Implementing technology to improve medication safety in healthcare facilities: a literature review.
Medication errors remain one of the most common causes of patient injuries in the United States, with detrimental outcomes including adverse reactions and even death. By developing a better understanding of why and how medication errors occur, preventative measures may be implemented including technological advances. In this literature review, potential methods of reducing medication errors were explored. Furthermore, technology tools available for medication orders and administration are described, including advantages and disadvantages of each system. It was found that technology can be an excellent aid in improving safety of medication administration. However, computer technology cannot replace human intellect and intuition. Nurses should be involved when implementing any new computerized system in order to obtain the most appropriate and user-friendly structure.
Medication errors are one of the most common but preventable causes of iatrogenic injuries in the United States. The Institute of Medicine report To Err is Human: Building a Safer Health System estimated that between 44,000 and 98,000 people die each year from medication errors (Kohn, Corrigan, & Donaldson, 2000). Outcomes of medication errors, including adverse reactions, may result in prolonged hospital stay with increased cost, patient disability, and even death (Kozer, Berkovitch, & Koren, 2006; Stetina, Groves, & Pafford, 2005). In Indianapolis, in 2006, two infants died and three were in critical condition after they were given overdoses of heparin ("Hospital changes procedures," 2006). The wrong concentration of heparin was stocked in the computerized drug administration system by an experienced pharmacy technician. Because heparin vials with different concentrations had similar packaging, the nurse retrieved the vial with a higher concentration and accidentally administered the lethal doses ("Hospital changes procedure," 2006). Similarly, in November 2006 a nurse in Wisconsin was prosecuted for administering a bag of epidural analgesia instead of the ordered penicillin (Holt, 2006). The prosecutors claimed the nurse failed to adhere to the five rights of medication administration (Table 1), which are considered crucial in reducing medication errors (Kozier, Erb, Berman, & Snyder, 2004; Lefrak, 2002). Additionally, the nurse was charged with neglecting to use the bedside barcode system in order to properly identify the patient and medication. She faced up to 6 years in prison for criminal neglect resulting in bodily harm (Holt, 2006).
Medication errors kill one person every day in the United States and contribute to multiple additional injuries (Menachemi & Brooks, 2006). Such errors are on the rise and, according to the MEDMARX voluntary medication error tracking system, there were close to 200,000 medication errors reported in 2002. The number of errors is estimated to be much higher, but many go unreported for fear of consequences (Lafleur, 2004).
With the implementation of computerized medication order and administration systems, many errors can be prevented (Rothschild, 2004). There are many technology systems currently available and the field of health information technology is growing rapidly. The importance of nurses' involvement in implementing computerized systems and appropriate training of personnel should be emphasized.
The purpose of this article is to describe the meaning of medication errors and explain reasons why medication errors are not appropriately reported. Furthermore, different technology tools available for medication orders and administration are reviewed, including advantages and disadvantages of each system. The overall goal is to explore how the healthcare field can make medication administration safer for patients and prevent the number of errors currently reported.
Defining medication errors
Medication errors are preventable events that may result in serious harm to patients. More specifically, medication errors are defined as "error[s] in prescribing, dispensing, or administering a medication" (Kozer et al., 2006, p. 1156). Furthermore, "medication error includes any preventable event that may cause or lead to inappropriate prescribing, dispensing, or administering of a medication." (p. 1156). Tragically, these errors result in the deaths of thousands of patients each year, especially in the pediatric population (Walsh et al., 2006). In an effort to reduce these preventable errors, Congress introduced two bills, the Health Information and Quality Improvement Act of 2001 and the Medication Errors Reduction Act. In combination these bills awarded more than $1 billion in grants for implementation of medication technology (Lafleur, 2004).
The types of medication errors are described by Kozer et al. (2006) as dosing errors including administration error, prescribing errors, and dispensing errors. An administration error happens when the wrong dose of medication is calculated and administered to patients (Papastrat & Wallace, 2003). Types of administration errors include tenfold errors, where the medication is calculated or transcribed to an amount 10 times higher than the recommended dose by using the wrong unit. The outcomes of tenfold errors can be fatal (Grissinger & Globus, 2004). For example, an order that is written in micrograms but is transcribed or calculated in milligrams can lead to fatal results, as with the medication digoxin.
