and presenting alerts.

Other evaluations: Studies report the degree of satisfaction of doctors, patients and pharmacists with e-prescribing systems. Training and implementation issues appear to have an important bearing. A study in Australia, comparing a number of commercial e-prescribing packages against pre-defined criteria, found important differences between systems.

System design: One paper considered the different features required in an e-prescribing system for paediatric use from those required for adult patients. In another study, integrating disease-specific order subsets into a single general admission order set significantly improved the overall adoption of order sets by clinicians.

Alert presentation: A common factor of studies of clinicians’ reaction to alerts provided by e-prescribing systems is that a high proportion are overridden. System design which avoids “alert fatigue” is thus important.

Drug interaction processing: Considerable differences exist in the way e-prescribing systems present potential drug interactions, and in the knowledge bases they use to identify them. Users appear to require a system which will grade drug interaction alerts by their (potential) severity, and make it more difficult to override the most serious alerts.

Other usability aspects: Usability problems appear to be frequent in e-prescribing systems (although they do not always affect users’ declared satisfaction). User-centred design is important. Poor usability is likely to result in users developing “workarounds”, which will have an impact on the integrity of data and potentially on patient safety.

Implementation: The papers listed describe implementation of e-prescribing systems in various settings. It is preferable to involve users in the design and implementation, and to provide adequate training for all users.

General: Several papers considering the future of e-prescribing are listed. It is noted that adoption has not been as rapid as expected in many countries, and that e-prescribing is not yet a mature technology. If the potential benefits of e-prescribing in terms of safety and efficiency are to be realised, it is important that systems should be able to communicate across the continuum of care. In practice, this happens to only a limited extent so far.
E-prescribing – effects on outcomes

Impact of computerized physician order entry (CPOE) system on the outcome of critically ill adult patients: a before-after study

HM Al-Dorzi, HM Tamim, AJ Cherfan, MA Hassan, S Taher, YM Arabi

BMC Medical Informatics and Decision Making19 Nov 2011;11:71

Background: Computerised physician order entry (CPOE) systems are recommended to improve patient safety and outcomes. However, their effectiveness has been questioned. Our objective was to evaluate the impact of CPOE implementation on the outcome of critically ill patients.

Methods: This was an observational before-after study carried out in a 21-bed medical and surgical intensive care unit (ICU) of a 900-bedtertiary care centre (King Abdulaziz Medical City-Riyadh, Saudi Arabia). It included all patients admitted to the ICU in the 24 months pre- and 12 months post-CPOE (Misys) implementation. Data were extracted from a prospectively collected ICU database and included: demographics, Acute Physiology and Chronic Health Evaluation (APACHE) II score, admission diagnosis and comorbid conditions. Outcomes compared in different pre- and post-CPOE periods included: ICU and hospital mortality, duration of mechanical ventilation and length of stay in ICU and hospital. These outcomes were also compared in selected high risk subgroups of patients (age 12-17 years, traumatic brain injury, admission diagnosis of sepsis and admission APACHE II higher than 23). Multivariate analysis was used to adjust for imbalances in baseline characteristics and selected clinically relevant variables.

Results: There were 1638 and 898 patients admitted to the ICU in the specified pre- and post-CPOE periods, respectively (age 52+/-22 vs 52+/-21 years, p = 0.74; APACHE II 24+/-9 vs 24+/-10, p = 0.83). During these periods, there were no differences in ICU (adjusted odds ratio (aOR) 0.98; 95% CI, 0.7 to 1.3) and in hospital mortality (aOR 1.00; 95% CI, 0.8 to 1.3). CPOE implementation was associated with similar duration of mechanical ventilation and of stay in the ICU and hospital. There was no increased mortality or stay in the high risk subgroups after CPOE implementation.

Conclusions: The implementation of CPOE in an adult medical surgical ICU resulted in no improvement in patient outcomes in the immediate phase and up to 12 months after implementation.

Effect on medication errors and patient safety

Impact and determinants of commercial computerized prescriber order entry on the medication administration process

I Abbass, S Mhatre, SS Sansgiry, J Tipton, C Frost

Hospital Pharmacy May 2011;46(5):341-348

Purpose: The purpose of this study was to evaluate the impact of commercial computerised prescriber order entry (CPOE) on efficiency outcomes in an 864-bed community hospital.

