Share this article on:

Discussion & Recommendations: Safe Medication Administration: An invitational symposium recommends ways of addressing obstacles.

Lewis, Laurie

AJN The American Journal of Nursing: March 2005 - Volume 105 - Issue 3 - p 42-47
State of the Science on Safe Medication Administration: Discussion and Recommendations

On July 16 and 17, 2004, more than 40 nurses and representatives from medicine, pharmacy, advocacy, industry, and consumer groups gathered in Philadelphia for an invitational symposium, “The State of the Science on Safe Medication Administration.” The symposium was supported by grants from the Agency for Healthcare Research and Quality (AHRQ 1 R13 HS14836-01) and manufacturers of products that promote safe medication administration (see the back cover for a list of sponsors).

After listening to several presenters whose articles are published in this supplement to AJN and the Journal of Infusion Nursing, the participants formed six small groups and spent several hours discussing barriers to safe medication administration and ways to overcome them. Specifically, they focused on the following questions:

  • What are the gaps in the research on safe medication administration?
  • What are the barriers to preventing medication errors and to ensuring safe medication administration? Consider those in nursing practice, management and administration, education, and policy.
  • What are the three most significant barriers?
  • What can be done to overcome, reduce, or eliminate them?

Each group attempted to come up with creative approaches to each barrier, listing three ways the problem has been customarily approached and then five possible new approaches. From this list of eight, the group selected what participants considered the three best strategies to address each problem.

At the symposium’s end, a plenary session convened in which participants reported on the small groups’ work. Each group presented its list of the three most significant barriers and the best—often new—ways of addressing them. After discussing issues raised at the symposium, participants left with a charge to work with nurses on improving the safety of medication administration.

After the symposium, a list of research priorities was generated. The list was then sent by e-mail to symposium participants for prioritization. Table 1 on page 5 of the executive summary lists these priorities for future research. (All tables mentioned in this report appear in the executive summary, starting on page 4.)

Back to Top | Article Outline


Each group listed barriers to safe medication administration as reflected in the following five domains: research, education, practice, policy, and administration. Each of the six groups focused on practice and on one other area from among the remaining four.

The groups identified numerous barriers in these areas (see Table 2, page 6). Some overlap was seen; for example, participants cited “research that does not translate readily into clinical practice” as a barrier to both research and to practice.

Back to Top | Article Outline

Barriers: research.

Participants observed that research outcomes are not always immediately applicable in clinical practice. One reason, they said, is that caregivers are not involved in determining research priorities. Research is most likely to be undertaken by academics and policymakers—people often far removed from clinical practice. On the other hand, bedside nurses may not understand how to translate the clinical issues into research questions. Even when research does result in clinically applicable outcomes, participants noted, there often is a considerable time lag between research and the application of its results.

One important topic for research was identified: finding the weaknesses in the health care system that increase the likelihood of medication error. Do systems on all units and shifts support safe medication administration by all staff, from new RNs to floats to travel nurses? Does the medication error–reporting system provide good information for remediation? Are there patterns of practice in certain clinical areas that greatly increase the risk of error?

There is little research on technology and whether it actually reduces medication errors. How is technology used in clinical practice? Which specific technologic innovations are best, and in which settings? Are new types of errors created when health care professionals depend upon technology to ensure safety?

Insufficient funding is a major concern of nurse researchers wanting to investigate medication error. With increased private and federal funding, research into some of the other gaps in research identified by symposium participants could be undertaken, including the role of patients and families in safety, why errors are not reported, and the economic impact of preventing medication error.

Back to Top | Article Outline

Barriers: education.

Barriers involving education at all levels have an impact, symposium participants observed. The problem begins in nursing school, where the curriculum does not present a systems approach to patient safety. Educators in nursing schools stated that with so much information to impart in so short a time, they must make difficult choices about what to teach and what not to teach.

And the barriers in education continue into clinical practice. Nurses are so busy, especially with short staffing, that finding time to learn about new technologies is difficult. Budgets do not allow for enough education. With multiple shifts and staffing by float nurses, some personnel miss in-service trainings and other opportunities.