In the category of prescribing errors, there is a high incidence of mistakes with verbal orders (Grissinger & Globus, 2004). Verbal orders should be used minimally because they lead to a greater risk of errors due to misunderstandings, misinterpretation, and miswriting (Kaplan, Ancheta, Jacobs, & Clinical Informatics Outcomes Research Group, 2006).
Finally, the category of dispensing errors includes wrongly dispensed medication or medication dose from the pharmacy, or technical errors when medications are administered through automated dispensing systems (Grissinger & Globus, 2004).
Medication errors are more prevalent in pediatric patients, complex patients such as those who have underlying disease or are very sick, and patients receiving medications in ambulatory care settings. In pediatrics, medication dosing is often calculated by the patient's weight or body surface area (BSA), which makes this population more prone to medical calculation errors (King, Paice, Rangrej, Forestell, & Swartz, 2003). Patient groups also prone to medication errors include patients with underlying diseases such as impaired renal function requiring renal dosing or very sick patients in an intensive care setting. The treatment plans for these patients are complex with multiple medications and healthcare providers often have little time to carefully check medication dosages (Rothschild, 2004). Outpatient and ambulatory care settings show an increase in the amount of medication errors based on the lack of teaching regarding over-the-counter medications. Patients may believe these non-prescription medications are harmless both in terms of dosing and combining medications. Without proper education for patients on the appropriate use of these non-prescription medications, morbidity and even mortality may result (Curry, Walker, Hogstel, & Burns, 2005).
The experience level of healthcare providers and their working conditions may contribute to an increase or decrease in medication errors. Based on recent studies, Kozer et al. (2006) reported that the risk of medication errors was higher when the order was written by an inexperienced physician rather than a staff physician. Physicians, practitioners, and nurses with advanced experience were more likely to detect medication errors than trainees. A retrospective cohort study conducted by Kozer et al. (2002) showed that in an emergency department of a pediatric tertiary hospital, the risk of error was higher when the order was given by an inexperienced physician than by one with experience. Although medication errors occur at all levels, based on these studies it is reasonable to believe that more experienced providers will recognize the mistakes.
Furthermore, it was previously denied by the medical community that the amount of work hours had an effect on performance of healthcare workers (Kozer et al., 2006). However, recent studies show a strong correlation between the workload hours of healthcare providers and the number of medication errors. For example, in a study (Landrigan et al., 2004) correlating working hours in an intensive care unit with the number of serious medication errors, the number of errors was significantly reduced when the on-call hours in a shift for medical interns were decreased to 16 hours with the maximum weekly work hours reduced from 80 hours to 63 hours. Additionally, the type of shift worked also influenced the number of errors. Reducing the hours in a shift demonstrated a reduction in medication errors (King et al., 2003; Lockley et al., 2004).
Although nurses have expressed concern about the quality of care and increase in errors when working extended or night shifts, few studies have examined the relationship between work hours of nurses and patient safety. In a study by Scott, Rogers, Hwang & Zhang (2006) on the effects of nurses' work hours on vigilance and patients' safety, findings indicated that shifts greater than 12.5 consecutive hours almost doubled the risk for nurses making an error. Furthermore, the risk of falling asleep and consequently having decreased alertness almost doubled when shifts exceeded 8 hours. These findings supported the Institute of Medicine recommendations to minimize the use of 12-hour shifts (Scott et al., 2006). (1)
The consequences of medication errors include diagnostic evaluations to evaluate organ function if toxic medication dosages are administered. Furthermore, treatments such as antidotes for the medication or even renal dialysis may be necessary. These interventions may prolong patients' hospitalizations and even result in death. Additionally, the cost of medication errors needs to be absorbed by patients, insurance companies, the healthcare system, and taxpayers (Kozer et al., 2006).
