Medications Errors Reduction, Research Paper Example

Abstract

Medication errors threaten the safety of patients. In addition, they have adverse effects on staff.  Fighting the threat that medical errors present to patients’ safety is one of the key tasks of healthcare organizations around the United States.  Reason’s Human Error Theory argues that to err is human, but different errors result from different cognitive processes.  Bar coding technology is one of the most effective ways of combating medication errors that can be classified as slips, as it is designed to ensure that the right patient gets the appropriate dose of prescribed medication in an appropriate way and at the right time. BCMA systems have their faults, but still they are effective enough to implement them.  But you just said that bar coding is one of the most effective ways to prevent medication errors.

Introduction

To err is human. Nevertheless, there are spheres of life where human error can be extremely dangerous. Healthcare is one of these spheres, as failure to provide adequate care may result in patients’ suffering and even death, not to mention the enormous economic costs that can be attributed to errors.  Medical errors occur in every hospital as well as in other healthcare settings. It is impossible to eliminate errors entirely, but efforts should be made to minimize them. According to the report of the Institute of Medicine (IOM) about building a safer healthcare environment, medication errors are the most frequent type of errors in hospital settings.  The government and healthcare organizations take exceptional measures to combat medication errors, but their incidence is still high.  The purpose of this paper is to analyze the issue of medication errors through the conceptual framework of Reason’s Human Error Theory and explore the theoretical relevance and effectiveness of bar-coding medication administration as a means to prevent medication errors.

Background

Medication errors cause seven thousand deaths yearly, and the direct cost associated with these errors is estimated to be two billion dollars yearly.  These errors are the eighth leading cause of death in the United States, and the leading cause for error-related inpatient deaths (IOM, 1999; Sakowski, Newman & Dozier, 2008).  Each hospitalized patient is estimated to be the subject of one medication error per day (Koppel at al., 2008).  In response to such statistics, Joint Commission, the Institute for Healthcare Improvement, the Agency for Healthcare Research and Quality, and many other research and professional organizations have acknowledged the danger and impact of medication errors on patient safety and made reducing these errors one of their main priorities (Fowler, Sohler & Zarillo, 2009).

Three terms are used to describe events that occur with medication administration during hospitalization: medication errors, adverse drug events and preventable adverse drug events.  Medication errors can be defined as any error that occurs during the process of prescribing, transcribing, dispensing, and administering a medication (Bates, Boyle, Vander Vliet, Schneider, & Leape, 1995).  For example, if the correct medication dosage is not given or if a patient fails to take a prescribed drug, these are all considered to be medication errors (IOM, 2006).  On the other hand, adverse drug events refer to any injury or harm that is caused by a medication that was not administered in error (Bates et al., 1995b).  Lastly, preventable adverse drug events are defined as injuries that result from a medication error (IOM, 2006).  Based on these definitions, not all medication errors lead to adverse drug events and not all adverse drug events are preventable. Williams (2007) defines medical “any error in the prescribing, dispensing, or administration of a drug, irrespective of whether such error leads to adverse consequences or not”.

Aronson states that medication errors can occur on several stages: when choosing a medicine, meaning irrational, inappropriate, and ineffective prescribing, underprescribing and overprescribing; when writing the prescription: prescription errors, including illegibility; when manufacturing the formulation to be used—wrong strength, contaminants or adulterants, wrong or misleading packaging; when dispensing the formulation—wrong drug, wrong formulation, wrong label; when administering or taking the drug—wrong dose, wrong route, wrong frequency, wrong duration; during monitoring therapy—failing to alter therapy when required, erroneous alteration (Aronson, 2009).

There are several stages in the medication delivery process when a medication error can occur.  Further, different healthcare providers at each stage assume primary responsibility for preventing different causes of a medication error.  Physicians are responsible for prescribing a medication and the dose, frequency, and route of administration as well as insuring accurate and timely communication of the prescription order to other health care providers.  Pharmacists are responsible for verifying, transcribing, and dispensing the medication that has been ordered.  Finally, nurses are assume responsibility for administering medications using the five “rights” as identified by Cescon and Etchells (2008).  These rights are the right patient, right drug, right dose, right route, and right time. Although errors are possible at any stage in the medication delivery process, errors made during the prescribing and administering stages are the most frequent, constituting an estimated 26-34% of all preventable medication errors.  Yet, errors that occur during stages of the process are rarely reported and even more rarely intercepted (Koppel et al., 2008). Moreover, they are far more likely to reach the patient than are errors that occur at any other stage (Cescon & Etchells, 2008; Koppel et al., 2008).

