Nurses improve medication administration accuracy

Many solutions are simple, low-cost

A group of seven hospitals in the San Francisco Bay area participated in an 18-month-long program designed to improve the reliability of medication administration by deploying nurse leadership and PI skills on a single med/surg unit. The results? Among the six hospitals that were included in the first analysis, accuracy improved from 85% to 92% in the first six months and to 96% 18 months after the intervention. The study was published in the Joint Commission Journal on Quality and Patient Safety.1

The study involved the implementation of the Integrated Nurse Leadership Program (INLP), which provides frontline nurses and other hospital staff with training, resources, and authority to devise and implement solutions. In addition to showing the importance of empowering frontline nurses, the most significant finding of the study is that "significant improvement in outcomes can be accomplished without very expensive fixes," says Julie Kliger, MPA, BSN, RN, INLP creator and program director at the Center for the Health Professions, University of California, San Francisco; principal and founder of The Altos Group; and lead author of the article.

"That's important because there are a lot of financial and resource pressures today, and people sometimes think they need to spend millions for bar-coding; this demonstrates through almost old-fashioned QI and engaging the people who are doing the work and providing them with tools, skills, and resources, that they can make statistically significant improvement."

Empowerment is critical

Why is it critical to empower frontline nurses to obtain such results? "I think it's essential to empower frontline clinicians because they are the ones doing the work," Kliger explains. "They know, see, observe, and live the problems, and have often thought about how to correct them — but are not typically put in the position where they can exercise that knowledge. When you give them the time, the resources, some tools, and support to do that in an organized framework and they are able to take what they intuitively know and funnel it in a framework and apply it, you not only get the right answers and solutions, you also get engaged workers."

That's why the INLP model is successful, says Kliger. "It weaves together both the technical framework we like to see in PDSA [Plan, Do, Study, Act], and it engages anyone who is closest to the issue. Having been a frontline nurse and been on the management end, I know you have to do that. As a frontline nurse I always had opinions, but the organizations were not set up to help me fix the problem. This model weaves together the frame and reliability, knowledge, understanding, using data, along with whatever outcome you are trying to advance."

In addition, she says, it weaves in "softer" skills, such as effectively communicating the key message. "Things will change depending on who you speak with, how well you understand the dynamics, your organizational savvy, and learning how to reach your environment," Kliger explains.

Customize your solutions

There is no "one size fits all" solution to medication administration accuracy, Kliger emphasizes; your solution must be customized to your facility. "Customization is very important because each unit is in itself a microsystem — even within a facility," she explains. "Its culture, attitude, flow, processes — they all function quite differently."

This is not to say, she continues, that the outcome or patient indicator can vary, but the solution to get there has to be customized. "You can't say the ED, for example, can have sloppier medication administration because it's hectic," Kliger asserts, adding that through variation, the changes are readily shared and become a "library" that can be compared across hospital and unit settings.

It also is this customization that led to creative and often low-cost solutions. For example, one of the hospital's solutions, which later was picked up by many of the others, was the notion of a communication tree. "One med/surg unit decided that too many nurses needed to communicate via e-mail, which was not very effective because not many people would actually read them," says Kliger.

If you picture a tree with branches, she continues, each member of the team had a name on a branch, and each leaf was the name of somebody they were responsible for calling. "Whenever there was a substantial change in policy, procedure, or you needed input from members of the team, 10 members reached out to 60-plus staff," says Kliger. "It was a more personal touch and made a big difference; somebody else downstream couldn't say they didn't know about the change. It also fostered buy-in."

Another solution addressed the problem of distractions and interruptions. "One of the units wanted to decrease interrupting phone calls during their protected hour of medication administration," Kliger explains. "They developed a process whereby the ward clerk identified who the caller was and read from a script telling how the nurses were involved in an important patient safety activity and couldn't be pulled out of it to talk."

The clerk then drew either a happy face or a frown face, indicating whether the customer was satisfied. "They amended the scripts over time and kept a log on whether there were more happy faces or frown faces," Kliger adds. "It did not cost any more money and directly determined if they were strategically communicating; people felt better about it. When someone absolutely had to be transferred, the charge nurse was brought up to speed on all the patients."

Ensuring compliance

The program also involved compliance tracking activities. For example, every month there were two small data measures of 10-measure sets. "Twice a month we would 'naively' observe 10 medication administrations," Kliger shares. "And every month we would have two data points — such as interruptions or labeling — and every month we'd see if the 'test of change' worked; that informed the next steps."

The goal, of course, is to keep the trend of errors on a downward slope. "We need to be at 90%-95%, so the goal is to strive that way," says Kliger. "When I visited with teams, they'd be reviewing how the data looked on each of the units. We developed an Excel-based electronic dashboard, which included raw data, percentages, and graphs, so the frontline staff could look at their progress, and at the governance level they can see an organizational error rate."

The method is apparently working. "We now have 36 months of data on the pilot units; all seven hospitals are either maintaining or improving their rates," Kliger reports. "And during that time we were spreading the program to all units in the hospital, adding complexity, which is hard to do."

The teams meet regularly, and it's important for them to have data to look at in order to sustain improvement, Kliger continues. "Data are like the headlights on a car," she notes. "If things are not working, then you draw on leadership skills and how you present the data; that's how you get people coming on board."

[For additional information, contact:

Julie Kliger, MPA, BSN, RN, Program Director, Integrated Nurse Leadership Program, Center for the Health Professions, University of California, San Francisco. Principal and Founder, The Altos Group. Phone: (510) 551-3330. E-mail:]


  1. Kliger J, Blegen MA, Gootee D, O'Neil E. Empowering frontline nurses: a structured intervention enables nurses to improve medication administration accuracy. Jt Comm J Qual Patient Saf. 2009; 35:604-612.