The Science of Diabetes Self-Management and Care2023, Vol. 49(5) 351–361© The Author(s) 2023Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/26350106231192362journals.sagepub.com/home/tde
AbstractPurpose: The purpose of the study was to examine the associations between perceived hypoglycemia and psycho-behavioral and clinical factors in persons with type 2 diabetes (T2D).
Methods: Adults with T2D were recruited from outpatient clinics in a university hospital in Korea. Sociodemographics, psycho-behavioral and clinical factors, and body composition were assessed. The participants were divided into 2 groups reporting perceived hypoglycemia or not in the previous month based on an item of the Control Problem Scale. Group differences were compared at α = .05 using SPSS (version 26.0).
Results: Of 177 participants, approximately one-third (n = 67) perceived hypoglycemia. The hypoglycemia group reported poor health-related quality of life, frequent blood monitoring and foot care, and sleep difficulties. However, no differences between groups were identified for diet, exercise, or glycosylated hemoglobin. The hypoglycemia group had a lower body mass index and a trend toward a lower skeletal muscle mass and fat free mass.
Conclusions: Perceived hypoglycemia was associated with psycho-behavioral factors and body composition. Importantly, some persons on oral antidiabetic medications that do not cause hypoglycemia still perceived hypoglycemia. Further investigation is warranted to examine the efficacy of strategies to minimize hypoglycemia and inappropriate fear of hypoglycemia. In addition, clinicians should be aware of the potential risk of hypoglycemia in persons with lower muscle mass.
Hypoglycemia is a major barrier to achieving optimal glycemic control and quality of life (QoL) in persons with type 2 diabetes (T2D).1,2 When people with T2D experience hypoglycemia, they can develop a negative perception of diabetes, leading to psychological distress (eg, fear, anxiety) and poor self-management behaviors.3,4 For instance, patients may accept chronic hyperglycemia to avoid hypoglycemia or engage in frequent blood monitoring out of fear of hypoglycemia.1,2,5 Furthermore, hypoglycemia generates a greater risk of glucose variability and associated diabetes complications.6,7 Frequent hypoglycemia impairs counter-regulatory hormonal responses and awareness of hypoglycemia, which can delay responses to hypoglycemia and lead to more severe hypoglycemia.8 Consequently, individual health and QoL are threatened.1,5
Problem-solving skills to prevent and manage hypoglycemia are currently emphasized as a basic essential component of diabetes education programs.9,10 One potential shortcoming of this approach is an increased fear of hypoglycemia. For instance, many people with T2D are concerned about hypoglycemia regardless of their actual risk.5,11,12 However, hypoglycemia in persons with T2D is mainly caused by medications (eg, dosing errors of insulin or sulfonylureas), excessive exercise and alcohol consumption, and illness related to malabsorption and metabolism (eg, gastroparesis).8 Therefore, hypoglycemia education should consider these factors, and a meticulous assessment of each patient’s actual risk is necessary prior to provision of this education.2,13 Namely, educational aims should be individualized based on each patient’s actual risk of hypoglycemia: While healthy coping and reduction of fear of hypoglycemia and diabetes-related stress can be emphasized for those at low risk, risk reduction and problem-solving skill development should be prioritized for those at high risk.14
There is limited research on the true prevalence of hypoglycemia in people with T2D including those on oral antidiabetic medications and the impact of perceive hypoglycemia on self-care activities.12,15 Previous hypoglycemia research has mainly involved persons with type 1 diabetes or those being treated with insulin.5,8,16 Because the causes of hypoglycemia are different in each person with T2D, hypoglycemia education needs to be tailored according to each patient’s risk profiles and subjective experiences.10,16,17 As a first step, this study examined the following: (1) the presence of perceived hypoglycemia in persons with T2D and (2) the difference in demographic, psychological, behavioral, and clinical factors between those who perceived hypoglycemia and those who did not. The findings of this study can assist researchers to design person-centered hypoglycemia education aimed at reducing inappropriate psychological distress and enhancing problem-solving skills to improve QoL and clinical outcomes.
A cross-sectional, correlational design was used to obtain information regarding practical strategies of holistic diabetes education and support programs for persons with T2D based on patients’ subjective experiences, herein referred to as “perceived hypoglycemia.”
Koreans with T2D age 20 years or older were recruited using a recruitment flyer distributed by trained research assistants and a poster hanging in the clinics. Those who were diagnosed with cognitive disorders or who had hearing or vision loss and who were thus unable to participate in the research activities despite having an assistive device were excluded. All participants were expected to communicate in Korean as well.
