Insulin Errors Potentially Plague Hospitals Treating Diabetes Patients

VA Has Had Problems Related to Insulin Pens, Concentrated Doses

Guidelines from the Institute of Safe Medication Practices have pointed out that insulin is associated with more medication errors than any other type or class of drugs. With more than 11,400 insulin-using veterans hospitalized in a recent two-year period at the VA, that is an especially critical issue for the healthcare system. Intense focus on avoiding problems with insulin, as well as new products, have improved the situation, both with subcutaneous administration – such as insulin pens – and with intravenous insulin in the critical care setting.

HORSHAM, PA — With a quarter of veterans treated in VHA facilities having a diabetes diagnosis, VAMCs deal with a high number of hospitalized patients requiring regular insulin administration.

While the VA has instituted a number of measures over the years to more safely administer insulin, especially in inpatient settings, it continues to face a dramatically increased risk of medical errors. Guidelines from the Institute of Safe Medication Practices pointed out that insulin is associated with more medication errors than any other type or class of drugs.

“As early as 1998, insulin was associated with 11% of all harmful medication errors in hospitals,” the ISMP explained. “More recent studies add evidence to the high frequency of insulin involvement in harmful medication errors. In 2004, a state reporting program established that 25% of all reported medication errors involved high-alert medications, and 16% involved insulin alone. Data published in 2008 showed that insulin was the leading drug involved in harmful medication errors, representing 16% of all medication error events with reported harm. A 2010 study found that the most common medical errors in critical care patients were insulin administration errors.”

And, the number of veterans facing those risks each year is large. A report looking at another issue—treatment intensification after discharge—noted that 11,430 insulin-using VA patients were hospitalized between 2012 and 2013.¹

The VA’s healthcare system has not been exempt from insulin errors. In October 2012, the VA Western New York Healthcare System chief of pharmacy discovered three insulin pens—designed for single-patient use only—with no patient labels in a supply drawer of a medication cart. Facility officials subsequently found three more pens without patient labels in medication carts on three other inpatient units, and, when queried, several nurses reportedly acknowledged using the pens on multiple patients.

In a report, the Office of Inspector General for the VA pointed out that inappropriately using single-patient use insulin pens on multiple patients could potentially expose patients to bloodborne pathogens.

In another, more-recent case, the VA is being sued over the death of a veteran from a wrongful insulin injection at the Louis A. Johnson VAMC in Clarksburg, WV. The lawsuit filed earlier this year alleges a “widespread system of failures” at the facility. Federal prosecutors, meanwhile, were reported to have begun presenting evidence to a grand jury in their criminal probe of at least 11 deaths suspected of being linked to improper insulin injections at that VAMC, which has not yet determined whether the deaths were related to a mistake or an intentional act by a former nursing assistant. The veterans, many of whom did not have diabetes, died of hypoglycemia.

According to the earlier OIG report on insulin pen misuse, the United States Pharmacopeia described 4,764 insulin errors over a two-year period that were reported to their voluntary reporting program. Of those, 6.6% resulted in patient harm, while, in another study, Tufts-New England Medical Center conducted an analysis of adverse events involving glucose lowering agents at 21 healthcare organizations and determined that, over about a three-year period, 2,125 inpatient errors involving insulin had occurred.

In its guidelines, the ISMP pointed out the following problems associated with dispensing, administering and monitoring subcutaneous insulin:

• Intermediate and long-acting subcutaneous insulin doses are not dispensed in the most ready-to-use form in inpatient settings, so that often the wrong vial of insulin is selected from unit stock and the wrong dose of insulin is measured when withdrawing it from a vial into a syringe.

• Errors with communicating and measuring doses of concentrated insulin. Most of the reports have been related to dosing confusion caused by the previous unavailability of a syringe with a U-500 scale. This required practitioners to measure U-500 doses with a U-100 syringe or tuberculin syringe and to teach the patient how to communicate their doses in “syringe units.”

• Coordinating glucose monitoring, meal delivery and insulin administration within the ideal time frame for rapid-acting insulin is a significant challenge often not being met in inpatient settings. In two studies, less than half of patients met the goal of receiving a rapid-acting insulin within 10-15 minutes of a meal, and 35% received glucose monitoring within one hour prior to insulin administration.

• Lack of protocols to guide insulin administration. One type of error seen in the absence of such protocols is the withholding of a basal dose of insulin when a patient’s glucose is within normal limits at the time a dose is due.

• No standardized process for alerting physicians, pharmacists and nurses as to when insulin doses must be adjusted, held or discontinued based on changes in the patient’s carbohydrate intake (e.g., changes in enteral feedings, parenteral nutrition, NPO status).

• Lack of prospective risk assessment to identify patients at high risk for hypoglycemia. Establishing and maintaining clinically appropriate glycemic targets in both the inpatient and outpatient settings has been difficult, because the risk of hypoglycemia increases with tighter glycemic control.