Abstract
Background. Non-standardized insulin injection has an impact on the efficacy of glucose control.
Objectives. The aim of the study was to explore the effectiveness of a nursing project in improving the insulin self-injection accuracy of diabetes mellitus patients.
Materials and methods. A total of 200 type 2 diabetes patients who received insulin therapy with an insulin pen were recruited at the First Affiliated Hospital of Army Medical University (Chongqing, China). Patients were randomly assigned to a control (n = 100) or intervention (n = 100) group. Conventional health education was conducted in the control group, while a nursing project and conventional health education were undertaken in the intervention group. The following parameters were analyzed between the 2 groups: standardized insulin pen use at admission and discharge, glycosylated hemoglobin (HbA1c), time in range (TIR), and adipose hyperplasia incidence rate 6 months after discharge.
Results. Concerning standardized insulin self-injection, the intervention group was superior to the control group, and the difference between the 2 groups was statistically significant (p < 0.05). The HbA1c levels (p = 0.000), TIR (p = 0.005) and adipose hyperplasia incidence rate 6 months after discharge (p = 0.000) all improved in the intervention group compared to the control group.
Conclusions. The application of the nursing project effectively improved the efficacy of glucose control in diabetes mellitus patients.
Key words: diabetes mellitus, insulin injection, nursing project, blood glucose management
Background
Diabetes mellitus is a serious public health issue worldwide.1, 2, 3 Based on the International Diabetes Federation (IDF) Diabetes Atlas (10th edition),4 nearly 537 million people between 20 and 79 years of age have diabetes mellitus globally, accounting for 10.5% of the adult population. Furthermore, the global population of diabetes mellitus patients is estimated to increase to 643 million (11.3%) and 783 million (12.2%) by 2030 and 2045, respectively.
According to domestic and foreign guidelines, novel hypoglycemic drugs have been most frequently recommended in recent years. However, insulin is still critical for diabetes mellitus management5, 6, 7, 8 and is the primary drug used for the lifelong treatment of type 1 diabetes patients and some type 2 diabetes patients.9, 10, 11, 12 The drug is widely applied in anti-diabetic treatment, but its glucose control effect is unsatisfactory. Studies have shown that the insulin utilization rate by diabetes patients is 50%, but the standard rate of blood glucose reduction is <35%.13
The efficacy of insulin injection therapy is affected by the dosage form, the injection device and the technique used.14, 15 When treating hypoglycemic patients using an insulin pen, most clinicians focus on dose adjustment and ignore the standardization and improvement of injection techniques.16 Although standardized treatment with an insulin pen plays a critical role in treatment,17 improper injection directly affects glucose control and poses an enormous risk to clinical treatment and patient self-management. As revealed by surveys, many problems occur during insulin treatment, such as inadequate disinfection, using the wrong injection mode, repeated injections at the same site, improper model of pen selection, repeated use of needles, improper drug loading and unloading, errors in injection time, and inappropriate drug storage. These issues can lead to large fluctuations in blood glucose and increase hypoglycemic risk, local skin infection, subcutaneous adipose hyperplasia, and other skin complications.17, 18, 19, 20, 21, 22 Furthermore, they gravely influence the effective management of blood glucose and place a substantial economic burden on patients and society.22, 23
A nursing project is a scientific approach used to improve nursing quality, defined as organizationally analyzing specific issues in nursing work with limited resources and timing, elucidating their root causes, proposing systematic solutions, and achieving specific goals to improve nursing quality and benefit patients. The project procedure follows a sequence of subject determination, status survey, goal setting, cause analysis, countermeasure development, countermeasure implementation, and effectiveness evaluation.
In this study, glycosylated hemoglobin (HbA1c) and time in range (TIR) were used to evaluate the effectiveness of the nursing project. The HbA1c is the gold standard for the clinical evaluation of blood glucose control and reflects the blood glucose level 2–3 months before enrollment, though it does not detect hyperglycemic and hypoglycemic events.24, 25 In recent years, this drawback has been offset by the emergence and development of continuous glucose monitoring (CGM) techniques that make the general blood glucose profile available.26, 27 Time in range is a new concept for improving glucose fluctuation monitoring described in the Chinese Guidelines for Prevention and Treatment of Type 2 Diabetes (2020 edition),5 and is the core indicator of the CGM technique. The application of TIR is a novel study design in the nursing field.
Objectives
The aim of the study was to explore the effectiveness of a nursing project in improving the insulin self-injection accuracy of patients with diabetes mellitus. We analyzed a comprehensive nursing intervention and therapeutic efficacy in patients with this disease who were treated with an insulin pen in the Department of Endocrinology of the First Affiliated Hospital of Army Medical University between January 2020 and March 2021.