Reporting medication errors
Ideally, voluntary reporting of medication errors should be expected from ethical, professional nurses, but for various reasons there is a lack of error reporting. Nurses may neglect to report medication errors in fear of being penalized and possibly losing their jobs (Lafleur, 2004). As a result, many errors go unreported and accurate statistical data are lacking (Kozer et al., 2006). A qualitative study by Stetina et al., (2005) confirmed these assumptions. In their study on how to manage medication error outcomes, nurses were found to be reluctant to report medication errors for fear of the consequences. Surprisingly, several of the nurses lacked knowledge of what qualified as a medication error and also when to report these mistakes. For example, the administration of a medication outside the given time limit according to hospital policy was not considered an error by the nurses. According to Kozer et al., (2006) one solution is to implement a surveillance system including chart review, incidence reports, and direct observation.
In an attempt to resolve the blame issue and improve the reporting of medication errors, a non-punitive and open approach with an emphasis on organizational learning is indicated as likely to facilitate reporting (Dickens, 2007). In a study investigating methods to detect medication errors, Dickens found that hospital units with a healthy attitude and openness towards error reporting also had a higher frequency of reported errors. Instead of attempting to hide a mistake, nurses took appropriate steps in medication error reporting (Dickens, 2007). Rather than blaming the individual, the entire healthcare system needs to be evaluated in order to make improvements and prevent similar medication errors from occurring (Papastrat & Wallace, 2003).
A variety of technologies can be implemented to reduce medication errors. Some of the methods used in healthcare settings to reduce medication errors and improve efficiency include the computerized physician order-entry system (CPOE), electronic medication administration record (eMAR), handheld computers, bar coding systems, automated decision support, and automated dispensing cabinets (ADCs) (Galanter, Didomenico, & Polikaitis, 2005; Grissinger & Globus, 2004; Kaplan, et al, 2006; Lafleur, 2004; McRoberts, 2005; Rothschild, 2004). Any technology used in healthcare settings should be considered as additional safety precautions. Additionally, any system implemented in a healthcare setting should not overshadow the advanced knowledge of providers or decrease basic safety precautions. Technologies, including medication administration systems, are not foolproof. Basic safety precautions such as the five medication checks can help in reducing errors (Lafleur, 2004).
Computerized physician order-entry (CPOE)
CPOE is a computer software program designed to reduce the number of medication errors and increase efficiency (Rothschild, 2004). Some believe that the acronym CPOE should stand for computerized provider order entry, to acknowledge that clinicians such as NPs and PAs, in addition to physicians, make medical orders. The system may improve communication and decision making within the multidisciplinary healthcare team and, as a result, could enhance overall patient outcomes. With the use of CPOE, orders are written directly into the computer and transferred to the pharmacy where they are reviewed. With appropriate input of patient characteristics, including weight, allergy status, and conditions altering the care plan such as impaired renal function, medication dosage calculations are performed by the software. In cases where the medical dosage is manually changed, the CPOE alerts the ordering physician of the patient's status affecting the medication adjustment (Rothschild, 2004).
Electronic medication administration record (eMAR)
The eMAR is usually a subsystem of the CPOE. It is a software program limited to medication orders that are transcribed prior to medication administration (Kaplan et al., 2006). The eMAR has been especially successful in reducing medication errors related to verbal orders. In addition to following standard protocol for verbal orders, including repeating back the order to the physician for confirmation and providing accurate identification of the prescriber and the patient, the order is immediately transcribed into the eMAR system. In a study by Kaplan et al., (2006), this process was shown to reduce the number of medication errors by 34%.
Bar coding systems
Bar codes from medications and vaccines have been used for several years after implementation by the federal Food and Drug Administration for safety reasons. For example, an adverse reaction to a measles, mumps, and rubella vaccine was easily tracked by bar code. The newer and more advanced use of bar codes is in the form of scanning. The bar-code scanning technology involves a hand-held scanner used at the bedside to confirm the five rights of medication administration. According to Lafleur (2004), the use of bar codes in medication administration can reduce errors significantly, in some studies by more than 20%. However, it is emphasized that the success of any bar-code scanning system depends on the amount of user training with the device.
Automated clinical decision support (CDS)
The automated clinical decision support system is another software program implemented with the goal of reducing medication errors. The software program makes suggestions in the course of care when patient data are entered into the system. In a study by Galanter et al. (2005), CDS was implemented on a unit in hope of reducing administration of medications contraindicated for patients with renal insufficiency. Although the implementation of a CDS system greatly reduced the administration of contraindicated medications, it was observed that healthcare providers were noncompliant with alert recommendations.
