Pre-mixed products cut admixture errors
Pre-mixed products cut admixture errors
HCFA dropping permissible error rate to 2%
A new study of admixture errors proves that ready-to-use products cut mistakes drastically, much more than system checks and balances currently achieve. The study, which found a 9% error rate, comes when the Health Care Financing Administration (HCFA) is about to lower the permissible error rate from 5% to 2% for hospitals receiving Medicare funding.
The observational study of five regional U.S. hospitals found a 9% overall error rate in the compounding of intravenous admixtures, using the American Society of Health-System Pharmacists’ (ASHP) 1993 Technical Assistance Bulletin (TAB) on Quality Assurance for Pharmacy-Prepared Sterile Products as a guideline.1 However, the error rate for ready-to-use products was a scant .3%.
The 9% error rate was well above the 5% standard HCFA adopted in 1984 to determine certification for federal Medicare and Medicaid programs. And the study comes just as HCFA is publishing its findings toward lowering that standard to a 2% error rate. HCFA is expected to push the Joint Commission on Accreditation of Hospitals to adopt the 2% standard for private hospitals as well as long-term care nursing homes. The regulation draft is awaiting release by the agency’s secretary, but at press time HCFA had not set a release date.
"We didn’t expect to find the rate of error to be that high," says the admixture study’s co-author, Kenneth Barker, PhD, head of the Department of Pharmacy Care Systems at Auburn (AL) University. "I think it’s reasonable to be alarmed and scared and concerned," Barker says of the results, published in the April 15 issue of the American Journal of Health-System Pharmacists. The researchers spent a week observing pharmacy operations at one hospital each in Mountain, Midwest, Pacific, Southeastern, and Northeastern parts of the country.
Barker notes that while compliance with 19 recommendations found in TAB was high overall (with 100% compliance found in 10 categories, dropping as low as 60% in others) that does not mean patients are out of danger.
"We found that, in part, following the rules didn’t seem to help much. TAB’s primary focus was preventing contamination, but our feeling is the emphasis on that was so great that not enough attention was given to other things that can go wrong, like simply giving a drug to the wrong patient. During the study, we saw no evidence of microbial contamination to give credit where credit is due but when nine of 100 doses have one or more things wrong with them, that’s still a problem," Barker says.
Among the study’s findings are that two of every 100 errors were deemed "potentially clinically important," based on HCFA’s significant medication error criteria for patient harm. Overall, 145 errors were found in 1,679 doses to reach the 9% error rate. (See chart, above.)
The study also supports recent alarm bells sounding for handling potassium chloride IV admixtures, where overall error rates ranged from 34.4% to 56.7% among both hospital technicians and pharmacists for that drug alone. The type of errors tracked included wrong dose errors the most common type found in the study along with the use of an unauthorized drug, wrong base solution, a product omission, or faulty preparation technique.
Products covered in the study included IV admixtures, antineoplastic preparations, parenteral nutrient (PN) solutions and ready-to-use preparations.The overall error rates ranged from 0 to 56.7%, depending on the drug and the use of manual, automated, or partly automated preparation. Again, the overall 9% error rate did not include ready-to-use products in the calculation. Error rates among the five individual pharmacies ranged from 6% to 10% overall, while the daily rate of compounding errors fluctuated from 2% to 24%.
PN solutions had the highest overall error rate of 37% for manual preparation and 22% for those that were partially automated. Errors involving automated compounding ranged from the omission of potassium phosphates solution or multi-vitamin products to the unauthorized use of heparin, potassium phosphates solution, or cysteine. (Only two errors in 746 doses using ready-made products were found by comparison.) Of the specific drugs followed, the error rate for potassium chloride was the highest. The study did not weigh in the amount of doses per given drug to effect the error percentage.
Although the administrations of the five hospitals studied were informed, the technicians, pharmacists, and nurses observed were unaware of the intent of observation, as researchers checked the compounding system and results against the labels used to prepare the doses.
The study also details differences in the workloads and systems used by the five hospitals. The preparation recurrence and amounts of patient-specific doses, the use of syringe pullback oversight and related methods, the use of manual or automated compounding, and the use of ready-to-use products were included.
"As a general rule, standardization is an approach to quality assurance," Barker says. "For instance, in airplane pilot training, there is a great deal of standardization, and every time there is an accident, the goal of standardization is increased. But the application of that principle has been poor in health care. A greater degree of variation has been tolerated than is probably justified in terms of patient safety. A pharmacist is wholly responsible for every professional act that he makes, which ignores the presence of environmental forces that may make it very difficult to avoid errors. [For example,] you can put a practitioner in a room with poor lighting, and the practitioner is trained to make the best of it, but the answer ought to be doing something about the lighting."
Throughout the study, two examples of an egregious, conscious error were noted. One involved a technician continuing to prepare an IV after pricking his finger with a needle, and the other involved five instances of technicians bypassing a safety check by scanning an empty vial through a bar code system instead of using a new, full vial to replenish the supply.