Materials and methods
Study design
A prospective, single-center, randomized, controlled, open trial was conducted in Chongqing in China. Patients with diabetes mellitus recruited from the Department of Endocrinology of the First Affiliated Hospital of Army Medical University between January 2020 and March 2021 were the participants in the study.
The Ethics Committee of the First Affiliated Hospital of Army Medical University approved the study (approval No. KY2020085), which was registered with the Chinese Clinical Trial Registry (registration No. ChiCTR2100052472). The study complied with the ethical principles of the Helsinki Declaration. All participants gave written informed consent.
Participants
A total of 200 diabetes mellitus patients admitted to the Department of Endocrinology in our hospital between January 2020 and March 2021 were recruited as study participants. The inclusion criteria were as follows: 1) patients diagnosed with type 2 diabetes mellitus according to the World Health Organization Diagnostic Criteria for Diabetes Mellitus (1999); 2) patients who received premixed insulin therapy using 32G 4-mm insulin pen needles (BD Biosciences, San Jose, USA)16; 3) patients who had complete cognition and showed good self-care, and who were capable of cooperating with the study and independently finishing the study questionnaire; 4) patients who continued receiving the same treatment after discharge; and 5) patients who understood and agreed to participate in this study and gave written informed consent. The exclusion criteria were: 1) patients with severe hepatic and renal impairment or serious complications; 2) patients with mental diseases or self-care and cognitive impairments; and 3) patients with a surgical history at the injection site (e.g., surgical scar) or skin diseases (e.g., dermatitis or psoriasis).
Participants were randomly divided into 2 groups using a random number generator and sealed envelope, blinded to investigators, including statisticians. A nursing project and conventional health education were performed in the intervention group, while conventional health education was conducted in the control group. Based on our experience, the mean TIR was estimated at 56% in the control group and at 65% in the intervention group.
The sample size in both groups was determined using IBM SPSS v. 23.0 statistical software (IBM Corp., Armonk, USA), with α = 0.0 and β = 0.10, i.e., n1 = n2 = 85. As estimated with a hypothesized lost-to-follow-up rate of 10%, the sample size would be n1 = n2 = 85 ÷ 0.9 = 94. Therefore, 100 patients were included in each group.
Materials
The following materials were used: FreeStyle Optium Neo H glucometer (Abbott Diabetes Care Ltd., Alameda, USA); blood glucose test strips employing the glucose dehydrogenase method (#4500184071; Abbott Diabetes Care Ltd.); FreeStyle Libre H Flash Continuous Glucose Monitoring System (Abbott Diabetes Care Ltd.); D-100 Hemoglobin Testing System and corollary high-performance liquid chromatography (HPLC) reagents (Bio-Rad Laboratories Inc., Hercules, USA).
Methods
Conventional health education for the control group on day 1 after admission involved primary care nurses carrying out an evaluation and scoring single items concerning the standardization of patients’ insulin self-injections, according to the Segmentation Scoring Form for Standardized Insulin Pen Use (developed in reference to the Indian Forum for Injection Technique; see Supplementary Table 1).19, 28 The nurses found weak links and incorrect operations, then immediately corrected these problems and provided guidance. In addition, they fully implemented health education on standard insulin pen injections and distributed an operational workflow chart and alternating chart of the injection site. Every Wednesday, the primary care nurses encouraged patients and their families to concentrate on watching a standardized insulin injection video and on-site operation demonstration. In this way, patients were educated with theoretical diabetes mellitus knowledge and guided to do an operational exercise within the education tool for the “Standard Injection Week for Chinese Patients with Diabetes Mellitus” (hereafter referred to as the “model”). On the last day before discharge, the primary care nurses guided patients to complete their self-injection of insulin. On the day of discharge, the primary care nurses finished patient assessments with on-site scoring of the standardization of each patient’s insulin self-injection, according to the Segmentation Scoring Form for Standardized Insulin Pen Use. Subsequently, they optimized the education plan according to assessment results.
After discharge, the patients made a follow-up visit to a specialist nursing clinic for diabetes mellitus once per month, consecutively, for 6 months, and continued using the health education leaflet, operation workflow chart and alternating chart of the injection site distributed during hospitalization.
Nursing project and conventional health education for the interventional group
Establishment of the nursing project team
The nursing project team consisted of a leader, 2 diabetes specialist nurses and 12 primary care nurses.