Methods: A retrospective study was developed to measure medication errors and medication order turnaround time in St. Luke's Episcopal Hospital located in the Texas Medical Center (USA). The study data were collected by stratified random sampling through a review of medication orders submitted to the pharmacy using a paper-based order system and the CPOE system. Descriptive frequencies, chi-squared test, Wilcoxon matched-pairs sign rank test, and logistic regression and multiple regression analyses were conducted to examine the relationship among variables.

Results: Of the 1110 total orders reviewed (563 paper-based and 547 CPOE), a total of 135 medication errors were found, with 10.5% in paper-based versus 1.6% in CPOE. The most prevalent errors in paper-based orders were inappropriate abbreviations (24.4%), incorrect doses (15.6%), occurrences of allergy (13.3%) and wrong administration frequency (9.6%). In CPOE orders, the errors were occurrences of allergy (10.4%), incorrect doses (2.2%) and drug interaction (0.7%). CPOE resulted in a 50% reduction of medication order turnaround time (median = 24 minutes CPOE vs 48 minutes paper orders). A potential medication error, unidentified prescribers within medication orders, urgency of medication order, and implementation of CPOE were the significant (P less than 0.05) determinants of medication order turnaround time.

Conclusions: The implementation of a commercial CPOE system reduced medication errors and improved medication order turnaround times.

Electronic prescribing within an electronic health record reduces ambulatory prescribing errors

EL Abramson, Y Barron, K Quaresimo, R Kaushal

Joint Commission Journal on Quality and Patient SafetyOct 2011;37(10):470-478

Background: Health policy forces are promoting the adoption of interoperable electronic health records (EHRs) with electronic prescribing (e-prescribing). Despite the promise of EHRs with e-prescribing to improve medication safety in ambulatory care settings - where most prescribing occurs and where errors are common - few studies have demonstrated its effectiveness. A study was conducted to assess the effect of an e-prescribing system with clinical decision support, including checks for drug allergies and drug-drug interactions, that was integrated within an EHR on rates of ambulatory prescribing errors.

Methods: In a prospective study using a nonrandomised, pre-post design with concurrent controls, 6 providers who used a commercial e-prescribing system were compared with 15 providers who remained paper-based from Sep 2005 to Jul 2008. Prescribing errors were identified by a standardised prescription and chart review.

Results: Some 2432 paper prescriptions at baseline and 2079 prescriptions at one year were analysed. Error rates for e-prescribing adopters decreased 1.5-fold - from 26.0 errors per 100 prescriptions at baseline (95% CI, 17.4 to 38.9) to 16.0 errors per 100 prescriptions at one year (95% CI, 12.7 to 20.2; p = 0.09). Error rates remained unchanged for nonadopters (37.3 per 100 prescriptions at baseline, 95% CI, 27.6-50.2, versus 38.4 per 100 prescriptions at one year, 95% CI, 27.4 to 53.9; p = 0.54). Error rates for e-prescribing adopters were significantly lower than for nonadopters at one year (p less than 0.001). Illegibility errors were high at baseline and eliminated by e-prescribing.

Conclusions: The preliminary findings from this small group of providers suggest that e-prescribing systems may decrease ambulatory prescribing errors, which are occurring at high rates among community-based providers.

Transitioning between electronic health records: effects on ambulatory prescribing safety

EL Abramson, S Malhotra, K Fischer, A Edwards, ER Pfoh, et al.

Journal of General Internal MedicineAug 2011;26(8):868-874

Background: Healthcare providers in the USA whopreviously used older electronic health records (EHRs) with electronic prescribing (e-prescribing) are transitioning to newer systems to be eligible for federal meaningful use incentives. Little is known about the safety effects of transitioning between systems.

Objective: To assess the effect of transitioning between EHR systems on rates and types of prescribing errors, as well as provider perceptions about the effect on prescribing safety.

Design, Participants: Prospective, case study of 17 physicians at an academic-affiliated ambulatory clinic from Feb 2008 to Aug 2009. All physicians transitioned from an older EHR with minimal clinical decision support (CDS) for e-prescribing to a newer EHR with more robust CDS.