Awareness and recommendations
Barker does sound at least one positive note: "We think there has been dramatic increase in awareness in the last two or three years, based on studies going on like this one. But an even newer development is willingness to talk about it," he says.
Barker points to a pair of multidisciplinary meetings held in 1996, the first of their kind covering the problems of adverse medication errors. (See story on medication errors, p. 117.) Along with the ASHP membership, the meetings were attended by the AMA, American Nurses Association, and American Hospital Association, among others.
Barker’s own study points to one clear path for improvement, demonstrated by the difference between a .3% error rate for ready-to-use products, and 9% error rate for compounded admixtures. "There is an increasing demand for premixed solutions, and pharmacies would prefer them. But sometimes cost is a factor. At one very simple level, hospitals must either invest more to do a better job or get out of the business and outsource it to someone who can," he says.
"We were not surprised to find that the operation is a high-stress one because of reductions of staff across the country in pharmacies, in general, which does include the IV admixture area. You couldn’t help but be sympathetic to the pharmacist in charge, but we also found system errors, not just individual performance, and more attention needs to be given to the whole system for IV admixtures," Barker says. Increased training for technicians and the growth of automated dispensing machines also should be considered, he says.
With potassium chloride dosage errors topping the individual drug error category in Barker’s study, the drug’s importance for patient care cannot be outweighed by the care of its usage, especially in the wake of the harmful and even fatal dosage errors being reported.
Everyone agrees with that, but at issue is what hospitals are doing about it. "Potassium chloride error is the number one cause of serious patient injuries and fatalities in the U.S.," says Stacy Wiegman, PharmD, a fellow at the nonprofit Institute for Safe Medication Practices in Warminster, PA.
"That 5% [HCFA] standard is bandied about a lot, but the answer is that a zero error rate is the only rate acceptable," Wiegman says. She points out that with potassium chloride, it’s not only the dosage amount, but the speed at which an IV including potassium chloride is given. "There are a lot of premixed bags, or hospitals make their own compound. The danger is when it’s infused too quickly, that can cause cardiac arrest," she says.
Avoid special orders
Wiegman says an infusion greater than 40 milliequivalents (mEq) per hour is dangerous. She also argues that ready-mades including potassium chlorides are not overly expensive for hospitals, and many are using them. But the danger there, she says, comes when an order is sent that doesn’t comply with the premixed concentrations, say at 20 mEq, for example.
"We recommend that hospitals avoid special orders and develop a protocol saying you must prescribe a premade amount, and that has worked well at most hospitals doing it," Wiegman notes. She also recommends that hospitals do not add potassium chloride to an existing hanging bag, as the addition tends to pool around the infusion port, especially if the mixture is not vigorously shaken. Premixed bags again provide greater safety, she says. Also, while dosage protocols, premixing, and paperwork checks and balances help, the bottom line for safe handling is simply removing potassium chloride from the nursing floors, a move that is becoming imperative, she says.
Samuel Kidder, PharmD, MPH, a pharmacy consultant with HCFA in Baltimore, allows that the 5% error rate established in 1984 is as much political as practical. "We wanted to establish a limit on errors; it’s not scientific at all. We said we don’t want any errors, but provider groups argued the case on nonconsequential errors, so we set it at 5% with no significant errors allowed," he says. "Now we are beginning to push a 2% level that we don’t think is an unreasonable limit."
HCFA’s criteria for a "significant medication error" apply to drugs for which the toxicity level and therapeutic index are close, meaning the nature of the drug and the patient’s condition are factors. "We don’t pretend to be able to say how to manage a drug distribution system, just that here’s an error rate to work with. We can detect them via observation, so our role is to stress no errors and develop a protocol to stop errors. Managed care could help by imposing cost controls. We know that many preventable errors will result in four days of extra stay, and that’s up to $4,000, so that’s motivation on the bottom line."
Kidder also notes that an observational technique like that used in Barker’s study is the best and really the only way to determine the scope of dosage error rates. "Observation is the best technique, even if technicians find it intimidating. Self-reporting just isn’t working. The biggest factor is nurses often don’t know they’ve made an error, and when they do report them they get sanctioned, so they won’t report them," Kidder says.
And in the wake of Barker’s recent study and HCFA’s desire to tighten its certification limits, more pharmacists can expect to find someone looking over their shoulders.
[Editor’s note: For more information, contact: Kenneth Barker, PhD, Department of Pharmacy Care Systems, 128 Miller Hall, Auburn University, Auburn, AL 36849-5506. Telephone: (334) 844-5152. E-mail: [email protected].
Stacy Wiegman, PharmD, Institute for Safe Medication Practices, 300 West Street Road, Warminster, PA 18974. Telephone: (215) 956-9181. World Wide Web: www.ismp.org.
Samuel Kidder, PharmD, MPH, Health Care Financing Administration, 7500 Security Blvd., Baltimore, MD 21244-1850. World Wide Web: www.hcfa.gov.]
Reference
1. Flynn EA, Pearson RE, Barker KN. Observational study of accuracy in compounding i.v. admixtures. Am J Health-Syst Pharm 1997; 54:904-912.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.