The administration phase includes verifying the prescription that has been ordered, preparing the medication, administering the medication as prescribed to the patient, and recording the administration of a medication in the patient record.  As such, administering medications can consume up to 40% of nurses’ working time.  (Fowler, Sohler & Zarillo, 2009).  Therefore, nurses are the key targets for developing error-prevention strategies in a hospital setting.  In addition to the potentially adverse health outcomes for patients and the financial costs that can result from medication errors, Cescon and Etchells (2008) indicate that medication errors at the administering stage may result in emotional distress and other personal and professional negative consequences for nurses.

Numerous steps have been taken to minimize the number of medication errors.  The Department of Health and Human Services (DHHS) created a Patient Safety Task Force to coordinate efforts in fighting this problem.  This organization works with the National Centers for Disease Control and Prevention, the Agency for Healthcare Research and Quality, the FDA, and the Centers for Medicare and Medicaid Services (CMS) (U.S. Department of Health and Human Services, 2001).  Each agency has introduced strategies for medication error prevention.  Nevertheless, the number of medication errors remains high. In the next part of this paper, Reason’s Human Error Theory will be reviewed and applied to explain medication errors that occur during the administration phase of the medical delivery process.

Human Error Theory

Reason defines error as a failure of a previously planned action to be completed or a failure to choose appropriate means for reaching a specific purpose. According to Reason, error, in fact, can be attributed to faulty cognitive processes which are seen as the foundation for intentional action.  According to Reason (1990) there are two types of errors: active and latent. Active errors are mistakes that have immediate effect, while latent error’s effect is postponed. Moyen, Camiré and Stelfox (2009) identify latent errors as “resident pathogens within the system”. They add that latent errors can influence the frequency of active errors. Latent errors happen when individuals perform actions that have unintended and unpredictable consequences in the future. Therefore, prevention of medication errors system has to have mechanism that allow to prevent the error. If it is not possible, the system should either have self-correcting mechanisms, or, at least, allow to detect and surface error as soon, as it is possible, in order to liquidate their consequences (Reason, 1990).

The examples of latent medication errors are: incomplete clinical information about the patient, lack of effective communication with the patient, incomplete or incorrect information about drug administration, incomplete or incorrect information about drug administering method, insufficient drug information, lack of specialized training/education of personnel who administers the drugs, lack of maximum-dose warning system in pharmacy computer, inconsistent procedure for double-check before administering the drug, mistakes in handwriting: communication of doze numerically instead of phonically, mistakes in drug name and doze, ineffective communication systems between personnel who prescribes the drug, and nurses, who administer them; inadequate drug references, and other (Committee on Identifying and Preventing Medication Errors, 2006).

Moyen, Camiré and Stelfox (2009) define active errors as “unsafe acts committed by people who are in direct contact with the patient”.  Active errors may include prescribing or administering wrong drug, or wrong doze of drug, misreading the drug name or doze, administering drug to the wrong patient etc (Committee on Identifying and Preventing Medication Errors, 2006). There are much more opportunities for preventing latent errors, than for preventing active ones.

Three types of errors, each associated with different cognitive processes, are identified in this theory: skill-based, rule-based, and knowledge-based errors (Reason, 1990).

Nursing skills are inserting catheters, making injections, measuring blood pressure etc. Skills are acquired through the human’s lifetime. As such, skills can be compared to habitual behaviors that are enacted with minimal conscious thought.  Because the performance of skills becomes automatic over time, they are especially vulnerable to slips or lapses, meaning that key steps in skill performance can be inadvertently omitted because the skill is performed with little conscious thought (Reason, 1990).Slip errors constitute about 60,7% of all errors (Reason, 1990).For instance, even the experienced nurse may fail to insert the intravenous catheter, or inject a medication. Skill-based errors are connected with three aspects of information processing: attention, recognition and memory (Reason, 2003).

Rule-based actions require more complex cognitive processes than do skill-based actions.  Among the cognitive processes that are used for rule based actions, the background and the problem to be resolved must be matched with appropriate application of an existing rule (Reason, 1990).  In a healthcare setting, rule-based processes are administering patients to the hospital and discharging them, and other.

Knowledge-based actions require the use of highly complex cognitive processes because these actions are based on decisions that are made in unfamiliar situations and, thus, not amenable to the use of pre-specified rules. Knowledge based actions require an analysis of previous experiences and the integration of knowledge to arrive at a decision that is unique to a situation that is ambiguous in terms of causality and possible outcomes (Reason, 1990).  In a hospital setting, knowledge based cognitive processes regulate nonstandard situations, like resolving ethical issues with patients and the staff, making decisions in critical situations etc..  Reason argues that both rule- and knowledge-based actions are vulnerable to mistakes rather than slips or lapses because they occur due to faulty cognitive processes that are consciously employed to make a decision.

