As a CNS specializing in diabetes, Mary Beth Modic, DNP, RN, CDE has seen clinical nurses spend a lot of time performing a two-nurse double-check when preparing insulin as recommended by the Joint Commission and Institute for Safe Medication Practices. “There are a lot of steps involved, and it requires seven minutes on average per check,” says Dr. Modic. If a nurse working a 12-hour shift has four patients requiring four doses of insulin daily, he or she spends nearly two hours on insulin double-checks.

“This is such a high-frequency intervention,” says Dr. Modic. “We need to be very mindful what we expect nurses to do.” To ensure that the practice was evidence-based, Dr. Modic led a research study to examine the effectiveness of a subcutaneous insulin double-checking preparation intervention on insulin administration errors. The results were published in the March 2016 issue of The Journal of Nursing Administration.

Studying a 6-step intervention

Mary Beth Modic, DNP, RN, CDE

The study utilized a randomized, controlled, non-blinded, intent-to-treat methodology. Nurses caring for patients with diabetes on five units at Cleveland Clinic’s main campus – three medical and two surgical – were randomized to usual care or completion of a double-checking insulin preparation intervention prior to administration. Those adhering to usual care followed hospital procedures that did not involve double-checking. Nurses involved in the intervention completed the following seven steps:

1. Request that another nurse colleague perform an independent double-check procedure.
2. Review the physician order for subcutaneous insulin administration in the electronic medical record (EMR).
3. Verify the blood glucose result in the meter or medical record for that administration period.
4. Review the EMR for insulin dosing parameters.
5. Verify the correct insulin preparation was used and the right dose was drawn into the syringe.
6. Compare the dose with the order.
7. Confirm that the dose was correct and necessary.

Any errors identified during the double-check were recorded on a card designed for the study and corrected before insulin administration. A subsequent patient chart, card audit and analysis measured six outcomes: no errors, wrong dose, wrong preparation, wrong time, combination of two errors and inappropriate omission of insulin administration.

Research results and recommendations

During the four-week study, 266 patients were enrolled, and there were more than 5,200 combined opportunities for insulin administration in the usual care and double-checking groups. Some of the findings included the following:

• 3 percent of the insulin preparation opportunities were error-free.
• The non-error rate was higher among intervention (71.2 percent) than among usual care (66.3 percent).
• The most common error was wrong administration time; it occurred in 31.1 percent of insulin preparation opportunities.
• After controlling for clinical nurses who administered insulin and differences in patient characteristics between groups, the double-checking intervention was effective in decreasing omission errors, but ineffective in reducing wrong time, wrong preparation, wrong dose or a combination of two errors.

“We found that on average, 97 percent of all errors were timing-related,” says Dr. Modic. “Because the double-checking intervention did not prevent the most prevalent error type – wrong time – nurse leaders must encourage innovative processes and systems to overcome insulin administration delays, rather than focusing on time-consuming verification practices.”

Work redesign processes could include equitable distribution of patients who need insulin, readily available insulin products, detailed insulin administration instructions and nurse notification of meal delivery to help prevent insulin errors.

In this study, Dr. Modic and colleagues found that there were a significant number of insulin administration events. Dr. Modic would like to see more research on insulin administration and specifically on double-checking or novel processes of ensuring error-free insulin administration. “Hopefully, our study has piqued people’s interest,” she says. “It is worthy of further investigation.”