Automated dispensing cabinets (ADC)
The ADC consists of an automated dispensing system with stock medications in patient-care areas rather than the traditional unit-dose dispensing system. In addition to reducing medication errors by dispensing the correct medication to be administered, the ADC is also cost beneficial in streamlining drug distribution and tracking drug charges. Close to 60% of hospitals in the United States now use the ADC system (Grissinger & Globus, 2004; Lafleur, 2004)
There are multiple studies that have shown positive findings in reducing medication errors with implementation of computerized systems. In a study by King et al. (2003) the implementation of CPOE was found to decrease medication errors by 40% in areas where it was implemented as compared to units using hand-written orders. However, one limitation to the study was that medication errors in the handwritten order group were based on incidence reports of medication errors. Since many medication errors are suspected to go unreported, the numbers may be higher than what was reported in the study. CPOE reduced medication errors by eliminating illegible handwriting and decreased errors associated with similar drug names. Because the computer calculates the medication dose, a decrease in errors was observed especially in high-risk areas such as pediatrics. Additionally, due to warnings of conditions such as renal or cardiac functions warranting dosage adjustments, certain medication errors were greatly reduced. For example, creatinine clearance values were entered into the system and dosage calculations were adjusted accordingly. Similarly, drug-to-drug interaction and the patient's allergy status were flagged by the system. Menachemi and Brooks (2006) also found similar benefits to implementing a CPOE system. Dosing guidelines were followed more closely, which added to the reduction in errors. Furthermore, there was greater clinician and patient satisfaction due to improved productivity and workflows with the reduction of a paper-based process.
Besides the direct impact on patient safety in terms of reducing medication errors, CPOE has been found to be helpful in tracking typical errors or prescribing habits and therefore improving future safety. Healthcare providers also found the system helpful in improving communication among members of the multidisciplinary healthcare team and increasing the efficiency of healthcare delivery. Finally, CPOE improved the use of evidence-based clinical guidelines (Kozer et al., 2006; Rothschild, 2004).
In a study by Kaplan et al. (2006), the implementation of CPOE greatly reduced the number of verbal orders given by providers and accepted by nurses. In their study, the rate of verbal orders went from 23% to 10% before and after initiation of a CPOE system. However, the encouraging results were also due to the reinforcement of nurses not accepting verbal orders and education regarding the risks of making errors. The nurses were also actively involved in the enforcement of policies discouraging verbal orders at any time. Finally, Kozer et al. (2006) reported a drastic decrease in tenfold error when using a CPOE system. The computer makes the correct calculation but, more importantly, the software will not allow unaccepted abbreviations.
The automated CDS system has also been found to greatly decrease medication errors. In a study by Galanter et al. (2005), there was a 42% drop in serious errors related to specific dosing for underlying conditions.
The negative findings of research studies were mostly technical in origin, but also included cost, limited selection of software, and reluctance to implement computerized systems from healthcare workers. In relation to the technology of the software, computerized medication ordering and administration systems are not flawless. One of the problems with a CPOE system is the danger of double dosing when there are different pathways in the system. In a pediatric study by Walsh et al. (2006), different medication errors in a CPOE system not usually seen in the traditional paper-ordering were detected. Some of these errors resulted from overriding alerts or warnings produced by the CPOE system. Sometimes the nurse would proceed in the ordering system without reading the alert. Additionally, standardized order sets created problems related to different pediatric age groups who varied in size. Kozer et al. (2006) found limitations with the CPOE system when the medications were required to be prepared immediately on the unit. Other disadvantages reported included the cost and limited choice of vendor offerings. Menachemi and Brooks (2006) suggested using internal systems rather than purchasing commercial CPOE technology in order to reduce the cost.
Finally, there is still resistance among certain healthcare employees to learn new technology. The key is to include nurses and nurse informatics in the process of implementing any new computer technology (Rothschild, 2004).
