QI project improves best practice use tenfold
September 1, 2002 8 minutes read
Collaboration across departments, institutions
Collaboration across multiple hospital departments and several institutions helped a hospital network in Massachusetts increase its use of medication best practices by 10 times in only 18 months. From the outset, the project was intended to produce changes that could be implemented rapidly, rather than focusing on long-term data collection.
Reducing adverse drug events was the overall goal of the quality improvement project launched by CareGroup, a network of six hospitals in eastern Massachusetts. The Massachusetts Coalition for the Prevention of Medical Errors and the Institute for Healthcare Improvement identified 17 best practices that could reduce medication errors, so the network decided to find a way to increase their use throughout the network. CareGroup has 13,000 employees, including 2,000 medical staff.
Implementing such changes across a network is less common than making the changes within a single institution, notes Dottie McCarthy, MS, RN, CS, medical surgical clinical nurse specialist at Mount Auburn Hospital in Cambridge, one of the CareGroup facilities. But when the CareGroup leaders launched the quality improvement project, they decided to take advantage of the network’s size. Collaboration and input from a variety of sources turned out to be key to the project’s success, she says.
Numerous leaders were needed throughout the network, so each facility CEO selected a leader for that hospital’s "medication reliability team." Those leaders came from a variety of disciplines, including peer review, risk management, pharmacy management, and staff physicians. Physician leadership is desirable but not necessary, McCarthy says. The teams formed in 1999, and CareGroup hired an outside consultant to train them in rapid-cycle techniques for their meetings, to emphasize quick implementation of change rather than a long discussion. Four teams had eight members, and the other two had four members and five members.
The size of the teams affected how they acted, McCarthy says. Larger teams tended to go right ahead and implement changes without testing them first, and they also tended to become fixated on hospital policies they couldn’t change.
The teams met first on Feb. 8, 1999. Rapid implementation became a central focus of the project, says Lisa Saubermann, RPh, manager for pharmacy quality assurance and process improvement at Beth Israel Deaconess Medical Center in Boston, another CareGroup hospital.
Data collection took a back seat to implementing change, she says. Getting everyone’s perspective was one of the first priorities, she says, because the multidisciplinary team members sometimes did not fully understand each other’s roles in administering medication.
Each team mapped out the members’ roles, Saubermann explains.
Data collection was an ongoing concern, with the teams monitoring issues concurrent with the changes. The teams used a variety of sources to collect data, such as incident reports and the pharmacy’s computer system. Some data collection was farmed out to patient care areas, McCarthy says, sometimes with team members hanging around a unit and listening to what nurses told patients. A local college student was recruited to help analyze some of the data.
"We didn’t have tons of data. We would trend things out when concentrating on a particular issue, but we weren’t sending out tons of data," Saubermann says. "We didn’t want to get so stuck on that aspect of it that we lose sight of where we’re trying to go. I think some people get paralyzed by all the data. We were more focused on getting changes implemented rather than gathering data just for the sake of gathering data."
The teams met together at one of the hospitals each month. Each team had to present its definition of "error" to make sure everyone was talking about the same thing, and each team presented a detailed flowchart of all the steps in providing medication for patients at their facilities. They also presented assessments of how their institutions were complying with the best practices so far and what steps were being taken to improve compliance. But more than anything else done at the meetings, the project depended on the teams from all six hospitals sharing information about actual medication errors and what they learned from them.
"This wasn’t so easy, because no one likes to share a lot of information about their mistakes," McCarthy says. "Initially, we heard a lot about near misses because it was safer to tell those. But as we went along, people got more comfortable and passed on more information."
To assess how the teams were improving and moving toward the best practices, they developed a scorecard to track their progress. The initial scorecard in May 1999, three months into the project, indicated that potassium chloride had been removed from the patient floors in all six hospitals, but that was the only best practice that had been fully implemented.
The consultant working with the groups helped them analyze the available information, both during and after the meetings. One of the first targets for improvement was heparin administration. The task force identified heparin as a high priority because it is prone to error, requiring titration and frequent recalculation of dosages to prevent over- or undercoagulation.
To address that issue, the task force created a new dosing schedule and revised the form used to order the drug. It also reduced the heparin bolus dose and infusion rates, while also improving the documentation forms.
After implementing those changes, the team members analyzed charts to see the effects. Saubermann says the analysis revealed that the new policies improved the quality of care while, presumably, reducing adverse events. The percentage of patients who achieved a therapeutic level within the first three days of heparin treatment more than doubled, to 93%. The percentage of patients reaching the therapeutic anticoagulation range by the end of the first day doubled from 16% to 31%.