Main Measurements: Prescribing errors were identified by standardised prescription and chart review. A novel survey instrument was administered to evaluate providers' perceptions about prescribing safety.

Results: We analysed 1298 prescriptions at baseline, 1331 prescriptions at12 weeks post-implementation, and 1303 prescriptionsat 1 year post-implementation. Overall prescribing error rates were highest at baseline (35.7 per 100 prescriptions; 95% CI, 23.2 to 54.8) and lowest1 year post-implementation (12.2 per 100 prescriptions; 95% CI, 8.6 to 17.4) (p less than 0.001). Improvement in prescribing safety was mainly a result of reducing inappropriate abbreviation errors. However, rates for non-abbreviation prescribing errors were significantly higher at 12 weeks post-implementation than at baseline (17.7 per 100 prescriptions; 95% CI, 9.5 to 33.0, vs 8.5 per 100 prescriptions; 95% CI, 4.6 to 15.9) (p less than 0.001) and no different at baselinefrom 1year (10.2 per 100 prescriptions; 95% CI, 6.2 to 18.6) (p = 0.337). Survey results complemented quantitative findings.

Conclusions: Results from this case study suggest that transitioning between systems, even to those with more robust CDS, may pose important safety threats. Recognising the challenges associated with transitions and refining CDS within systems may help maximise safety benefits.

Can new technologies reduce the rate of medication errors in adult intensive care?

(Les nouvelles technologies permettent-elles de réduire les erreurs médicamenteuses en soins intensifs adultes?)

E Benoit, J Beney

Journal de Pharmacie de Belgique Sep 2011;(3):82-91

Technology tocontrol the monitoring and administration of critical drugs to unstable patients is widespread in the intensive care environment. Since the early 2000s computerised physician order entry (CPOE), bar code assisted medication administration (BCMA), 'smart' infusion pumps (SIP), electronic medication administration records (eMAR) and automated dispensing systems (ADS) have been recommended to reduce medication errors.Some 10years later, the extent to which they have been adoptedisincreasing butis stillmodest. The objective of this study is to determine the impact of these technologies on the rate of medication errors (ME) in adult intensive care. CPOE permits amarked and significant reductionin ME, especially the least critical ones. Onlyby adding a clinical decision support system (CDSS), can CPOEachieve a reductionin serious errors. Used alone, it could even increase them. The available studies do not have sufficient power to demonstrate the benefits of SIP or BCMA on ME. However, practices such as overriding of alerts have been demonstrated with these devices. Power or methodology problems and conflicting results do not allow the ability of ADS to reduce the incidence of ME in intensive care to be established. Studies investigatingsuch technologies are not very recent, are of limited number andshow defectsin their methodology, which does not allow usto determine whether they can reduce the incidence of MEs in the adult intensive care. Currently, the benefits appear to be limited which may be explained by the complexity of their integration into the care process. Special attention should be given to the communication between caregivers, the human-computer interface and the training of caregivers.

Can an electronic prescribing system detect doctors who are more likely to make a serious prescribing error?

JJ Coleman, K Hemming, PG Nightingale, IR Clark, M Dixon-Woods, RE Ferner, RJ Lilford

Journal of the Royal Society of MedicineMay 2011;104(5):208-218

Objectives: We aimed to assess whether routine data produced by an electronic prescribing system might be useful in identifying doctors at higher risk of making a serious prescribing error.

Design: Retrospective analysis of prescribing by junior doctors over 12 months using an electronic prescribing information and communication system. The system issues a graded series of prescribing alerts (low-level intermediate, and high-level), and warnings and prompts to respond to abnormal test results. These may be overridden or heeded, except for high-level prescribing alerts, which are indicative of a potentially serious error and impose a 'hard stop'.

Setting: A large teaching hospital in the UK.

Participants: All junior doctors in the study setting.

Main outcome measures: Rates of prescribing alerts and laboratory warnings and doctors' responses.