After Institutional Review Board approval (Chungnam National University Protocol code: 201910-SB-184-01), data were collected from December 2019 to May 2020. Of 602 contacts, a total of 191 participants (31.73%) were recruited from endocrinology or nephrology outpatient clinics affiliated with a university in South Korea. The main reasons to refuse study participation were time restriction, health condition, not being interested in the research, or having insecure feelings about a longer stay in the hospital clinics due to COVID-19.
Once the participants showed interest, a trained research assistant explained the research procedure using a talk-back method and obtained a written consent in a quiet and secured location. Then, participants were asked to complete self-reported questionnaires. Once participants completed the questionnaires, a trained research assistant measured body composition and height. Data collection took approximately 60 to 90 minutes for each participant. A gift certificate of 30 000 won (equivalent of $25 USD) was given as compensation for their time after data collection.
An individual’s health behaviors are influenced by the ability to adapt one’s environment and changing health conditions18 and, for some, a degree of spirituality, in support of acceptance of having diabetes and the efforts of exploring alternative health behaviors to improve health outcomes. Thus, study variables in the current study were selected with physical, mental, behavioral, and spiritual components, as follows.
Sociodemographics and severe hypoglycemia history. Sex, age, education, perceived financial condition, smoking and drinking history, and self-reported severe hypoglycemic episodes in the previous year were assessed.
Psychological factors. Spiritual health and health-related QoL were assessed. Spiritual health was assessed with the self-reported Functional Assessment of Chronic Illness Therapy - Spiritual Well-Being 12 Item Scale (12 items) in Korean.19 This 5-point scale assesses 2 domains of spiritual health, including meaning/peace (8 items) and faith (4 items).19,20 A reported Cronbach’s α in Koreans was 0.80.19
Health-related QoL was assessed using the 12-item Short Form Health Survey questionnaire. Based on responses regarding 8 concept domains (physical functioning, role-physical, bodily pain, general health, energy/fatigue, social functioning, role-emotional, mental health), T-scores for physical (PCS-12) and mental (MCS-12) functioning were calculated according to published algorithms.21,22 A higher score indicates better physiological and psychological health-related QoL, and a reported Cronbach’s α in Koreans was 0.88.23
Behavioral factors. Self-reported or interviewer-administered questionnaires were used to assess diabetes self-care activities, exercise habits, and sleep behaviors. Using the original Summary of Diabetes Self-Care Activities Questionnaire in Korean, diabetes self-care activities including diet (4 items), exercise (2 items), blood glucose monitoring (2 items), foot care (2 items), and smoking (1 item) were assessed.24 A reported Cronbach’s α in Koreans with T2D was 0.77.25
The International Physical Activity Questionnaire (IPAQ)-short form in Korean was used to assess exercise habits and amount in the previous week.26 This self-reported questionnaire required participants to recall exercise days, types of exercise, and minutes during the past week to calculate the metabolic equivalent (MET) minutes per week. Also, the time sitting in days and minutes over the past week were calculated. The Pearson’s correlation coefficient between IPAQ in Korean and accelerometer was 0.43.27
The Pittsburgh Sleep Quality Index (PSQI) in Korean was used to assess sleep behaviors and quality. Participants were asked about bed time and get-up time, sleep troubles, taking sleep pills, daytime sleep problems, and perceived sleep quality. Then, the 7 components of sleep (ie, sleep hours, latency, efficacy, subjective sleep quality, daytime sleepiness, problems maintaining enthusiasm, and sleep difficulties) were assessed based on a scoring guide. Global PSQI score (≥5 implies poor sleep quality requiring professional health care provider’s counseling) was also calculated.28 A reported Cronbach’s α in Korean with T2D was 0.89.29
Clinical factors. Body composition using InBody 970 (Biospace, Republic of Korea) and height (while barefoot) were assessed by a trained research assistant. Weight, waist and hip circumference, body mass index (BMI), skeletal muscle index, body fat, fat free mass, total body water, and intercellular water were also assessed.30,31
To obtain clinical laboratory results, electronic chart reviews were completed within 1 month of obtaining the self-reported questionnaires. Laboratory values related to dysglycemia (plasma random blood glucose, A1C, insulin, C-peptide, high sensitivity C-reactive protein), dyslipidemia (total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglycerides), liver function (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase), kidney function (blood urea nitrogen [BUN], creatinine, glomerular filtration rate [GFR], uric acid), and blood pressure were collected. Furthermore, information regarding antidiabetic mediation(s), duration of T2D, and history of diabetic complications or comorbidities, and emergency room visits and hospitalizations in the previous 6 months was collected.