The problem isn’t just what is taught to whom, but how it’s taught. What’s the best way to present new practices to nurses? What’s the best way for educators to help novices make the transition from administering medications as a nursing student to administering them as a licensed RN? Which information will result in the best practices? How are new skills incorporated into practice without the loss of valuable old skills? A case in point is dosing calculation. Nurses may rely on technology to calculate doses, but what happens when the equipment isn’t working? Nurses must be taught how to calculate doses without technologic assistance in case of equipment failure.

Finally, nurses do not operate in a vacuum where safety is concerned. They need to understand the entire medication process, from manufacturing and labeling to prescribing, dispensing, and administering, when attempting to create a safer environment for patients.

Back to Top | Article Outline

Barriers: practice.

A topic that was widely discussed during the symposium was the distinction between current practice and best practice. Who determines “best practices”? Research is not always available to delineate the standards of best practice, and nurses must consider a range of sometimes conflicting expert opinions, as sometimes comes from various professional organizations. Furthermore, given the role of technology in safe medication administration, best practice is a rapidly evolving concept.

Without such guidelines there can be no standardization, and a lack of standardization also is a problem in information technology, drug delivery systems, policies, and procedures. But as one nurse asked, is there any documentation of the effectiveness of standardization?

When discussing practice barriers, nurse participants voiced a common frustration: their heavy workloads make it difficult for them to incorporate safety processes into their daily work. An example is the time needed for “medication reconciliation” when a patient is transferred from unit to unit (see “Medication Reconciliation,” page 31). Nurses also cited nonnursing duties that divert attention and time from the focus on safety. The work environment often poses obstacles to safety. An example of an improvement strategy is basing work flow on patient acuity so that nurses can concentrate on the patients in greatest need of care.

Nurses, other clinicians, and administrators agreed that poor communication among providers is a major hindrance to patient safety. Interdepartmental discord could be decreased, symposium participants said, with interdisciplinary teams.

Presentations at the symposium emphasized the low reporting rate of medication errors, which results in lost opportunities for improving practice. Better relationships among clinicians might make nurses more willing to question other professionals when they see an error about to occur, and they might be less reluctant to report errors for fear of being belittled by the prescriber or pharmacist. The support of coworkers and administrators for those involved in an error also might improve reporting rates.

Symposium participants said they would like to see a greater connection between research and practice. They wondered which educational approaches would best change behavior and practice, and they noted the need to make sure nurses have all the information they need to provide the best possible care when technology fails.

Back to Top | Article Outline

Barriers: policy.

The subject of policy involves not only standard procedures within facilities, but also policy-related national quality measures. The biggest barrier to medication safety seems to be that policies aren’t set by the people closest to clinical care. Competing groups tend to set their own policies (for example, the Centers for Disease Control and Prevention, the Food and Drug Administration, the Joint Commission on Accreditation of Healthcare Organizations, the National Quality Forum, professional groups like the ANA and the American Medical Association, specialty organizations like the Infusion Nurses Society and the Association of periOperative Nurses, and consumer advocacy groups like the National Patient Safety Foundation), and their interests and agendas may not agree. Symposium participants asked how academics, regulators, and institutional policymakers find out what happens at the bedside.

As it affects practice, insufficient standardization also affects policy. Policies differ from one institution to another and from one state to another. Even terminology is not standardized. For example, is a near miss an error that should be reported? And even when policies exist, inadequate enforcement may lead to lapses in safety. Huge discrepancies can exist between written policy and what actually happens in clinical practice.

Back to Top | Article Outline

Barriers: administration.

These are most evident in the lack of support for a “just culture of patient safety”—that is, a nonpunitive environment that promotes the reporting of errors and correction of error-prone situations. Administrators may give lip service to patient safety, but they must back up their words with action: putting systems and structures in place that facilitate safer practices. Administrators should distinguish between errors caused by systems failures and those caused intentionally or through negligence.

