Hypoglycemia Incidence Utilizing Insulin Therapy and A Continuous Glucose Monitor

Patients with hypoglycemia unawareness are signs and symptoms of low blood the of effects including acute cerebrovascular infarction, neurocognitive dysfunction, retinal cell death, and loss of vision as well as problems related to activities of daily living such as ABSTRACT Purpose: The purpose of this study was to compare the rates of hypoglycemia in patients 18 years and older who use an insulin pump with those who take insulin injections. Hypoglycemia is a consequence of diabetes management and leads to a higher risk of morbidity and mortality in people with diabetes. Furthermore, hypoglycemia unawareness is a major risk factor for individuals with diabetes. Procedure: This study was a retrospective chart review of 128 patients, with 64 patients utilizing insulin pump therapy and 64 patients utilizing multiple daily injection therapy. All patients were monitored using a continuous glucose monitor. Results: There was no statistically significant difference in the time spent in hypoglycemic range between the two groups. However, additional findings included a clinical significance in that 63.13% of all patients reported hypoglycemia unawareness. Furthermore, patients in both groups experienced hypoglycemic events. The multiple daily injection group also had more outliers showing greater glycemic variability, suggesting subpar glycemic control. Conclusion: Further research is suggested to determine the prevalence of hypoglycemia and interventions to help prevent hypoglycemic events. It is imperative to advocate and educate for evidence- based, best practice guidelines by conducting and supporting research related to continuous glucose monitors and glucose control.

sleeping and driving [3]. Continuous glucose monitoring requires a prescription to obtain and is much more expensive than a selfmonitoring blood glucose (SMBG) monitor (National Institute of Diabetes and Digestive and Kidney Diseases [7]. In spite of this, a continuous glucose monitor (CGM) has the potential to be beneficial in reducing the incidence of hypoglycemic occurrences [2]. The CGM displays blood glucose values and trends in real time with alerts to allow the individual to monitor if values are hypoglycemic, normoglycemic or hyperglycemic. Flash glucose monitoring does not provide real-time and continuous data with alerts for hypoglycemic, normoglycemic or hyperglycemic occurrences [2]. Individuals who may benefit from CGM therapy include individuals on intensive insulin therapy, hypoglycemia unawareness, and/or with frequently low or high blood glucose levels [7]. Many of the individuals who currently use CGMs have type 1 diabetes. Research is being conducted on the benefits and effects of CGM use of those with type 2 diabetes [7]. According to one study, the use of a CGM helped to reduce the time individuals spent in hypoglycemia, compared with those using flash glucose monitoring [2]. The benefits of CGM compared to self-monitoring of blood glucose (SMBG) includes better management of glucose levels throughout the day, less incidence of low blood glucose emergencies, and the need for fewer finger sticks via SMBG monitors [7].

Purpose and Significance of Research
A greater understanding of hypoglycemia regarding the use of insulin therapy, whether by pump or injection, is needed to promote a decrease in the incidence of hypoglycemic occurrences, tighter control of consistent blood glucose values, as well as the prevention of severe hyperglycemic occurrences. Continuous glucose monitoring has the potential to aid in prevention of hypoglycemic occurrences and the morbidity, mortality, and secondary complications that occur as a result or in relation to diabetes [8].

Design
A retrospective chart review was used for this study. The sampling strategy was a convenience sample gathered from patient charts from a diabetes specialty clinic in the Southern United States.
Data from individual charts of those who met study criteria was collected and recorded on a study tool created by the researchers and approved by the owner of the clinic. Study inclusion criteria included individuals who were aged 18 or older using either insulin pump or insulin injections. Individuals using oral medications for the treatment of diabetes were also included if they were on insulin injections or insulin pump therapy. Additional inclusion criteria included the use of a CGM for a minimum of two weeks consecutively. Exclusion criteria for the study included individuals younger than 18 years old and/or individuals only using oral medications for the treatment of diabetes. Other exclusion criteria included those monitoring glucoses through finger stick only or having used a CGM for less than 2 weeks. Prior to data collection, the review board of Northwestern State University approved the pilot study and approval was obtained from the owner and operator of the diabetes clinic. There were no ethical conflicts with this study.
Data was collected until there were 64 subjects in each group. The CPT code 95251 was used to determine data eligible to study.

Population
A sample size calculation was performed to identify the needed amount of study data to ensure results were valid. Sample size needed per group was 64 individuals in the group using insulin injections for blood glucose management on the continuous glucose monitor and 64 individuals in the group using an insulin pump for blood glucose management on the continuous glucose monitor.
The subject population included 128 subjects' total. Health status, gender and ethnicity of these individuals varied, but all subjects had a diagnosis of type 1 or type 2 diabetes. All individuals were on continuous glucose monitoring and met inclusion criteria discussed.
The continuous glucose monitor used by study individuals provided a continuous measurement of the study subjects glucose readings. Individual electronic medical records were pulled by the CPT code 95251 (CGM data interpretation) to ensure no subject was duplicated and all included subjects were using continuous glucose monitoring. There was a modifier in the report to ensure only subjects 18 years of age and older were included.

Demographics
Participants were selected using a convenience sampling method and included all patients who met inclusion criteria.
Information was gathered retrospectively, therefore no informed consent was needed. A total of 72.66% of all patients had type 1 diabetes and 27.34% had type 2 diabetes. Race and ethnicity were as follows: 14% were black, 84% Caucasian, and <1% were Hispanic or Asian. Fifty percent of patients utilized insulin pump therapy and 50% utilized MDI therapy for the treatment of diabetes. Of all patients with type 1 diabetes, 63% used an insulin pump and only 14% of patients with type 2 diabetes used an insulin pump. For all patients included in the study, the average age was 49 years old, and the average HbA1c was 7.8%.

Data Collection
The investigators gathered all information obtained from the diabetes clinic. A data collection tally form, created by the investigators, listed all data obtained from the electronic medical

Data Analysis
The Shapiro-Wilk Test of Normality uses the distribution platform to examine a continuous variable [9,10]

. Kruskal-Wallis
Test is a non-parametric test used to determine statistically significant differences between two groups [9,10]. The null hypothesis is that data is normally distributed. If the alpha level is 0.05 and the p-value is less than 0.05, the null hypothesis is rejected [9,10]. Therefore, the data is not normally distributed. If the p-value is greater than 0.05, the null hypothesis is accepted [9,10]. The SAS version 9.4 software was used to run statistics with a 0.05 level of significant difference. The following data was collected from the patient charts using the data collection tool developed by the researchers: age, gender, race/ethnicity, diagnosis were Asian (n=1) ( Table 1).      in time spent in hypoglycemic range between the two groups, the Kruskal-Wallis Test was used. There was no significant difference in the time spent in hypoglycemic range (p=0.27). Therefore, the null hypothesis is accepted (Table 9).