23 April 2012
A study has shown that two commercial electronic prescribing systems were associated with a statistically significant reduction in total prescribing error rates (66.1%, 57.5% and 60.5% in a total of three intervention wards in two hospitals). This was driven by a substantial reduction in incomplete, illegal and unclear prescription orders (90.2% in Hospital A and 93.6% in Hospital B). However, both hospitals experienced errors related to the new system, which accounted for 35% of prescribing errors in the intervention wards.
Electronic prescribing systems should be introduced with care so any system-related errors are minimised. Prescribers and other healthcare professionals who will be using the system should be appropriately trained and errors should be continuously monitored following the introduction of the system.
It is likely that many system errors can be reduced by modifying the system features (e.g. listing frequently used options first on drop-down menus or creating pre-structured orders to reduce the need for users to construct complex order sentences). For any errors for which this is not possible, targeted training to highlight specific risks should be considered.
What is the background to this?
Commercial electronic prescribing systems are being introduced in many countries. Although there is evidence that home-grown electronic prescribing systems are associated with a reduction in medication errors, little is known about the efficacy of commercial systems and there have been concerns that they may introduce new types of system error. This study evaluated the effectiveness of two commercial prescribing systems in reducing prescribing error rates and their propensities for introducing new types of error.
The researchers looked at the medication charts in two Australian teaching hospitals before and after the introduction of electronic prescribing systems. In Hospital A, the Cerner Millennium electronic prescribing system was implemented on one ward, and three wards, which did not receive the electronic prescribing system, were used as controls. In Hospital B, the iSoft MedChart system was implemented on two wards. Both of these systems had limited decision support. Procedural (unclear, illegal and incomplete prescriptions) and clinical (e.g. wrong drug, strength, dose, frequency or route; duplicated therapy; drug not prescribed; interaction) errors were identified.
What does this study claim?
At baseline, an average of 5.8 prescribing errors per admission was seen across the six wards. Implementation of electronic prescribing significantly reduced overall prescribing errors from 6.25 to 2.12 per admission (p<0.0001) in Hospital A and from 3.62 to 1.46 in Hospital B (p<0.0001). No significant change in overall prescribing errors was seen in the Hospital A control wards.
A marked reduction in procedural errors drove the reduction in overall prescribing errors. The procedural error rate fell from 4.89 to 0.48 per admission (p<0.0001) in the intervention ward in Hospital A, and from 2.66 to 0.17 per admission (p<0.0001) in Hospital B.
There was little change in the rate of clinical errors for the intervention wards (and an increase in one) following the introduction of electronic prescribing. However, prescribing errors classed as serious were significantly decreased from 0.25 to 0.14 per admission across the intervention wards (p=0.0002). Serious errors were those considered likely to lead to death; a major permanent loss of function; or permanent reduction in bodily functioning, increased length of hospital stay or surgical intervention. The Hospital A control wards experienced no significant change in serious errors.
Both hospitals experienced errors related to the electronic prescribing system (0.73 and 0.51 per admission). These accounted for 35% of overall prescribing errors in the intervention wards following the introduction of the new systems. Of the system-related prescribing errors, 99% were clinical errors. Each electronic prescribing system was associated with different types of system-related errors.
Prescribing errors are common. The EQUIP (Errors—Questioning Undergraduate Impact on Prescribing) study investigated prescribing errors by foundation trainees (doctors in the first 2 years after leaving medical school) and compared them with errors made by more senior doctors. The researchers examined 124,260 medication orders across 19 hospitals. Of these, almost one in 10 prescriptions (8.9%) were incorrect, with almost one in 50 errors (1.7%) considered to be potentially lethal. When examined by level of training, the error rate was 8.4% for foundation year 1 doctors, compared with 10.3% for those in foundation year 2, 8.3% for those in fixed-term specialty training posts, and 5.9% for consultants. Very few prescribing errors caused harm to patients because almost all were intercepted and corrected. The intervention of nurses, senior doctors and, in particular, pharmacists was vital in picking up errors before they impacted upon on patients.
In this study, the two electronic prescribing systems were associated with a statistically significant reduction in prescribing errors, largely due to a substantial reduction in incomplete, illegal and unclear orders. There was limited change in clinical error rates. The researchers state that system-related errors were a major reason for the electronic prescribing systems not delivering a significant reduction in clinical errors.
The high number of system-related errors suggests that, if a new electronic prescribing system is introduced, users should receive sufficient training and errors should be monitored continually and the design of the system refined accordingly to improve the safety and appropriateness of medication use.
A single observational study, such as this one, can prove only association not causation and observational studies are prone to confounding. This study has limitations. For example, it was not possible to randomise intervention wards, to obtain a control ward at Hospital B, or choose which wards should be included in the study. In addition, data collection periods were different at the two sites which may have influenced the results.
Electronic prescribing systems are only one way in which prescribing errors can be reduced. The EQUIP study investigators made the following recommendations:
- A standard drug chart should be introduced throughout the NHS.
- Electronic prescribing systems introduced to support safe prescribing should be monitored closely for new types of error introduced by their use.
- Clear information regarding information services and reference sources should be provided in clinical workplaces.
- Coping strategy training should be offered in undergraduate medical education.
- Clinical governance systems should make clear to senior doctors their duty to exemplify good practice and not to resolve their lack of competence by delegating prescribing to their trainees.
- Trusts should discharge their responsibility to optimise the working conditions of hard-pressed health professionals to ensure safe prescribing.
Westbrook, JI, et al. Effects of two commercial electronic prescribing systems on prescribing error rates in hospital in-patients: A before and after study. PLoS Med 9(1): e1001164. doi:10.1371/journal.pmed.1001164
Design: Before and after study involving medication chart audit of 3,291 admissions (1,923 at baseline and 1,368 post electronic prescribing system) at two Australian teaching hospitals.
Intervention: At Hospital A, the Cerner Millennium e-prescribing system was introduced on one ward (geriatric); three other wards (geriatric, renal/vascular and respiratory) acted as controls. At Hospital B, the iSoft MedChart e-prescribing system was introduced on two wards (psychiatry and cardiology); no control ward could be obtained.
Comparison: The researchers examined medication charts for procedural errors (unclear, incomplete, or illegal orders) and for clinical errors (for example, wrong drug or dose)
Outcomes and results: Use of an electronic prescribing system was associated with a statistically significant reduction in error rates in all three intervention wards—reductions of 66.1% (95% confidence interval [CI] 53.9% to 78.3%); 57.5% (95% CI 33.8% to 81.2%); and 60.5% (95% CI 48.5% to 72.4%) were seen. The use of the system resulted in a decline in errors at Hospital A from 6.25 per admission (95% CI 5.23 to 7.28) to 2.12 (95% CI 1.71 to 2.54) and at Hospital B from 3.62 (95% CI 3.30 to 3.93) to 1.46 (95% CI 1.20 to 1.73).
See also What did this study claim? section above
Sponsorship: National Health and Medical Research Council
Further information on reducing risk and medication errors can be found on NHS Evidence and in the Increasing safety and managing risk in medicines management systems and processes e-learning section of the on the NPC website
Please comment on this rapid review using our feedback form.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.