Chemotherapy prescribing errors: an observational study on the role of information technology and computerized physician order entry systems
- BACKGROUND:Chemotherapy administration is a high-risk process. Aim of this study was to evaluate the frequency, type, preventability, as well as potential and actual severity of outpatient chemotherapy prescribing errors in an Oncology Department where electronic prescribing is used.
- METHODS:Up to three electronic prescriptions per patient record were selected from the clinical records of consecutive patients who received cytotoxic chemotherapy between January 2007 and December 2008. Wrong prescriptions were classified as incomplete, incorrect or inappropriate. Error preventability was classified using a four-point scale. Severity was defined according to the Healthcare Failure Mode and Effect Analysis Severity Scale.
- RESULTS:Eight hundred and thirty-five prescriptions were eligible. The overall error rate was 20%. Excluding systematic errors (i.e. errors due to an initially faulty implementation of chemotherapy protocols into computerized dictionaries) from the analysis, the error rate decreased to 8%. Incomplete prescriptions were the majority. Most errors were deemed definitely preventable. According to error presumptive potential for damage, 72% were classified as minor; only 3% had the potential to produce major or catastrophic injury. Sixty-eight percent were classified as near misses; adverse drug events had no or little effect on clinical outcome.
- CONCLUSIONS:Chemotherapy prescribing errors may arise even using electronic prescribing. Although periodic audits may be useful to detect common errors and guide corrective actions, it is crucial to get the computerized physician order entry system and set-ups correct before implementation. 
Chemotherapy is an extreme source for medication errors due to complexity of the regiments, the narrow therapeutic ranges of the drugs, frequent does changes, medication delivery is very intricate, and on top of all that cancer patients are more susceptible to drug interactions than most other patients. In 2002 the American Society of Health-System Pharmacists released guidelines for preventing chemotherapy prescribing errors. This guidelines claims that CPOE systems offer superior results over pre-printed prescription forms. Though relatively few studies have studied CPOE systems in oncology. The aim of this study is to evaluate the frequency of errors that occur in an ontology ward with a CPOE system as well as categorize these errors.
The study was conducted at the Department of Oncology, University Hospital of Udine, Italy. The information system used, G2, is entirely home grown and was first used only by the oncology unit in 2001.
Prescriptions were selected from the medical records for outpatients who received treatment from Jan 2007 to Dec 2008. To limit the data exculsion criteria was implemented: prescriptions issued by not fully qualified oncologists, prescription treatments within a clinical trial, and if an incorrect prescription was corrected. A maximum of three prescriptions for each patient record were analyzed by two specially trained oncology residents. A special record form was used to collect the data.
The collected errors were then evaluated as a whole and classified by error type, potential preventability, potential severity, and actual clinical impact. Different systems and methods were used to classify all collected errors.
335 eligible records were used that gave a total of 835 eligible prescriptions.
- Incomplete prescriptions were the most common (66%), followed by incorrect and inappropriate (28% and 6%).
- 99% of all the errors were deemed preventable.
- possibility of severity: 72% of the errors were minor, 25% were moderate, and 2% were major, and 1% were catastrophic
Medication errors is one of the most common causes of morbidity and death in patients, and most of these errors occur in the first stage of the medication process, prescribing. CPOE is used to reduce these errors while IT solutions help to reduce errors that occur in the dispensing and administrating of the drug. Even with the use of CPOE the error rate in this study was around 20%. That being said the errors in this study had no or little impact on patients.
IT was found that even with CPOE systems there is a prescribing error occurring. To fix this, it is crucial that protocols be set up correctly before using the CPOE system. It is also thought that additional evaluation of CPOEs in oncology centers are needed.
I found it odd that there was such a high prescription error rate in oncology wards. I was always under the impression that oncology wards had very few medication errors comparatively. Though this paper does shed light onto misconceptions that an average individual would have. I did find this paper that talks about evidence based medicine in the field of oncology, which fits nicely with what was discussed in the discussion.
- Aita, M., Belvedere, O., De Carlo, E., Deroma, L., De Pauli, F., Gurrieri, L., … Fasola, G. (2013). Chemotherapy prescribing errors: an observational study on the role of information technology and computerized physician order entry systems. BMC Health Services Research, 13, 522. http://dx.doi.org/10.1186/1472-6963-13-522