Since cost is a major factor in implementing new technology in any healthcare facility, it is important to evaluate the alternatives when a computerized system is not financially feasible. These include templates, color-coded standard dosing systems, pharmacy auditing, regulations, emphasis on education, and teamwork. Templates consist of standardized preprinted order sheets with specific patient information (Kozer et al., 2006). The color-coded system includes devices such as the Broselow Pediatric Emergency Tape, which implements color-coding of drug doses for children of different sizes and weights (Shah, Frush, Luo, & Wears, 2003). Careful pharmacy auditing, especially in high-risk areas such as intensive care units, has been shown to decrease the incidence of medication errors considerably (Grissinger & Globus, 2004). Implementation of hospital regulations and policies for certain procedures and medications reduces the incidence of errors (Stumpf, 2007). For example, with certain drugs, two nurses are required to sign when the medication is administered. Implementation of strict regulations is beneficial in terms of safety related to working hours. Since fatigue has a strong correlation with medication errors, additional safety precautions are necessary (Scott et al., 2006). Finally, the number of educational hours allocated to prescribing and preventing of medication errors needs to be increased both in medical and nursing schools (Stetina et al., 2005). Open communication and teamwork among healthcare professionals are crucial in order to prevent medication errors and provide maximal patient safety (Kozer et al., 2006).
Technology can be an excellent device in improving safety, but any system is only as good as its users. Computer technology cannot replace the human intellect and intuition. By relying too much on computerized systems, nurses may ignore simple, yet crucial safety steps such as the five rights. In terms of medication dosage calculations, most of the problems require only basic arithmetic skills. Without practice and daily use of these calculations, however, many physicians, practitioners, and nurses become incompetent in proper medical dosage calculations. Nurses need to question whether technology is used to support their own judgment or acts as the decision maker. An extreme example is the fatal heparin medication error mentioned in the introduction.
When implementing a new computerized system such as CPOE or eMAR, it is crucial that nurses be involved from the planning stage. By choosing a user-friendly system for nurses, efficiency and work satisfaction will likely improve. Additionally, a computerized system customized to the users' needs may increase their confidence in the use of technology. By incorporating these factors, the predicted outcome will be a reduction in the total number of medication errors.
By implementing a systems approach to medication errors, it is hoped that nurses will be less intimidated by and afraid of reporting medication errors. Without the direct blame for mistakes, nurses will likely report more errors, resulting in a realistic statistical picture of medication errors. As a result, a plan to prevent future mistakes can be developed. There is no doubt that medication safety is an ongoing challenge that cannot be changed overnight. Through the use of appropriate technology, education, training, and open communication, however, the benefits will outweigh the negative factors.
Curry, L. C., Walker, C., Hogstel, M. O., & Burns, P. (2005). Teaching older adults to self-manage medications: Preventing adverse drug reactions. Journal of Gerontological Nursing, 31(4), 32-42.
Dickens, G. (2007). Inpatient psychiatry: Three methods to detect medication errors. Nurse Prescribing, 5(4), 167-171.
Galanter, W. L., Didomenico, R. J., & Policaitis, A. (2005). A trial of automated decision support alerts for contraindicated medications using computerized physician order entry. Journal of the American Medical Informatics Association, 12(3), 269-74.
Grissinger, M., & Globus, N. J. (2004). How technology affects your risk of medication errors. Nursing2004, 34(1), 36-41.
Holt, C. (2006, November 13). Wisconsin to prosecute nurse for medical error. Retrieved March 6, 2007, from http://nursing.advanceweb.com/common/ EditorialSearch/AViewer.aspx?CC=79183
Hospital changes procedures after two premature infants die of overdoses. (2006, September 18). Associated Press. Retrieved March 6, 2007, from http:// www.whas11.com/sharedcontent/APStories/stories/D8K7IAP80.html
Kaplan, J. M., Ancheta, R., Jacobs, B. R., & Clinical Informatics Outcomes Research Group. (2006). Inpatient verbal orders and the impact of computerized provider order entry. Journal of Pediatrics, 149, 461-474.
King, W. J., Paice, N., Rangrej, J., Forestell, G. J., & Swartz, R. (2003). The effect of computerized physician order entry on medication errors and adverse drug events in pediatric inpatients. Pediatrics, 112, 506-509.