Like policymakers, administrators often are remote from clinical practice. They need to keep up with the latest standards and research, as well as actual practice in the facility. They should make it possible for practicing nurses to be involved in making decisions on new technology options and other safety initiatives.

The way administrators allocate funds is another factor affecting patient safety. Funds should be earmarked for safety initiatives, distributed equally between technology and staffing.

Back to Top | Article Outline


Each symposium participant ranked the barriers identified, selecting the three most important ones for further deliberation. They discussed strategies for addressing the barriers, both standard responses and innovative approaches, and voted on the three best strategies for each barrier.

The groups often identified similar barriers, but they used different language to describe the problems. Sorting through the terminology in the 18 barriers reported in total by the six groups revealed certain themes. In the end, seven significant barriers to safe medication administration emerged:

  • the failure of institutions to create a just culture of safety
  • a lack of interdisciplinary collaboration and communication
  • a working environment for nurses that does not emphasize safety
  • the absence of front-line nurses in decision making and system design
  • the difficulty of applying research to practice
  • policy that’s not driven by evidence
  • insufficient funding for research

Numerous strategies were recommended to address these barriers (see Table 3, page 8). The lists of strategies revealed that the five areas of focus (research, education, practice, policy, and administration) must be considered in addressing barriers to safety in medication administration. For example, a working environment for nurses that emphasizes safety depends on the support of administrators and policymakers, on research that demonstrates the benefits and cost-effectiveness of safety initiatives, and on a public educated in safe health care. Rather than isolate research, education, practice, policy, and administration barriers and strategies, the following discussion examines common topics and recommendations that emerged during the symposium.

Back to Top | Article Outline

Working together to create a culture of safety.

A term heard frequently throughout the symposium was “culture of safety.” Although they may have had in mind different definitions, participants said that the lack of a culture of safety is a major barrier to safe medication administration. They emphasized that safety must be addressed in a just and fair manner, without adverse consequences for practitioners who commit errors unintentionally or who report errors others have made.

Many participants cited lax reporting of errors and near misses as an example of a deficient culture of safety. Is this because of a lack of agreement on what constitutes an error? Is it because safety is not emphasized in education and, consequently, ensuring safety isn’t considered a responsibility? Is it because nurses are afraid of reprisal if they report errors? Is it because nurses, physicians, and pharmacists do not work well together?

Another theme of the day was the general lack of standardization in health care. Terminology isn’t standardized (there’s no consensus on what constitutes an error, for instance, or on what particular abbreviations mean). Procedures and policies aren’t standardized, either within institutions or across them. Who reads bar codes? How should they be read? Which information is contained in them? If patient safety is to improve, multidisciplinary groups—which must include nurses—are going to have to standardize policies, terminology, and procedures.

Back to Top | Article Outline

Giving voice to nurses and consumers.

A concern expressed repeatedly during the symposium was the limited decision-making power of clinical nurses in ensuring safe medication administration. Nurses must be equal participants in multidisciplinary safety initiatives, research, policymaking, and the design and implementation of new systems. Administrators should give nurses time away from direct care to participate in such activities. As one group of participants noted, nurses are the only health care workers who are expected to undertake such activities on their own time.

Another voiceless group is health care consumers. Agreeing that patients and families should be more involved in matters of health care safety, symposium attendees called for educating patients and giving them representation on policymaking teams. Patients can be strong advocates of safety with, for example, legislators and the media.

Back to Top | Article Outline

Basing policy and practice on research evidence.

Optimal 21st-century health care is evidence based, and nurses should be major figures on research teams. More research must be conducted on the effects of safety initiatives on outcomes and on the cost-effectiveness of safe medication administration.

Humans can’t be used as guinea pigs in research on an issue as critical as patient safety; “simulation research” can demonstrate how new approaches will work before they are instituted. No policy should be written or implemented until its value and safety are demonstrated clearly.