Group classification by perceived hypoglycemia. To obtain answers for the research aims, the participants were divided into 2 groups according to an item reporting perceived hypoglycemia from the Control Problem Scale (CPS), which is an appendix of Diabetes Care Profile.32 The question asks about the frequency of hypoglycemiarelated symptoms such as sweating, weakness, anxiety, trembling, hunger, or headache during the previous month. Those who answered that they experienced these symptoms zero times were classified as the no perceived hypoglycemia group, while those who answered that they experienced these symptoms more than 1 time were assigned to the perceived hypoglycemia group. An item from the CPS asking about severe hypoglycemia experience during the previous year was included to obtain a better understanding of patients’ experiences of hypoglycemia. This information is reported in Table 1 with the other sociodemographic variables.
Persons who answered “don’t know” for the item asking about perceived hypoglycemia during the previous month were excluded. In addition, persons who did not answer more than 20% of items in the questionnaire were excluded. One hundred seventy-seven participants remained for the final data analysis.
All statistical analyses were performed using SPSS Statistics version 26 (IBM, Armonk, NY). Based on the data characteristics, descriptive statistics, cross-tabs, independent sample t tests, or Mann-Whitney U tests were performed to examine the group differences. The level of significance was set at P < .05 (two-tailed).
Of the 177 participants, 113 (63.8%) were men. Participants in the no perceived hypoglycemia group were older than those in the perceived hypoglycemia group, but the difference was nonsignificant (59.06 ± 12.66 years vs 62.05 ± 12.80 years, P = .131). The majority of participants reported a perceived financial condition that was poor (n = 40, 22.6%) to moderate (n = 117, 66.1%). More than 10% of the participants reported living alone. The drinking prevalence was higher than the smoking prevalence, and there were no group differences. Approximately one-third of the perceived hypoglycemia group (n = 23, 34.3%) reported severe hypoglycemic episodes (eg, passing out, needing help) in the previous year (Table 1).
There were no significant differences in spiritual health between the 2 groups. However, the perceived hypoglycemia group reported significantly lower MCS-12 scores, indicating poorer mental health functioning than the no perceived hypoglycemia group. There was a trend of a lower score for PCS-12, physical health functioning with regard to health-related QoL, in the perceive hypoglycemia group than in the no perceived hypoglycemia group, but the difference was nonsignificant (Table 2).
During the past week, frequent monitoring of blood glucose and feet was performed in the perceived hypoglycemia group, while no group differences were identified in diet, exercise, and smoking habits. However, the perceived hypoglycemia group reported more physical activity than the no perceived hypoglycemia group, indicating a trend toward a higher median MET-min/wk.
Sleep behavior, sleep hours, latency, efficacy, daytime sleepiness, and self-reported sleep quality were similar between the 2 groups. However, more participants in the perceived hypoglycemia group reported problems in maintaining enthusiasm. The reported reasons for sleep difficulties were feeling cold, having bad dreams, having pain, and others (ie, leg cramps or stiffness, coffee). More than half of the participants (n = 99, 55.83%) needed professional sleep counseling according to the global PSQI score (Table 3).
The perceived hypoglycemia group showed a significantly lower weight, BMI, and weight-hip ratio than the no perceived hypoglycemia group. In addition, this group showed a trend toward a lower body fat mass, fat free mass, skeletal muscle, total body water, and intercellular water (Table 4).
The duration of T2D was 3 years longer in the perceived hypoglycemia group. They, however, were 3 years younger than the no perceived hypoglycemia group. More than half of the participants in the perceived hypoglycemia group were prescribed medications that could potentially cause hypoglycemia (ie, insulin, sulfonylureas). In particular, this group showed a higher rate of insulin administration. No differences in diabetic complications and comorbidities or emergency room visits and hospitalizations in the past 6 months were identified between the 2 groups.
Most clinical values showed no differences between the 2 groups. For instance, there were no statistically significant differences in blood pressure, lipid profiles, or liver function. However, with regard to kidney function, the perceived hypoglycemia group showed a higher mean BUN and creatinine and a lower GFR that were outside the normal ranges for age. While A1C was almost identical between the 2 groups, random glucose levels were higher, and the ranges were wider in the perceived hypoglycemia group; however, these differences were not statistically significantly different. The serum insulin level was significantly higher in the perceived hypoglycemia group (Table 5).