Kohn, L. T., Corrigan, J. M., & Donaldson, M. (Eds). (2000). To err is human: Building a safer health system. Washington, DC: National Academy Press.
Kozer, E., Berkovitch, M., & Koren, G. (2006). Medication errors in children. Pediatric Clinics of North America, 53, 1155-1168.
Kozer, E., Scolnic, D., Macpherson, A., Keays, T., Shi, K., Luk, T., et al. (2002). Variables associated with medication errors in pediatric emergency medicine. Pediatrics, 110, 737-742.
Kozier, B., Erb, G., Berman, A., & Snyder, S. (2004). Fundamentals of nursing concepts, process, and practice (7th ed.). New Jersey: Pearson.
Lafleur, K. J. (2004). Tackling med errors with technology. RN, 67(5), 29-34.
Lefrak, L. (2002). Moving toward safer practice: Reducing medication errors in neonatal care. Journal of Perinatal & Neonatal Nursing, 16(2), 73-84.
Landrigan, C. P., Rothschild, J. M., Cronin, J. W., Kaushal, R., Burdick, E., Katz, J. T., et al. (2004). Effect of reducing interns' work hours on serious medical errors in intensive care units. New England Journal of Medicine, 351, 1838-1848.
Lockley, S. W., Cronin, J. W., Evans, E. E., Cade, B. E., Lee, C. J., Landrigan, C. P., et al. (2004). Effect of reducing interns' weekly work hours on sleep and attentional failures. New England Journal of Medicine, 351, 1829-1848.
McRoberts, S. (2005). The use of bar code technology in medication administration. Clinical Nurse Specialist, 19, 55-56.
Menachemi, N., & Brooks, R. G. (2005). Reviewing the benefits and costs of electronic health records and associated patient safety technologies. Journal of Medical Systems, 30, 159-168
Papastrat, K., & Wallace, S. (2003). Teaching baccalaureate nursing students to prevent medication errors using a problem-based learning approach. Journal of Nursing Education, 42(10), 459-464
Rothschild, J. (2004). Computerized physician order entry in the critical care and general inpatient setting: A narrative review. Journal of Critical Care, 19(4), 271-278.
Scott, L. D., Rogers, A. E., Hwang, W., & Zhang, Y. (2006). Effects of critical care nurses' work hours on vigilance and patients' safety. American Journal of Critical Care, 15(1), 30-37.
Shah, A. N., Frush, K., Luo, X., & Wears, R. L. (2003). Effect of an intervention standardization system on pediatric dosing and equipment size determination: A crossover trial involving simulated resuscitation events. Archives of Pediatrics & Adolescent Medicine, 157(3), 229-36.
Stetina, P., Gorves, M., & Pafford, L. (2005). Managing medication errors: A qualitative study. Medsurg Nursing, 14(3), 174-178.
Stumpf, P. G. (2007). Taking aim at the top 3 patient safety errors in ob/gyn. Contemporary OB/GYN, 52(1), 58-62.
Walsh, K. E., Adams, W. G., Bauchner, H., Vinci, R. J., Chessare, J. B., Cooper, M. R., et al. (2006). Medication errors related to computerized order entry for children. Pediatrics, 118, 1872-1878.
(1) Nurses continue to debate the perceived benefits of minimizing 12-hour shifts versus their value for recruitment and retention.
Unn Hidle, MS, MEd, CRN, CPNP
Unn Hidle is an assistant professor of nursing at LaGuardia Community College in Long Island City, NY.
Table 1. The Five Rights of Medication Administration Five rights Questions to ask Right medication Is the right medication ordered for the right patient? Right dose Is the medication dose accurate? In pediatrics, is the dose according to the patient's weight? How often is the medication dose administered? Right time What time is the medication to be administered? Should the medication be given before, during, or after a meal? Right route How is the medication administered? What route is correct for administration of the medication? Is the route of administration accurate according to the order? Right patient Was the right patient identified using appropriate institutional policy measures such as patient name, date of birth, and medical record number?
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of the New York State Nurses Association|
|Date:||Sep 22, 2007|
|Previous Article:||Consider all options.|
|Next Article:||Nurses' attitudes about end-of-life referrals.|