The trusted source for
healthcare information and
Project aids rural hospitals in drug error reporting
A powerful new resource improves safety
If your hospital is a large urban facility, you may have large numbers of staff with individual departments responsible for performance improvement, patient safety, and data collection and abstracting. Unfortunately, many quality managers at smaller facilities are not so lucky.
"At many facilities, the same individual is in charge of infection control, quality assurance, and performance improvement," says Katherine Jones, project director of the Omaha, NE-based Small Rural Hospital Medication Error Reporting Project. "Collecting and analyzing medication error rates is one of many tasks."
At Jefferson Community Health Center in Fairbury, NE, a single individual is responsible for quality improvement, risk management, worker’s compensation, and long-term care education. "I do not have a secretary or any other staff," says Sharon Vandegrift, RN, ONC, MNEd, the facility’s quality improvement/risk manager/patient safety coordinator. "At a critical access facility, it’s very common for one person to wear five or six hats."
Nebraska’s medication error reporting project gives solutions for the unique patient safety improvement challenges of small rural hospitals with 50 or fewer beds, and is a collaboration between the Nebraska Center for Rural Health Research, Nebraska Medical Center, and 11 participating hospitals. The project recently was expanded to allow enrollment for hospitals outside of Nebraska, Jones says.
For a $1,500 enrollment fee, the project provides a system of data collection, systems analysis, feedback, and workshop opportunities. "The problem is that nobody has been collecting information on the critical access hospitals," says Vandegrift. "We are on the cutting edge of collecting these data to share with other facilities."
Here are key benefits of the program:
Small rural hospitals with fewer than 25 beds and an average daily census between two and 15 face a daunting challenge — even those with the best voluntary, nonpunitive error reporting systems, according to Keith J. Mueller, PhD, the project’s principal investigator and director of the Rural Policy Research Institute’s Center for Rural Health Policy Analysis in Omaha.
The project aggregates data across hospitals of similar size and operation so that a larger pool of medication error data is created in less time. Currently, the database has almost 1,000 medication error reports representing two years of reporting from eight different hospitals.
As a critical access hospital, Jefferson Community has an average daily census of only 2.3 patients per day. "It would take an awfully long time for any of us to collect the information by ourselves," Vandegrift explains. "But with the hospitals all grouped together, it helps us function like a larger facility."
The information has shown that the biggest problem in the facilities overall was omissions, she says. "When reviewing the data with our staff here, we realized we weren’t using some of the best practice guidelines out there," she says, referring to recommendations from the Agency for Healthcare Research and Quality and the Joint Commission on Accreditation of Healthcare Organizations’ 2004 National Patient Safety Goals.
To decrease actual errors while increasing near misses that are caught before they reach the patient, the facility’s policy was changed to include the "five rights of medication administration." That is probably one of the biggest changes we have made," Vandegrift says.
The program assumes the burden for creating a database, managing the data, performing analysis, and generating reports. This is a major obstacle for rural hospitals, Jones notes. "You need to find somebody who knows how to put it together, maintain it, and be able to extract information from it. That’s not a skill you learn in nursing school."
The program also gives the facilities the ability to submit data to MedMarx, a national medication error reporting database. This gives the hospital the ability to look at its own data and compare them with a nationwide peer group, she adds. "We plan to dump data periodically. On their own, it isn’t something they would necessarily think to do."
The program is working with MedMarx to determine the effect of staffing mix on error reporting, says Jones, adding that critical access hospitals often lack even a single full-time equivalent (FTE) pharmacist. A recent survey conducted by the University of Washington, shows critical access hospitals have an average FTE of 0.67 pharmacists on staff, and 38% contract with a consultant pharmacist to provided limited on-site services.1
"When you don’t have pharmacists participating in the medication dispensing process, you are removing a reporter who generally catches errors upstream before they reach patients," says Jones, adding that possible solutions include faxing orders to an off-site vendor pharmacy practice that reviews them at night. "We are exploring different interventions to broaden the participation of pharmacists in the dispensing process."
Participation in the program gave staff a golden opportunity to have their current system evaluated and compared with other participating hospitals. "All of a sudden, it wasn’t one QA nurse in a small rural hospital saying we need to change — we had input from a professional team and our peers. The feedback was invaluable," says Judy Glynn, RN, a nurse at Pawnee County Memorial Hospital, which has a staff of 12, including the director, nine nurses, and three LPNs.
Previously, staff were reluctant to report drug errors. "Reporting was not viewed as helpful, but a requirement to cover liability," Glynn explains. "There was no tracking of the errors we prevented from reaching the patient."
Other than summarizing the findings of each report, documenting its review, and calling staff’s attention to ways to avoid errors in the future, nothing much was done, she says. "In general, people hated to make an error, hated to have to make a report and, if minor, sometimes asked the doctor to cover with an order to avoid having to make out an incident report."
When Glynn discovered that 50% to 75% of drug errors involved a medicine card in some way, nurses voted to change this practice by using the medication administration record (MAR) to set up, give, and chart medications. Staff were impressed when she showed them a graph with the dramatic results: Errors that reached the patient decreased from 75% to 20% after the new MAR system was implemented. "Near-miss errors are presented to the staff as a compliment on how well they prevent an error from reaching the patient," she says.
Another key change that was made is encouraging staff to double-check high-alert drugs such as intravenous antibiotics with another nurse before administration. "This list of high-alert drugs is posted in the med room with an invitation to add to the list," she says. "For example, glucophage was added when it was involved in three of our five errors in February."
[For more information, contact:
1. CAH/FLEX National Tracking Project. Health workforce recruitment and retention in critical access hospitals. Findings From the Field 2003; 3(5). Web site: www.rupri.org/rhfp-track/results/vol3num5.pdf.