The time elapsing between the initiation of research and the validation and implementation of findings is another issue. Symposium participants recommended using computer technology to give nurses, other clinicians, and policymakers immediate and easy access to the latest data related to safety. The development of a compendium of databases related to safety is one way computer technology can be used to disseminate information on safety. Another is to provide patients with “smart cards”— cards that contain electronic data on an individual’s health history.

A lack of research funding is a common lament. Symposium participants came up with some innovative approaches to securing more funding. One suggestion was to earmark a portion of government reimbursements for safety research. Unexplored sources of funding include businesses, philanthropies, and consumer groups.

Back to Top | Article Outline


Patient safety, like all aspects of health care, is rooted in science. Research is necessary to lay the foundation for policy and practices that will ensure safe medication administration. Nurses need to participate in safety-related studies and to have a voice in setting policy and practice guidelines. Ongoing education of nurses and other providers must emphasize the importance of safety.

Barriers to safe medication administration do exist, but they can be broken down by a variety of strategies. Collaborative working arrangements will help to create a culture in which safety comes first and nobody is penalized for pointing out safety concerns. Ensuring safety is a joint endeavor requiring the active participation of nurses, other providers, administrators, policymakers, educators, and consumers.

Back to Top | Article Outline

The Clarian Safe Passage Program

Increasing safety at Clarian Health Partners in Indianapolis.

By Kathy Rapala, JD, RN

Kathy Rapala is director of risk management and patient safety at Clarian Health Partners in Indianapolis, IN. Contact author:

The Clarian Safe Passage program is an information-management program that supports the dissemination of patient safety information among staff members within a facility. This program is integrated into nursing practice through the American Association of Critical Care Nurses (AACN) Synergy Model of patient care. In the AACN’s model of care, the ultimate patient outcome is “safe passage” through the health care system; hence the name of Clarian’s program.

Safe Passage seeks joint solutions to safety problems. Representatives from nursing, pharmacy, clinical engineering, laboratory, radiology, and the ED serve on a multidisciplinary team. The goal is to have a Safe Passage representative in all clinical and nonclinical areas. These representatives attend monthly Safe Passage meetings, as well as “local” monthly meetings in each department, where participants share information and identify and solve patient safety issues.

Because of the multidisciplinary nature of the program, staff are no longer isolated in their own work areas. For example, clinicians are not always adept at describing the scenario surrounding a piece of malfunctioning equipment. And clinical engineers don’t necessarily think of a patient’s connection with the machine. By working on a multidisciplinary team, representatives can look at all facets of safety issues.

The initial curriculum of the Safe Passage program, developed in conjunction with the Indiana University School of Nursing, includes general patient safety principles, emphasizing the complexities of the work environment. Understanding this environment is essential for practitioners to improve policies and procedures, introduce technology, analyze patient events, and prevent gaps in safety. Staff from both the Indiana University School of Nursing and Clarian provide the training. In addition, Safe Passage representatives receive patient safety intuition training developed with Klein and Associates in Dayton, Ohio, a group that specializes in cognitive task analysis. This training helps representatives analyze how events unfold and how to plan to prevent future difficulties. Goals of the Clarian Safe Passage program include

  • providing staff with a patient safety knowledge base, including the most current patient safety information available.
  • providing “vertically integrated” communication (among all levels of staff and Clarian board members).
  • preventing errors by planning for change and identifying gaps in safety.
  • creating a mechanism to analyze and learn from errors. • increasing work efficiency and effectiveness (for example, by standardizing work processes and eliminating duplication).
  • providing a mechanism for the improvement of care through evidence-based practice.


By implementing the Safe Passage program and incorporating it into the organization’s goals, Clarian has made a commitment to a culture of safety. This includes providing time for staff members to create the infrastructure for the Safe Passage program and allowing them to leave their units to study and solve patient safety problems.

The Safe Passage program helps the institution meet regulatory and evidence-based practice standards, such as the National Quality Forum–Leapfrog measures for the culture of safety. The culture of safety is usually regarded in relation to the patient—that is, it should prevent medical errors and hazards in the environment and be conducive to sharing information about errors. The program also creates a safe environment for the nurse by identifying and solving workplace hazards, reducing the chance of error, and simplifying the work process.