This study added evidence that perceived hypoglycemia in persons with T2D is associated with psychological, behavioral, and clinical factors. Importantly, some patients taking oral antidiabetic agents that do not cause hypoglycemia still perceived hypoglycemia. Because hypoglycemia experiences decrease patient health-related QoL and increase inappropriate anxiety and concerns,3,4,33 holistic, person-centered hypoglycemia education that is individualized for the patient at the right time is necessary for persons with T2D reporting perceived hypoglycemia.1,33
Previous research has found that the efficacy of home self-glucose monitoring can be maximized with adequate health care provider feedback based on regular A1C test results.34,35 However, A1C does not provide a full understanding of patient experiences with hypoglycemia, as noted in the current study. A variation of 10% to 61% of hypoglycemia at home was reported to health care providers.2,12,15 Two reasons for the lack of reporting of low blood glucose by self-monitoring may exist: Patients may be concerned about being viewed as noncompliant or subsequent implications of revealing hypoglycemia, such as losing their job or driver’s license.12,36 Health care providers in busy clinical settings may miss an opportunity to explore patients’ hypoglycemia concerns or experiences.2,9,37 A proactive, stigma-free, person-centered approach by providers is necessary to identify unmet education needs.2 Additionally, health care providers should consider multidimensional factors such as patient’s psychological traits (eg, excessive anxiety), lifestyle habits (eg, unbalanced diet and exercise), substance use influencing counter-regulatory responses (eg, alcohol, caffeine, medication), or comorbidities (eg, nutrition malabsorption)5,38 prior to the feedback to empower patients to prevent and manage hypoglycemia.
Many people with T2D consider diabetes self-care activities as the end of a normal life,4 and thus effective strategies to improve self-care activities are necessary.5,6,9 A potential strategy is to provide accurate information of the consequences of hypoglycemia or glucose variability, including impaired hypoglycemia awareness, cognitive function decline, and diabetic complications.1,6,38,39 Also, a tailored approach depending on hypoglycemia experience may improve a patient’s problem-solving skills and motivation to control hypoglycemia. For instance, if patients have already experienced hypoglycemic episodes, these experiences may provide an opportunity to teach them how to be aware of and respond to hypoglycemia.5,14 This education should include information about the physiologic mechanism (eg, impairment of counter-regulatory and central autonomic systems) and avoidance of inappropriate psychological distress causing intentional hyperglycemia and depression.11,14,38 If patients have not experienced actual hypoglycemia, they need appropriate training to identify and treat true hypoglycemia.14
In the current study, the perceived hypoglycemia group frequently monitored their blood glucose and their feet, which may be a result of fear and anxiety of hypoglycemia. However, approximately 45% of the perceived hyperglycemia group did not take any medications that can potentially cause hypoglycemia. As previous research has addressed,12,40 patients’ nonsevere hypoglycemic or self-treated episodes are rarely or never reported to health care professionals, and it is difficult to determine whether the frequent monitoring of their blood glucose and feet was associated with the patients’ perceived hypoglycemia. In the future, data regarding patients’ subjective experiences of hypoglycemia (eg, when/why hypoglycemia occurs) and their subsequent reactions need to be collected at the time of such episodes using an objective measure (ie, glucometer) or via longitudinal tracking using continuous glucose monitoring (CGM) with a journal to fully assess these incidents, the timing of their occurrence, and the responses to treatment.17,36
Sleep influences glucose sensing and symptomatic responses to hypoglycemia, leading to nocturnal hypoglycemia and poor sleep quality.14,38 For instance, vivid dreams/nightmare and restless sleep can be symptoms and signs of nocturnal hypoglycemia.16 In the current study, the perceived hypoglycemia group reported poor sleep quality. This may be related to their higher reporting of feeling too cold and experiencing muscle cramps during sleep. Advanced technology (eg, CGM, a wearable device to assess sleep behaviors) may help researchers to obtain objective data of glucose variability and sleep cycles/patterns with time stamps to determine whether patient-reported sleep problems are associated with nocturnal hypoglycemia.41,42 Furthermore, behavioral interventions, such as sleep hygiene and hypoglycemia awareness training, should be followed to improve sleep quality and to reduce risk of nocturnal hypoglycemia.16,17
The current study suggests potential associations among glucose variability, body composition, and hypoglycemia. Uncontrolled diabetes and insulin sensitizers have been recognized as risk factors that decrease muscle quality regardless of diabetes duration,39 and adequate muscle maintenance is important to maintain body temperature, circulation, and tissue perfusion.31 In the current study, there was a trend toward lower body fat mass, fat free mass, skeletal muscle mass, total body water, and intercellular water in the perceived hypoglycemia group, which may be associated with patient-reported symptoms such as feeling cold, leg cramps, or stiffness during sleep. Further research with a larger sample is warranted to determine how body composition or muscle strength is associated with hypoglycemia, extensionally patient-reported symptoms.