Preliminary analysis of the nursing project
We retrospectively analyzed the comprehension of an insulin injection technique using an insulin pen by patients with diabetes mellitus in 2019. The team members discussed, analyzed and summarized the following 10 items: insulin type, disinfection mode, standardized needle use, injection time, indwelling time, injection site, insulin pen installation, pre-injection exhaustion, premixed insulin vial-shaking, and insulin storage.14, 19 Then, they conclusively characterized issues under 4 headings: patient, nurse, method, and environment (Figure 1).14 Improvement measures of the nursing project are shown in Supplementary Table 2.
Summary after discharge
The primary care nurses finished admission and discharge statistical analyses and compared and evaluated patients’ injections. Six months after discharge, they conducted analyses, comparisons and evaluations of HbA1c, adipose hyperplasia incidence rate, and the 14-day TIR in the 2 groups.
Indicator evaluation
Comprehension of insulin injection
The standardized insulin injection was comprehensively evaluated by segmentation scoring according to the Segmentation Scoring Form for Standardized Insulin Pen Use (Supplementary Table 1). Statistical analyses were performed before and after the nursing project intervention.
Blood glucose control criteria
In accordance with the Chinese Guidelines for the Prevention and Treatment of Type 2 Diabetes (2020 edition)5 and the interpretation of the International Consensus on Time in Range,29, 30 the analysis showed HbA1c ≤ 6.5%. Blood glucose within the 24-hour TIR was 3.9–10.0 mmol/L for 70% of the time.
Adipose hyperplasia grading
Adipose hyperplasia grading was calculated based on the guidelines outlined in the study by Conwell et al.28
Statistical analyses
The research utilized biomedical and demographic data (age and gender) of trial participants to understand the effects of the intervention across treatment and control groups. The descriptive statistics (mean, median, min, max, 1st quartile (Q1), 3rd quartile (Q3), and standard deviation (SD)) were used to analyze the distribution of baseline characteristics, including age, body mass index (BMI), cognitive ability score, fasting blood glucose, HbA1c, and diabetes mellitus. Percentage values were utilized for the categorical variable (gender).
The normality of discrete and continuous variables, such as insulin installation scores, pre-injection exhaustion scores, injection time scores, standardized use of insulin needle scores, premixed insulin vial-shaking scores, HbA1c, and TIR, were assessed using graphical methods (histograms, density plots, box plots, and quantile-quantile (Q-Q) plots). The outliers revealed in box plots were examined for typos and human errors. No outliers were removed as they were free from human error.
After assessing the data distribution of continuous/discrete variables (HbA1c, discharge scores of insulin TIR, injection time scores, standardized use of insulin needle scores, and other related indicators) using histograms with density plots and Q-Q plots, these variables were found to be non-normally distributed. The non-normally distributed data were compared between 2 groups (control and intervention) using a Mann–Whitney U test at a 5% significance level. The median value, interquartile range (IQR) and 95% confidence intervals (95% CIs) for the IQR were estimated and presented with U test statistics and p-values. The χ2 test assessed the differences in categorical variable proportions across groups (control and intervention).
All statistical analyses employed IBM SPSS v. 23.0 (IBM Corp.) and Stata v. 13.0 (StataCorp LLC, College Station, USA) software.
Results
The medical professional recruited 210 participants who satisfied the inclusion criteria. Ten participants who refused to participate in the research were excluded, and the remaining 200 individuals were randomly assigned to a control or an intervention group. Figure 2 shows the study participation flowchart. Due to an epidemic, all patients undertook their daily activities locally and showed good compliance. There was no loss at follow-up.
The proportions of males who participated and were followed up during the trial across control and intervention groups were 62% and 63%, respectively. Around 38% of females were included in the control group, while there were 37% of females in the intervention group (Figure 3).
The baseline profile of the 200 trial participants (100 per group) is shown in Table 1. The mean age of the participants in the intervention group was 59.33 years (Q1: 52.00, Q3: 67.50), which was relatively lower than the mean age in the control group (60.24 years, Q1: 54.00, Q3: 67.00).
The mean BMI of the intervention group was 24.17 kg/m2 (Q1: 22.15, Q3: 26.35). In the control group, the mean cognitive ability score was 27.22 (Q1: 26.00, Q3: 28.00), and it was 0.20 points higher than that in the intervention group. Mean fasting glucose levels in the control group (11.81, Q1: 7.65, Q3: 15.10) were higher by 0.26 points than those in the intervention group (11.55, Q1: 7.90, Q3: 13.90). Meanwhile, HbA1c levels among control participants (9.83, Q1: 7.85, Q3: 11.50) were 0.25 points lower than those in the intervention group (10.08, Q1: 8.40, Q3: 11.25). Finally, the mean course of diabetes mellitus score was 0.92 points lower in the control group (10.70, Q1: 6, Q3: 15.50) compared to the intervention group (11.62, Q1: 6.00, Q3: 15.00).