Look-alike, sound-alike drug events
Another focus was the problem of look-alike and sound-alike drugs; four adverse events had been traced to this problem at CareGroup hospitals in the previous six months. The teams came up with an unusual way to address the problem.
"We held a contest to see who could identify the most look-alike or sound-alike drugs, with all the different patient care areas competing against each other to identify the most," McCarthy says. "We came up with more than 100 and posted a list of all of them in the medication rooms. It raised awareness quite a bit, more than if we just handed out a list and told them to be careful."
Any look-alike drugs stored next to each other were rearranged or removed, and when possible, the hospital switched to a different supplier with different packaging. The pharmacy also placed special warning stickers on some look-alike drugs considered high risk.
For sound-alike drugs, CareGroup programmed its Pharmacy Drug Data File to show indications when a nurse calls up the medication to place an order. The system shows the common indication for the drug and prompts the nurse to confirm that it is the correct medication for the patient’s needs.
Other forms were standardized and then computerized later in the process.
"There’s a lot of talk about computerizing these days, and that was one of our goals, but we had to get it right on paper first," Saubermann says. "If you don’t have it right on paper first, computerizing it will just compound those problems."
Forcing functions also turned out to be key to reducing adverse drug events. Reducing the reliance on a clinician’s memory always will help improve medication safety, Saubermann says. To that end, the CareGroup teams looked for ways to "force" safer practices by standardizing forms, for instance. With heparin, order forms were changed to make a certain type of stat test the default rather than relying on the physician to actually write in that order each time.
"We’re not taking away the physician’s ability to make his own decisions, but that’s an example of an order that every physician wrote if he or she remembered to write it," Saubermann says.
"The change just made it the default, and if they wanted to change it, they had to actually write in that change," she says.
Saubermann and McCarthy say the task force didn’t follow the typical quality improvement rules too strictly if that would have meant slowing the pace. A hospital’s team often initiated several changes at once. The downside to that approach is that it can be difficult to determine what change was responsible for the improvement. By changing many variables at once, the teams sometimes could not point out which one was effective, or most effective.
"The team leaders did not see this as a problem, or at least not a big enough problem to slow down the improvements," McCarthy says.
"Protecting patients was the goal, even if it meant we wouldn’t have the rock-solid data that you might need for publishing study results and getting a lot of credit for the idea," she adds.
The results are hard to argue with. Among the six CareGroup hospitals, adoption of the adverse drug event best practices increased tenfold in 21 months.
At the beginning of the project, one best practice was adopted fully by all six of the hospitals; and less than two years later, there were 60 best practices in place through the network. (Not all of the best practices were in place at all the hospitals, so the 60 is a compilation of how many were in place at each site. The average works out to about 10 of the 17 best practices in place at each hospital.)
Almost all the changes involved no additional expense. Only the standardization of patient-controlled analgesia pumps required additional funds.
"The commitment at all levels was a key to making this work. We had people discussing this project at every level, at all kinds of meetings throughout the system," McCarthy says.
"We were constantly feeding information back to the staff so they knew that this was something we were serious about, not just a problem of the week," she adds.. "We emphasized that this was an important patient safety issue, not just a problem we’re focused on now but will forget next week."
[For more information, contact:
- Dottie McCarthy, MS, RN, CS, Medical Surgical Clinical Nurse Specialist, Mount Auburn Hospital, 330 Mount Auburn St., Cambridge, MA 02138. Telephone: (617) 492-3500.
- Lisa Saubermann, RPh, Manager for Pharmacy Quality Assurance and Process Improvement, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215. Telephone: (617) 667-7000.]
These are the 17 best practices for reducing adverse drug events, developed by The Massachusetts Coalition for the Prevention of Medical Errors and the Institute for Healthcare Improvement:
- Chemotherapy protocols and preprinted orders
- Weight-based heparin protocols
- Enforcement of standardized prescribing
- Removed potassium chloride concentrates from all units
- Patient partnering
- Colored allergy wristbands for patients
- Limited numbers of dosages and pump types for IV solutions
- Computerized drug profiling
- Unit dosing
- Pharmacy-based admixture of IV solutions
- Pharmacist on unit at rounds
- Warning labels for look-alike and sound-alike drugs
- Emergency department condition management protocols
- Maximum 12-hour shifts for all nursing personnel
- Critical pathways for complicated care
- 24-hour-a-day pharmacy service available
- Nonpunitive error reporting
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content