Back to Top | Article Outline

The MEDMARX Database

By Rodney Hicks, RN, MSN, MPA

Rodney Hicks is a research coordinator at the Center for the Advancement of Patient Safety, U.S. Pharmacopeia, in Rockville, MD. Contact author:

In 1999 subscribers of MEDMARX, the country’s largest anonymous Internet-accessible subscription database for reporting medication errors, reported 6,224 such errors. There were 41,296 reports of medication errors in 2000, 105,603 reports in 2001, 192,477 reports in 2002, and in 2003, the most recent year for which data have been analyzed, there were 235,159 reports. The increasing numbers do not reflect an explosion of medication errors but rather greater participation in MEDMARX’s voluntary reporting system. In 1999 only 56 hospitals subscribed to MEDMARX. The number increased to 184 facilities in 2000, 368 in 2001, 482 in 2002, and 570 in 2003.This growing participation demonstrates that health care professionals are increasingly willing to share information about medication errors.

Most of the errors reported to MEDMARX did not result in harm. The percentage of harmful errors was 2.8% in 1999, 3% in 2000, 2.4% in 2001, 1.7% in 2002, and 1.53% in 2003. Medication errors were associated with three deaths in 2000, 14 deaths in 2001, 20 deaths in 2002, and 24 deaths in 2003. Analysis of the 2002 data revealed that geriatric patients were especially vulnerable to harmful or fatal medication errors.

Vigilance on the part of health care providers has prevented many errors from causing harm. In 2002 35% of errors were intercepted by a practitioner and did not reach the patient; however, 49% of errors did reach the patient. In 2003 45% of errors reached the patient.

An examination of the five years of MEDMARX reports shows that about 75% have involved four types of errors: an omission, an order involving an improper dose, a prescribing error, and administration of a drug that was not prescribed. Together, omission errors and improper doses account for about half the errors (Figure 1, above right); however, wrong administration technique tends to be responsible for the highest percentage of harmful errors. Although wrong administration technique occurred in only 1.4% of cases in 2002, it was the type of error most likely to harm the patient. Of errors that resulted in harm, 6.2% involved wrong administration technique. The next most common type of error that harmed the patient was the wrong route of administration, involved in 2.6% of harmful errors. These patterns repeated in 2003. Wrong administration technique was seen in 1.3% of cases in 2003, but was associated with 6.1% of harmful cases, and wrong route was reported in 1.5% of cases overall, but was associated with 2.2% of harmful cases.



Not surprisingly, high-alert medications are most likely to be involved in errors that harm the patient. The biggest offenders in 2002 and 2003 were insulin (8.1% of harmful errors in 2002 and 8.7% in 2003), morphine (5.5% and 5.3%, respectively), heparin (4.6% and 3.9%), injectable potassium chloride (3.0% and 2.9%), warfarin (3.0% and 2.6%), hydromorphone (2.8% and 2.3%), and fentanyl (2.2% and 2.6%).

Performance deficit, which MEDMARX defines as an error that cannot otherwise be explained, occurs when a trained professional, who has the skills and knowledge to safely discharge his or her duty, fails to do so in a particular instance. This is the most common cause of error, accounting for 38% of medication errors in 2001, 46.6% in 2002, and 38.2% in 2003. The next most common cause is failure to follow a procedure or protocol, responsible for 20% of errors in 2001, 29.4% in 2002, and 18.3% in 2003. These factors also are the biggest causes of harmful errors. Performance deficit and failure to follow a procedure or protocol together were responsible for 76% of the harmful errors in 2002 and 55% in 2003. Absent, confusing, or intimidating communication accounted for 17.7% of harmful errors in 2002 and 13% in 2003.

MEDMARX is one project handled by the United States Pharmacopeia, a nonprofit, nongovernmental organization that sets standards for drug products. To learn more about medication error reporting, visit To learn more about MEDMARX, visit

© 2005 Lippincott Williams & Wilkins, Inc.