No statistically significant differences (p > 0.05) were found with regards to age, BMI, fasting blood glucose, HbA1c, course of diabetes mellitus, or insulin injection frequency between the 2 groups at a 5% significance level.
Table 2 shows the improved effectiveness of insulin pen injections between admission and discharge in the intervention and control groups. The median scores of insulin installation at discharge were significantly higher (U = 12.12, p < 0.001) in the intervention group (median/IQR: 10.0/2.0, 95% CI: 1.04–2.96) compared to the control group (median/IQR: 4.0/1.0, 95% CI: 0.25–1.75). Similarly, median scores of pre-injection exhaustion were significantly higher (U = 12.09, p < 0.001) in the intervention group (median/IQR: 10.0/0.0, 95% CI: −1.65–1.65) than in the control group (median/IQR: 2.0/5.0, 95% CI: 4.22–5.78). The median score for standardized needle use at discharge was also higher in the intervention group (median/IQR: 10.0/1.0, 95% CI: −0.87–2.87) compared to the control group (median/IQR: 4.0/0.0, 95% CI: −1.84–1.84).
Scores of injection time at discharge revealed a significant increase (U = 10.59, p < 0.001) in the intervention group (median/IQR: 10.0/0.0, 95% CI: −0.51–0.51) compared to the control group (median/IQR: 5.0/0.0, 95% CI: 0.00–0.00). Meanwhile, median scores of premixed insulin vial-shaking were lower at discharge in the control group (median/IQR: 5.0/5.0, 95% CI: 0.63–9.37) than in the intervention group (median/IQR: 10.0/0.0, 95% CI: 0.00–0.00).
Table 3 outlines changes between the 2 groups in HbA1c levels on admission and after discharge and TIR at discharge. After 6 months of the nursing project intervention, a statistically significant decrease in the HbA1c level (U = 2.97, p = 0.003) was observed in the intervention group (median/IQR: 7.0/2.1, 95% CI: 1.47–2.70) in comparison to the control group (median/IQR: 7.7/2.5, 95% CI: 2.10–3.24). Similarly, a significant increase was observed in TIR after discharge (U = 2.72, p = 0.007) in the intervention group (median/IQR: 69.5/30.0, 95% CI: 24.42–35.58) compared to the control group (median/IQR: 59.5/32.0, 95% CI: 23.39–40.61).
The χ2 analysis demonstrated a statistically significant difference in adipose hyperplasia incidence rate after insulin application between the 2 groups (χ2 = 26.4, p = 0.048). Improvements in the intervention group compared to the control group were as follows: 13.0% for mild, 6.0% for moderate and 2% for severe adipose hyperplasia. The nursing project effectively improved the therapeutic efficacy of insulin injections in diabetes mellitus patients (Table 4).
Discussion
Diabetes mellitus patients who require insulin injection have a low standard-reaching rate for glucose control, with non-drug factors being a major concern.16 In this study, we explored the effectiveness of a nursing project for improving patient insulin self-injection. The main difference between the 2 groups was that the intervention group underwent 6 health education courses implementing the latest course design of comprehensive and intensive lectures. Most diabetes mellitus patients are elderly with poor memory and have great difficulty rapidly accepting and storing knowledge, so such a course design is helpful for them. The improvements in insulin pen injections (installation) were significantly superior in the intervention group compared to the control group (p < 0.001). Thus, the nursing project successfully standardized the injection technique in a short period of time and is therefore worthy of implementation.
The HbA1c and time in range, the latest study method used in the nursing field and an important indicator for the clinical evaluation of the efficacy of glucose control, were used as primary and secondary endpoints, respectively, to evaluate the effectiveness of the nursing project in this study. We found a statistically significant difference in TIR (p = 0.007) and HbA1c (p = 0.003) between the 2 groups at 6 months, indicating an obvious improvement in the intervention group. Time in range reached 65% of the expected target in the intervention group, and it reached the recommended level (70%) after long-term intervention.31 These findings are similar to the results of a previous study.16
Comprehensive non-drug interventions work more slowly than drug interventions. However, most hypoglycemic drugs have serious side effects and may be used as adjuvant therapy to reduce the daily insulin dose, as shown in a previous study.16 Standardized insulin pen use effectively reduced variations in insulin absorption32, 33 and stably controlled blood glucose so that it approached or reached the target level, confirming the efficacy of the nursing project and improving patient satisfaction with treatment. In this study, we explored the relationship between TIR and standardized insulin injection but did not elucidate influential factors and their effects, which will be the direction of subsequent nursing research.