Patient safety is compromised by both slips and mistakes. Nevertheless, medication administration errors are more often caused by slips, as nurses are not responsible for prescribing drugs, which involves conscious cognitive processes as is the case with rule- and knowledge-based actions. Based on this conclusion, the main objective of reducing the number of medication administration errors in a healthcare organization is to introduce technologies and strategies that will recognize and prevent skill-based errors that are classified as slips.  One such strategy is the use of drug labeling that minimizes the potential for confusing drugs with similar names.  Even so, these strategies are not enough. In the next part of this paper the effectiveness of bar-coding medication administration as a method to prevent medication administration errors in healthcare settings is discussed

Bar-Coding Medications as a Method to Prevent Medication Errors

Bar-coding technology or bar-code medication administration systems are one of the most reliable methods of preventing slips during medication administration. A barcode can be described as a symbol (a combination of dots, lines, numbers, and figures) that contains information about the particular product and can be read by a specially designed optical scanner. Bar-coding technology was patented in 1952 and, since 1974, has been used commercially. At first, bar-coding was used in grocery stores to mark products, but then the use of bar-codes spread to engineering, aviation and healthcare industries. Today more than one million companies use bar-coding to identify their products. The estimated number of bar-code scans each day is five billion (Fishman, 2008).

It was not until recently that bar-coding technology became common for preventing medication administration errors.  Koppel et al. (2008) refers to three methods through which bar-codes can enhance medication administration safety. The first is integrating the patient’s medication administration record with the process of administering medications, the second is insuring adherence to the “5 rights” of medication administration, and the third is enhancing the probability that nurses will use equipment protocols during medication administration. BCMA systems have a potential for reducing skill-based errors described by Reason (1990), as they reduce the amount of actions a human being needs to perform in order to administer a medication, and allow for double-check before giving medicine to the patient.

According to one of the Reason’s principles of error management, humans cannot be changed, but their working conditions can be modified (Reason, 2003). The main task of BCMA systems is, in fact, changing the working conditions of the hospital staff in order to reduce the opportunities for slips and lapses. The algorithm of using bar-codes for preventing medication administration errors is uncomplicated. Bar codes are placed on medications and patient-identification bands.  This allows the prescribed drugs to be matched with the patient’s identity so that administration of medications to the right patient can be monitored.  Cescon and Etchells (2008) argue that the bar-coding administration (BCMA) system includes several stages.  These are dispensing individually packaged bar-coded medications by the pharmacy, scanning s a patient’s wristband and getting a list of medications that have been prescribed and getting a list of warnings and instructions (usually performed by the nurse).  The following stages are administering the medications and logging all medications taken by the patient into an electronic medication record.  The BCMA system includes hardware and software parts. Hardware components are label printers, bar-code scanners, and robots dispensing the medications and software components include, for example, user interfaces, error-preventing mechanisms, and personal medication records. All of these components should work together in order to ensure the effectiveness of BCMA systems for preventing medication administration errors and ensuring patient safety (Cescon & Etchells, 2008).

Research indicates that using a bar-coding administration system prevents medication administration errors in approximately 1–2% of all medication administration attempts.  This percentage is consistent with a 50% reduction of medication administration errors (Sakowski, Newman and Dozier, 2008).  According to research, conducted by The University of Pittsburgh Medical Center, the reduction in medication administration errors may be as high, with an estimate of 55%. Fowler, Sohler & Zarillo, year) According to Fowler et al’s (2009) literature research, the effectiveness of using a BCMA system in a chain of twenty seven hospitals in northern California showed that medication errors were prevented in 1.1% of all medication administrations. The most frequent errors that were intercepted before reaching the patient were related to an attempt to administer a medication that had been cancelled or expired, attempts to administer the wrong dose of a drug, and attempts to administer a medication before the scheduled time. Therefore, it is obvious that those mostly were slip errors, defined by Reason (1990) as skill-based errors.

Moreover, research conducted by Fowler, Sohler and Zarillo in 2009 indicated that implementing bar-coding medication administration increased the level of satisfaction among hospital nurses. The nurses were most satisfied with access to the physicians’ prescriptions and orders, and the availability of other information that helped to insure that medications were administrate correctly.  Nurses were also satisfied that the BCMA system helped to protect the five rights of medication administration.  At the same time, the nurses complained that the required to administer medicines increased, resulting in less time to spend with patients. The nurses who participated in the survey also indicated that the bar-code system had little effect on the quality of their communication with pharmacists and physicians who prescribed the drug, as doctors and pharmacists often failed to use the two-way communication option available in the system (Fowler, Sohler & Zarillo, 2009).