Although a lack of statistically significant differences in clinical markers was found in this study, scrutiny of potential glucose variability is warranted. The A1C levels were almost identical between the 2 groups, and the mean plasma glucose values showed no statistically significant differences. However, the ranges and standard deviations of glucose values in the perceived hypoglycemia group were larger than those in the no perceived hypoglycemia group, which may imply a clinical difference. Technology advances, such as CGM, offer benefits for optimal glucose control.42 However, their availability, convenience, and cost remains problematic, especially for people with T2D. For instance, real-time CGM is currently endorsed by many professional organizations.43 However, real-time CGM data alone cannot provide vivid descriptions of patients’ experiences with mild/moderate self-treated hypoglycemia and their subsequent reactions on a “regular day,”17 and the cost-effectiveness of these methods for patients with T2D who are seen as at low risk but may be at high risk of psychological distress is a concern.4,44 Future studies using CGM along with patient self-reported journals of symptoms of hypoglycemia and the associated time of day, activities, diet, and medication could provide insights for individualizing education to prevent repetitive hypoglycemia in those with T2D.
This study has several limitations. First, we do not know whether the perceived hypoglycemia group had actual hypoglycemia. However, one aim of this study was to examine how perceived hypoglycemia was associated with psychological and behavioral factors. A future study with CGM would be beneficial to overcome this shortcoming. Second, we used a chart review to collect clinical outcomes. Patients were asked to undergo blood testing with at least a 12-hour fast. Some patients, however, underwent blood testing during an afternoon clinic visit without reporting their last mealtime. Consequently, we decided to report random plasma glucose values rather than fasting plasma glucose values. Finally, our participants consisted of those with relatively well-managed diabetes, which may not represent the actual clinical and behavioral characteristics of persons with T2D experiencing hypoglycemia.
The findings of the study supported the associations between perceived hypoglycemia and psychological, behavioral and clinical factors. Thus, further investigation is warranted to identify effective strategies to minimize hypoglycemia and inappropriate fear of hypoglycemia. Additionally, this study proposes future research to investigate the associations between glucose variability and time in range with objective measures and patient-reported symptoms, including sleep quality, psychological distress, and self-care activities, to tailor hypoglycemia education by actual risk and concerns for individual patients. As clinical implications, health care providers need to be aware of potential risk of hypoglycemia and lower quality of life in persons with lower muscle mass.
Parts of this study were presented in poster form at the 83th Scientific Sessions of the American Diabetes Association, June 23-26, 2023 (536-P in category 08 Diabetes Education).
The authors are grateful for the research team who collected data and conducted the preliminary data analysis. Also, the authors thank essay review company (editor Gretchen V) and Chungnam National University academic research promotion fund for English language editing services.
None.
This study was supported by the Korea Research Foundation (NRF-2019R1A2C1087199 & NRF-2023R1A2C2004541, PI: ES Cha).
EunSeok Cha https://orcid.org/0000-0001-6115-567X
From College of Nursing, Chungnam National University, Daejeon, South Korea (Dr Cha); Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta, Georgia (Dr Cha, Dr Faulkner); Department of Internal Medicine, Chungnam National University, Daejeon, South Korea (Dr Joung, Dr Kim); College of Sport Science, Dankook University, Cheonan, South Korea (Dr Shin); College of Natural Science, Department of Statistics, Keimyung University, Daegu, South Korea (Dr Son); and Byrdine F. Lewis School of Nursing and Health Professions, Georgia State University, Atlanta, Georgia (Dr Faulkner).
Corresponding Authors:EunSeok Cha, College of Nursing, Chungnam National University, 266 Munhwaro, #710, Daejeon, 35015, South Korea.Email: echa5@cnu.ac.kr
Kyong Hye Joung, Department of Internal Medicine, Chungnam National University, Daejeon, 35015, South Korea.Email: babehorse78@naver.com