Similar to a previous study,22 patients using premixed insulin were selected at enrollment since a lack of standardized vial-shaking of premixed insulin may lead to an increased proportion of short-acting insulin and thus increase the risk of hypoglycemia, affecting the glucose control effect.5 In our study, patients were unfamiliar with the vial-shaking skills required for premixing insulin, though the nursing project achieved a limited improvement by reinforcement. This observation is largely related to patients not giving sufficient importance to a standardized injection technique, a lack of formal pre-hospital training, and a lack of references to guidelines, as well as the intervention time of the nursing project. A 7-day nursing intervention during hospitalization failed to induce a successful switch from a conventional injection technique to a post-intervention standardized injection technique in patients, highlighting that the intervention time of the nursing project needs to be prolonged.
In contrast to previous investigations,14, 19 this study demonstrated that the injection time was associated with pre-injection hunger in patients. Newly admitted patients showed excessive hunger due to medical examination requirements. Patients often self-injected insulin and then ate several snacks before the provision of staple foods. Some patients, especially those requiring 3 injections, missed regular meals because of the demand for a medical examination. As a result, insulin injections could not be determined as being pre-breakfast or pre-lunch. For the fasting examination, some patients underwent a fasting state of 10 h or longer due to a long waiting period for various reasons and probably presented with hypoglycemic symptoms. The evaluation and intervention of these patients are critical and will be investigated by a subgroup study in our subsequent nursing project plan.
Adipose hyperplasia is an insulin injection-related complication that manifests as skin swelling or subcutaneous induration, but its pathogenesis is unknown. It is perhaps associated with the overall time of insulin treatment, a lack of change in the injection site and repeated needle use.34 The overall time for insulin treatment is uncontrollable and is a blind spot in our study, thus it will be the subject of future research. The lack of change in the injection site and repeated needle use can be improved by the nursing project to decrease the adipose hyperplasia incidence rate and reduce insulin dose and treatment costs.16, 19 In the present study, 4 conventional injection sites (abdomen, upper arm, thigh, and buttocks) were selected, and a 32G 4-mm insulin pen needle was used in all enrolled patients. Buttock injections were difficult for patients, and they required assistance from their families or doctors/nurses to change injection sites in order to reduce the occurrence of adipose hyperplasia. Studies have suggested that the skin thickness of the Asian population is 1.7–2.8 mm on average.35, 36 A 4-mm needle is sufficiently long to penetrate the skin by vertical insertion and short enough to effectively reduce the risk of injection into the muscle layer37 while reducing and preventing the occurrence of hypoglycemia. Previous studies have demonstrated the safety of the 4-mm needle in all populations.16 Therefore, other needle models were excluded at patient enrollment.
From a technical perspective, conventional health education and a nursing project alongside conventional health education were conducted in the control and intervention groups, respectively. As shown by the study results, the incidence rate of adipose hyperplasia 6 months after discharge was lower in the intervention group than in the control group (p = 0.048), highlighting the efficacy of the nursing project. However, such improvement was insignificant due to the short intervention time of only 6 months. The occurrence of adipose hyperplasia is a long process; thus, good comparability requires standardized insulin injections over the long term.
Limitations
The study was limited by the small sample size and not employing ultrasonic diagnosis of skin and subcutaneous tissues in patients with adipose hyperplasia. Future research will focus on the following aspects: 1) the nursing of patients with concomitant diseases and diabetes complications, such as diabetes with Alzheimer’s disease or Parkinsonian syndromes, and impaired vision caused by diabetic optic neuropathy, as well as how these affect standardized insulin injections; and 2) the relationship between the TIR and chronic diabetes complications.
Conclusions
In conclusion, the application of a nursing project had a positive role in providing knowledge on insulin, using standardized injections and patient self-management, and effectively improved therapeutic efficacy, reducing the occurrence of adipose hyperplasia. Therefore, a nursing project intervention has the potential for clinical application and is worthy of implementation. Another type of intervention, telenursing, is also feasible for standardizing the insulin injection technique.
Supplementary data
The supplementary materials are available at https://doi.org/10.5281/zenodo.8137362. The package contains the following files:
Supplementary Table 1. Segmentation scoring form for standardized insulin pen use.
Supplementary Table 2. Improvement measures of nursing project.
Supplementary Fig. 1. Distribution of data of baseline characteristics of the trial participants across groups.
Supplementary Fig. 2. Distribution of data of glycosylated hemoglobin before and after treatment across groups.
Supplementary Fig. 3. Distribution of data of insulin type scores across groups.