The effectiveness of bar-coding medications and the types of errors it can prevent were studied by Sakowski, Newman and Dozier (2008),. Findings from this study indicated that, while 91% of near miss errors detected by bar-coding  would have minimal or no consequence, 8% led to moderate adverse effects, and 1% were judged as potentially life-threatening. Errors that involve administering high-alert drugs were also more likely to have acute health outcomes than errors involving other types of medications.

After the effectiveness of bar-coding for reducing the number of medication errors and enhancing patient safety were shown, the U.S. Food and Drug Administration (FDA) accepted a rule that requires bar-coding most prescription drugs and some over-the-counter drugs.  Drug manufacturers, labelers, private label distributors of unit-dose medications, and re-packagers are required to put machine-readable bar-codes on medications.  At the same time, there is no requirement for hospitals to scan the bar-coded medications. The FDA hopes that the hospitals provided with bar-coded medications will purchase scanning equipment and introduce scanning medications into their practice in order to enhance safety (Fowler, Sohler & Zarillo, 2009).

Problems with Introducing Bar-Code Technology

Despite of the advantages of bar-coding technology, it is not ideal. This part of the paper analyses the problems healthcare settings have encountered when introducing a BCMA system.. For instance, Koppel and colleagues identified fifteen workarounds in healthcare settings when they studied the implementation of bar-coding system. This was, for instance, scanning medication from patient’s drawer without performing visual examination of medication list, name, and dose. It may lead to administering the patient wrong medication, or wrong dose of drug. The doctors and nurses rely on computer system alerts, but without visual check slips are still possible. Another type of mistake is that physicians do not review patients’ current records before administering a medicine. Doctors and nurses may have impeded access patient records and, therefore, be unaware about medications that have been prescribed by other doctors, or may not stop unneeded medications or change the dose or route as intended. Sometimes nurses and other users administer medications without reviewing the complete guidelines for administration. The information on administering a specific drug may require viewing more than one screen. Nurses can also bypass the requirement of double confirmation of high-risk medications. Some DCMA systems are set up to require double confirmations of the patient’s name and dose for high-alert medications. User can override such confirmation in the system or confirm the administration using another nurse’s login data. In some cases, nurses fail to check and verify new administrations done by physicians. When a doctor prescribes a new drug, this entry appears in the system, and nurse has to verify it before administering the drug. If she does not check and verify it, she may administer a drug using the previously prescribed dose or  route.etc. Sometimes nurses administer drugs without scanning the patient’s wristband. The wristband may be missing or damaged, or the patient might be sleeping, therefore, nurses sometimes administer drugs without performing the scan. Failure to scan the wristband may lead to administering medications to the wrong patient. Another type of mistake is when nurses administer medications without scanning the barcodes. The barcode labels may be damaged or the scanner might be unavailable, so the nurse gives the medication without checking its barcode.  In this case, the nurse may administer the wrong drug or the wrong dose of medication.

Another common mistake is documenting administration of medications before the actual administration. Nurses are prompted to document administration right after scanning the medication, as it may take time to prepare the medication or access to a computer in the patient’s room may be problematic. Therefore, an unadministered medication can be recorded as administered. Some nurses scan patient ID barcodes when they are placed on objects other than the patient. The patients’ ID codes may be placed on the scanner machines, in the nurses’ pockets, on the doors of the patients’ rooms or elsewhere. Therefore, the opportunity exists to scan the wrong ID and, thus, administer the wrong medication.

In some cases BCMA users may prepare and carry medications for more than one patient at a time. Nurses scan the drugs’ barcodes and take them to the patient’s room in their pocket or on a cart. Even when the ID is scanned, therefore, the patient may get a drug that has been prescribed for another patient. Yet, nurse can scan medication labels after they are removed from the medication. Sometimes nurses open packages with several medications and put the labels in one place to scan them all together. Therefore, the patient may be given the wrong medication since the bar code is not scanned at the time of administration. At the same time, users can scan one label only when different medication packages are thought to be the same. This practice may result in administering the wrong drug or the wrong dose obtained from a package that looks exactly like the one that was scanned.

Sometimes, the scanner is located on the cart with medications. When the cart it is too large to fit in the patient’s room, the warning alarm that indicates that the wrong patient or the wrong medication is being administered may not be seen. In some cases users can give a partial medication dose but scan the label for the full dose. Therefore, the dose of the administered medication is seen as correct in the system but, in fact, is wrong. Finally, users can mute the warning alarm, so that error warnings are not heard (Koppel et al., 2008).

