PHILADELPHIA—Sixty-six percent of antibiotic prescriptions were overdosed in older patients with Stage 4-5 chronic kidney disease (CKD), according to research from Ontario, Canada, presented here at the American Society of Nephrology Kidney Week 2011 (FR-OR-190).
“There needs to be a significant improvement in drug dosing in CKD patients, particularly in the outpatient setting, but others are seeing the same thing in the inpatient setting,” said senior author Arsh Jain, MD, Assistant Professor in the Division of Nephrology at the University of Western Ontario. “We are hoping that this study will serve as a call to action or an impetus for jurisdictions to implement changes in the way they deliver medication to patients or the way drugs are dosed for patients.”
The study had two primary research questions: How common are antibiotic dosing errors in patients with non–dialysis Stage 4-5 CKD, and did the initiation of estimated glomerular filtration rate (eGFR) reporting reduce such errors? The first author of the study was Alexandra Farag, a medical student at the University of Western Ontario.
The researchers conducted a population-based retrospective analysis of ambulatory antibiotic dosing errors in southwestern Ontario, analyzing medications that had maximum recommended daily doses for patients with non-dialysis CKD regardless of indication. For data sources, they linked Ontario health care databases with serum creatinine results from the largest outpatient laboratory provider in the region.
The authors used time-series modeling of the monthly rate of dosing errors from January 2003 to April 2010, for a total of 88 intervals—36 before and 52 after eGFR reporting commenced.
An average of 650 patients met the inclusion criteria each month, 65% of whom were female, 30% diabetic, and 40% hypertensive. All patients were 66 or older, had an eGFR less than 30 mL/min/1.73 m2, and had not recently been hospitalized or given prescriptions.
Of 1,464 prescriptions for study antibiotics filled over the period of analysis, 970 (66%) were dosed in excess of guidelines. For example, even though nitrofurantoin, which is used to treat urinary tract infections, is contraindicated in patients with creatinine clearance less than 60 mL/min, the antibiotic was prescribed 169 times in the study.
The initiation of eGFR reporting was not associated with a decline in the rate of antibiotic dosing errors. Prior to eGFR reporting, the average error rate was 636 per 1,000 antibiotic prescriptions, and after eGFR reporting, the rate was 680 per 1,000 antibiotic prescriptions.
Finding a Solution
“We have pretty definitively shown that in Ontario there have been a number of changes that have resulted because of eGFR reporting, but what's surprising here is that there was no change in drug dosing,” Dr. Jain said. “This can be because of a few things.
“One, are the physicians not aware of the dosing guidelines? They may be aware that patients have CKD, but do they know that they should not be giving certain drugs or changing the dose of the drugs in the patients?
“Are some of these physicians unaware of the fact that these patients have CKD? Particularly in elderly patients, their serum creatinine may be close to the normal range, but their kidney function based on eGFR reporting would be significantly reduced. If primary care providers were not paying attention to eGFR reporting and simply going with the serum creatinine reports, then they may not have been as apt to change behavior as a result of eGFR reporting.”
Investigating the reasons behind the errors could help determine how to prevent them, noted Michal L. Melamed, MD, co-moderator of the session during which the results were presented and Assistant Professor of Medicine and of Epidemiology & Population Health at the Albert Einstein College of Medicine.
“Overall, it was a nice study,” Dr. Melamed said. “It would have been interesting to follow up with physicians—which would be a different study—to figure out what the doctors were thinking. If you understand what physicians are thinking you can better target interventions to the root cause of the problem.”
She shared a common scenario that could have contributed to the findings.
“If I can imagine the way things happen, a lot of antibiotics are prescribed when a patient comes to an office complaining of some symptom, and the doctor who is writing the prescription does not have the patient's eGFR right in front of them,” Dr. Melamed said.
“One potential way I see this being remedied is if there is an electronic prescription system where as soon as you are going to put in the prescription the eGFR pops up, and I think that is the intervention that would lead to a decrease in these alarming rates of overprescription.”
Dr. Jain agreed that such technology has the potential to address this issue.
“We have done a review on computerized decision support systems in the past, and they have been shown to reduce dosing errors in patients, so absolutely this sounds reasonable,” Dr. Jain said.
“This is very easy to do in the inpatient setting, but in the community setting, that's a little more difficult. We need to start adopting electronic medical records in primary care provider offices that have computerized physician order entry associated with them. This absolutely could be a good way to go.”
The next steps are to pinpoint the source of the dosing errors and evaluate how best to reduce them.
“This opened a big can of worms,” Dr. Jain said. “The biggest thing we want to determine now is who is making these mistakes. Is it all physicians or a specific group of physicians?
“It also begs the question of how do we actually improve. This may lead to a randomized trial adopting some of these computerized systems and changing the eGFR reporting prompt to make sure the idea of drug dosing is in the back of mind when you get the report.”