Researchers have developed a new “hard-stop alert” that pops up when clinicians are trying to prescribe carbohydrate-containing medications to hospitalized children on a keto diet — a ketogenic eating plan that is high-fat and low-carb.
Notably, this type of diet is recommended for patients with treatment-resistant epilepsy, such as Lennox-Gastaut syndrome (LGS), due to its beneficial effects.
The implementation of the alert at two pediatric hospitals cut such prescribing errors by half, data shows, supporting its application to improve the safety of children with seizures on this type of diet.
Still, future studies are needed to determine the effects of reducing the prescription of such medications on ketosis maintenance, and on the frequency of seizures in these pediatric patients. Of note, achieving and sustaining ketosis, a metabolic process in which the body uses stored fat, rather than sugar, as its main energy source, is the main goal of the keto diet.
The study, “Reducing prescribing errors in hospitalized children on the ketogenic diet,” was published in the journal Pediatric Neurology.
The ketogenic diet, known as a keto diet, is a high-fat, low-carbohydrate, limited-protein diet developed by physicians in the 1920s to treat epilepsy. While the exact mechanisms by which the diet helps drop seizure frequency remain poorly understood, several studies have supported its use as an add-on therapy to conventional anti-seizure medication in epileptic patients, including those with LGS.
This restrictive diet “makes hospitalized children vulnerable to clinically significant prescribing errors because the most commonly prescribed intravenous [into-the-vein] fluids and oral medications contain carbohydrate in sufficient quantities to exceed daily carbohydrate allowance,” the researchers wrote.
A previous study, published in August 2019, showed that almost half of patients on a keto diet in a Turkish pediatric emergency department received carb-containing fluids, impairing ketosis maintenance. The researchers in that study noted that such errors “might be life-threatening.”
As such, increasing efforts have been made across most hospitals to minimize exposure to carb-containing medications in epileptic patients on keto diets, with an aim toward preventing lower seizure thresholds and serious safety events.
Clinical decision support (CDS), which encompasses a variety of tools to enhance decision-making in the clinical workflow, can be integrated into computerized order entry systems in hospitals. Such tools can help reduce medication errors by “providing the right information to the right person in the right format through the right channel at the right time in workflow,” the researchers wrote.
However, evidence suggests that current CDS systems covering keto diet considerations are inefficient at reducing prescribing errors in hospitalized patients on keto diets.
Now, researchers at Children’s Healthcare of Atlanta and the Emory University School of Medicine, in Atlanta, set out to develop an improved, user-centered CDS alert and test it at two Children’s Healthcare of Atlanta’s tertiary-care pediatrics hospitals.
First, a team consisting of two pharmacists, a ketogenic registered dietitian, and a clinical informaticist, compiled a list of carb-containing medications available on the hospital formulary that should not be prescribed to patients on the keto diet. A computerized, hard-stop alert was then created to fire whenever a clinician attempted to order one of these medications for these patients.
Next, the alert was optimized based on feedback from two pediatric clinicians, after it was compared with the previously implemented CDS. This CDS consisted of an alert — providing information on keto diet patients’ unique considerations — that popped up whenever clinicians opened the chart of such a patient, but not at any point of medication selection and order.
The final version of the entry-order alert was validated in 20 pediatric providers at varying levels of experience, who gave the hard-stop alert high ratings for ease of use, efficacy, usefulness, and user satisfaction. Overall, feedback from both groups favored the use of the new entry-order alert.
Finally, the team compared the number of prescribing errors in keto diet patients, before and after the implementation of the new alert, at the two hospitals.
During the 17-month study period, 82 pediatric patients (42 boys and 40 girls, median age 6.4 years) on the ketogenic diet were admitted across both sites, for a total of 280 admissions and 1,219 patient-days (414 pre-implementation and 805 post-implementation). Of note, patient-days means the total number of days of service provided to inpatients over a one-year period.
The results showed that the implementation of the new alert dropped prescribing errors in these patients by 49%, or 0.34 errors per hospital day, preventing an estimated 256 inappropriate prescriptions.
“The reduction in error rate was observed at both study sites and was sustained despite very little educational effort at the start of the project or maintenance interventions,” the researchers wrote.
The alert’s beneficial effect was not associated with a high impact on alert burden, as after its implementation, clinicians triggered the alert an average of two times, and many (57%) alerts were accepted — meaning the order was changed to an appropriate medication.
“Implementation of a CDS alert at order entry resulted in a sustained reduction in carbohydrate-containing medication orders for hospitalized ketogenic diet patients without an increase in alert burden,” the researchers wrote.
Both the observed alert acceptance and the change in prescribing error were higher than those reported in previous studies, the investigators noted, which may be due to the use of the user-centered design in developing and implementing the new CDS.
The team emphasized, however, that further studies are needed to assess how this reduction in prescribing errors helps to maintain ketosis and prevent seizures in children with epilepsy on the keto diet.
They also noted that the significant amount of institutional effort to create and maintain a list of inappropriate medications for keto diet patients and the fact that sugar content in medications is not always provided challenges an effort to scale this CDS to outpatient settings.