Results: Altogether 848,678 completed prescriptions issued by 381 doctors (median 1538 prescriptions per doctor, interquartile range (IQR) 328-3275) were analysed. We identified 895,029 low-level alerts (median 1033 per 1000 prescriptions per doctor, IQR 903-1205) with a median of 34% (IQR 31-39%) heeded; 172,434 intermediate alerts (median 196 per 1000 prescriptions per doctor, IQR 159-266), with a median of 23% (IQR 16-30%) heeded; and 11,940 high-level 'hard stop' alerts. Doctors vary greatly in the extent to which they trigger and respond to alerts of different types. The rate of high-level alerts showed weak correlation with the rate of intermediate prescribing alerts (correlation coefficient, r = 0.40; P less than 0.001); very weak correlation with low-level alerts (r = 0.12, P = 0.019); and showed weak (and sometimes negative) correlation with propensity to heed test-related warnings or alarms. The degree of correlation between generation of intermediate and high-level alerts is insufficient to identify doctors at high risk of making serious errors.

Conclusions: Routine data from an electronic prescribing system should not be used to identify doctors who are at risk of making serious errors. Careful evaluation of the kinds of quality assurance questions for which routine data are suitable will be increasingly valuable.

Using an enhanced oral chemotherapy computerized provider order entry system to reduce prescribing errors and improve safety

CM Collins, KA Elsaid

International Journal for Quality in Health Care Feb 2011;23(1):36-43

Objective: To reduce the probability of failure in the oral chemotherapy order, review and administration process and to reduce oral chemotherapy-related prescribing errors intercepted by clinical pharmacists prior to reaching the patient.

Design: A before-and-after cohort study.

Setting: A 719-bed multidisciplinary tertiary care institution in the USAwith a paediatric division and an outpatient cancer centre.

Participants: A multidisciplinary team characterised key elements of the oral chemotherapy process using healthcare failure modes and effects analysis (HFMEA).

Interventions: Oral chemotherapy computerised provider order entry (CPOE) was developed and implemented.

Main Outcome Measures: Pharmacist-intercepted oral chemotherapy prescribing errors over a 24-month period (before) and over a 6-month period (after) were analysed according to the error type (errors in clinical decision making, errors in transcription or errors related to prescribing policy). The incidence of prescribing errors prior to and following CPOE implementation was compared by calculating the odds ratio (OR) and the 95% confidence interval (CI).

Results: HFMEA hazard analysis revealed7 potential failure modes, with the highest hazard scores in the prescribing and administration components of the process. CPOE implementation significantly (P = 0.023) reduced prescribing error risk by 69% (OR = 0.31; 95% CI, 0.11-0.86) and eliminated certain types of errors that can lead to significant patient harm.

Conclusions: Prescribing oral chemotherapy is a failure mode with significant risk of inducing patient harm. CPOE is effective in reducing prescribing errors of oral chemotherapy and should be considered part of a fail-safe process to improve safety.

Impact of electronic prescribing support on the reduction of transcription errors at the administration stage

(Impacto de la prescripción electrónica asistida en la reducción de los errores de transcripción a la hoja de administración)

SE Garcia-Ramos, G Baldominos Utrilla

Farmacia Hospitalaria Mar-Apr 2011;35(2):64-69

Objective: To assess the impact of administration errors when transcribing treatments to nurses' administration forms, and to estimate the impact of electronically assisted prescribing (EAP) in minimising these errors.

Method: A prospective, observational study in hospitalised patients in Spain. Changes in treatment in the 24 hours before examination were analysed for a representative sample. Transcription errors were detectedby checkingfor discrepancies between thedoctors' prescriptions and the nurses' treatment administration forms. Error incidence was calculatedoverall, and by ward, type of error, administration route and potential seriousness. The possible reduction in new errors per day if EAP were introduced in all wards was estimated.

Results: Of the 416 prescriptions recorded, the overall percentage of transcription errors was 12.4%, 9.8% in medical units and 15.2% in surgical units. Most of the errors were made when a new medicine was added (29.4%) and the frequency of administration was changed (27.4%). With regard to their seriousness, 98% did not harm the patients, and 57.7% werelisted as 'Category C'.Given thatone change of treatment is made per patient per day, introduction of EAPwould bepredicted to prevent 64 new errors daily in the hospital.