According to Reason (2003), each error has a history: it is a result of the chain of events that includes organizations, people, teams, tasks and organizational factors. Reason developed a number of recommendations for creating systems that would reduce the negative influence of human factor. Unfortunately, he overlooked the fact that people are likely to sidestep the equipment. Therefore, in addition to the mechanism of preventing slips and lapses, medication error preventing systems should include protection from sidestepping. These may include removing the possibilities to switch off the audio alarm on the cart, introducing the system that would register the time medication was administered, and inform the nurse’s supervisor and the patient’s doctor when regime is violated, or modifying the software so that it would inform the nurse’s supervisor that the medication was not administered, or administered with violations. Reason (2003) writes that error is rather the consequence than the cause, therefore, the best approach would be dealing with the causes that make nurses sidestep the rules of BCMA systems operation. For instance, as nurses complain that administering with BCMA takes too much time (Fowler, Sohler & Zarillo, 2009), efforts should be made in order to shorten the procedure, and make it more convenient. At the same time, nurses should be educated about the possible consequences of sidestepping the system, and taught to operate it efficiently.

Unfortunately, it is impossible to eliminate sidestepping by modifying BCMA software only; hospital staff should also be included in this process. In cases when the amount of slip errors is high despite of the presence of BCMA, human control over the medication preparation and administering process should be enhanced.

Conclusions

Medication errors are one of the leading causes of deaths in the healthcare setting. The estimated cost of medication errors is 2 billion a year, and their consequences may be disastrous both for patients and staff. Therefore, a nationwide goal to decrease the number of medication errors and enhance patient safety have been established.  In this paper, the problem of medication errors was analyzed through the framework of Reason’s Human Error Theory.  According to this theory, errors cannot be eliminated entirely, but equipment can be designed to minimize slips that result in errors during skill-based actions. Bar coding administration (BCMA) systems have emerged as one of the most useful tools for decreasing the number of medication errors.  Based on studies that indicate as much as a 55% reduction in errors through the use of BCMA systems, the effectiveness of bar coding medications is clear.  Moreover, introducing bar coding systems increases nurses’ satisfaction levels.

Nevertheless, these systems are not the perfect or final solution to the problem of medication administration errors since, as human beings, users can violate the guidelines for administering medications using the BCMA systems. Measures should be taken to reduce the number of infringements when bar-coding systems are used.

References

Aronson, J.K. (2009). Medication errors: what they are, how they happen, and how to avoid them. QJM: An International Journal of Medicine, 102, 513-21.

Cescon, D. W. & Etchells, E. (2008). Barcoded medication administration: A last line of defense. Journal of the American Medical Association, 299, 2200-2202.

Committee on Identifying and Preventing Medication Errors. (2006). Preventing Medication Errors: Quality Chasm Series, P.Aspden, J.Wolcott , J. L. Bootman & L. R. Cronenwett (Eds.). Wasington, DC: The National Academies Press.

Fishman, C. (2008). The killer app – bar none. American Way. Retrieved from http://www.americanwaymag.com/so-woodland-bar-code-bernard-silver-drexel-university

Fowler, S. B., Sohler, P., & Zarillo, D. F. (2009). Bar-code technology for medication administration: Medication errors and nurse satisfaction. Medsurg Nursing, 18, 103-109.

ISMP Institute of Medicine. (1999). To err is human: Building a safer health system. Washington D.C.: National Academy of Press.

Koppel, R., Wetterneck, T., Telles, J. L., & Karsh, B. (2008). Workarounds to barcode medication administration systems: Their occurrence, causes, and threats to patient safety. Journal of the American Medical Association, 15 (4), 408-423.

Reason, J. T. (1990). Human error. Cambridge, England: Cambridge University Press.

Reason, J.T., & Hobbs, A. (2003). Managing maintenance error: a practical guide. Hampshire, England: Ashgate Publishing.

Sakowski, J., Newman, J. M., & Dozier, K. (2008). Severity of medication administration errors detected by a bar-code medication administration system. American Journal of Health-System Pharmacy, 65, 1661-1666.

U.S. Department of Health and Human Services. (2001, June 22). Patient safety task force established. Retrieved fromhttp://www.highbeam.com/doc/1G1-79370081.html

Williams, D.J. (2007). Medication errors. The Journal of the Royal College of Physicians of Edinburg, 37, 343–346. Retrieved from http://www.rcpe.ac.uk/journal/issue/journal_37_4/Williams.pdf