insulin pens have been widely used in hospital settings, the accuracy of pen delivery depends on the use of proper technique.38 For example, some pens will not deliver the full dose of insulin unless the needle is held in place for 6 sec after administering the dose. If the pen is removed before 6 sec has elapsed, the full dose is not injected into the site. Likewise, although needle guards are designed to prevent needle-stick injuries, when unfamiliar with the type of insulin pen, nurses not infrequently suffer needle-stick injuries despite these guards.
Smart IV pumps are IV infusion pumps designed not to allow infusion errors by prohibiting rates of infusion outside of what the pump memory has programmed to be the limits of safe infusion for that individual drug. They are a technological advance and have been shown to reduce some types of errors, but even these smart pumps may present complexities for the health-care provider unfamiliar with their proper function and errors may result.
Glucose Monitoring Errors
The ASHP recommendations regarding glucose monitoring emphasize the training of those who perform the glucose monitoring and the establishment of minimum frequencies of glucose monitoring: no less than every 4–6 h if a patient is on insulin but as often as hourly with insulin infusions. These recommendations focus on the need to document and communicate the results of glucose monitoring to the clinical team and to provide both critical values on an urgent basis and alerts to notify appropriate caregivers when there is an increased risk for hypoglycemia. The ASHP also emphasizes the need to verify POC glucose values when a critical result is noted. They also recommend the need for observation of the patient if self-monitoring is allowed. These are all sound recommendations.39
Recommendations for Change
Systemic Recommendations
• Establish a culture of safety with specific attention to assuring patient safety throughout insulin therapy. The need for timely and effective communication, coordination, and teamwork is key. The environment should promote a nonpunitive learning environment.
• A multidisciplinary management team should be formed and given authority to manage or comanage critical clinical problems or emergencies in glycemic control. They should have an ability to provide education and consultation for hospital staff, health-care providers, and patients as resources allow, and to advise the hospital administrator on suggested improvements in glycemic control.
• The hospital administrator should seek input from the glucose management team and other relevant people or groups regarding the need for additional staffing or training or other resources relevant to improving diabetes care and reducing errors in insulin administration. The hospital administrator should strongly support a culture of safety at the institution and focus care staff on clinical goals, not on nonclinical goals.
• The ASHP professional practice recommendations for the safe use of insulin in hospitals should be incorporated into the hospital’s policies and procedures. Double checks should be the norm for each of the steps in insulin therapy.
• An EHR with capabilities of CPOE should be instituted hospital-wide and all parts of the hospital should use either the same system or one that communicates transparently with the main EHR. Decision support tools and forcing functions, if carefully validated, may be used to improve safety of insulin therapy.
• Hospital-wide insulin order sets should be made available for each of the clinical situations in which errors in insulin use may occur.
• Processes of care for each step in the provision of insulin therapy, from prescription to administration and monitoring, should be written unambiguously in preprinted, approved, evidence-based orders and should be reviewed at least annually to ensure that they are being used correctly and remain valid.
• The quality of POC glucose monitoring should be a high priority. The POC glucose meter performance should be tested at regular intervals under the direction of the hospital clinical laboratory. All users of the POC glucose monitors should receive thorough education and must regularly pass proficiency testing. POC meters or strips that do not meet the new FDA standards for POC meters used in clinical units should be replaced.
• The hospital should provide guidelines delineating the strengths and weaknesses of the POC meters, and only POC meters that can demonstrate accuracy at the new FDA standard should be allowed in operating rooms or intensive care units. The hospital should consider whether a POC blood-gas analyzer using arterial blood samples for more optimal POC glucose measurement should be made available for use in critical care areas.
• Metrics for glycemic control should be obtained on a periodic, regular basis, and feedback should be provided to the clinical care areas. Benchmarking to national standards should be performed and a continuous quality improvement strategy followed. When problem areas emerge, nonpunitive steps to remedy the problem should be provided.
• Extensive education programs that provide a general understanding of diabetes care standards in the inpatient setting and specific staff responsibilities should be provided. Proficiency standards should be established. Health-care providers and staff should be tested initially and annually to ensure continued proficiency. Education programs should be tailored to the learning styles and strengths of the intended audience.
• All never-events should be given a root-cause analysis and efforts to educate and correct the errors should be undertaken.
• Encourage education of all staff, including ancillary staff such as transportation staff, on safety in insulin therapy.
• Educate nurses on the new technologies they will be responsible for, particularly POC glucose monitoring systems, smart pumps, and other new technologies used in the care of the patient with diabetes.
• Perform studies on a regular basis to ensure that the interventions to improve insulin safety are working as expected. Use objective metrics such as test scores and timely feedback of near misses and errors to obtain a more complete assessment of what is actually taking place in the hospital.
• Checklists should be developed for each provider or specialized care unit so the key steps in their tasks in insulin therapy can be clearly noted and always performed in the correct sequence.
Recommendations to Reduce the Frequency of Diagnostic Errors
• Utilize system-related factors to decrease diagnostic errors, as for example, to provide timely feedback on diagnostic studies that are needed to verify a diagnosis, both radiologic, laboratory, and other consultant-generated studies.
• Educate clinicians as to common pitfalls of clinical reasoning and provide back-up checks on their own diagnoses and real-time team discussions of the patient’s clinical status.
• Encourage clinicians to consider whether their initial diagnoses are correct or whether new information will alter their initial judgment. Premature closure is the most common cognitive error that leads to faulty diagnoses.
• Provide timely consultative expertise to the responsible provider of the diabetes care.
• Encourage the use of diagnostic simulations and case studies to provide training and education and review to improve diagnostic accuracy.
• Transitions of care must be done with the highest level of mindfulness because of the high error rates, particularly with insulin therapy.
• Errors that cause injury should be studied in depth to learn where processes of care and education need to be modified.
• Encourage feedback from patients and families regarding insulin therapy and errors, and incorporate that data in reports.
Conclusion
This chapter discussed different types of errors in insulin therapy and provided examples of each category. Unfortunately, prescribing errors, transcribing errors, dispensing errors, storage errors, administration errors, and monitoring errors are still far too common. The systemic causes of error are most important, because many of the remedies that can be done at this level are much more likely to have the largest impact on making insulin therapy safer. It is also important to look at the performance of individuals involved