Insomnia and Type 2 Diabetes: Measurements, Impacts and Interventions

Insomnia and type 2 diabetes (T2D) are common chronic health conditions in modern life. In the general population, 1 out of 10 are diagnosed with T2D, and 2 out of 10 have complained of insomnia symptoms. Further, more than half of people with T2D report insomnia symptoms. Previous research has shown associations between insomnia symptoms and risk of T2D. These associations could increase also the risk of medical morbidity such as cardiovascular disorders, hypertension, obesity, neuropathic pain, and psychological symptoms such as depression, anxiety, and pain. Although the underlying mechanisms that explain these associations are unknown, there is a need to understand the additive effect of insomnia symptoms on sleep, diabetes and health outcomes. People with T2D and people with sleep disturbances share several common health issues, including fatigue, low self-care, high sleep variability, poor sleep quality, daytime sleepiness, and severe symptoms of depression, anxiety, and pain. The literature explaining the actual impact of insomnia symptoms on T2D health outcomes is lacking. Because insomnia symptoms are common for people with T2D, it is imperative to understand the health issues in this population to promote preventive strategies and treatment options. In this dissertation, we examined the impact of insomnia symptoms on sleep, diabetes, and health outcomes for people with T2D. In addition, we studied the effectiveness of cognitive behavioral therapy for insomnia on sleep, diabetes, and health outcomes for people with T2D and insomnia symptoms. In Chapter 2, we compared the average and variability of sleep parameters in people with T2D with and without insomnia symptoms. As a secondary aim, we assessed the relationship between the average and variability of sleep parameters in people with T2D with and without insomnia symptoms. This study assessed between-group differences in the averages and variability of sleep efficiency and total sleep time for 59 participants with T2D with and without insomnia symptoms. The data suggested poor average sleep efficiency and high sleep efficiency variability in people with T2D and insomnia symptoms compared to people with T2D without insomnia symptoms, with statistical analyses suggesting psychological symptoms may explain the observed differences. For the secondary analysis, we observed a negative relationship between the average sleep efficiency and the variability of sleep efficiency for the entire sample. In addition, we observed a negative relationship between the average total sleep time and the variability of total sleep time for people with T2D and insomnia symptoms. The results indicated the potential effect of insomnia symptoms on regular sleep efficiency, in which improving sleep efficiency may minimize the variation of sleep efficiency or regulating the sleep schedule optimize sleep efficiency for people with T2D without or without insomnia symptoms. After understanding the possible impact of insomnia on sleep outcomes for people with T2D, insomnia symptoms may also impact diabetes outcomes. In Chapter 3, we compared the diabetes self-care behavior and glycemic control for people with T2D with insomnia symptoms to those without insomnia symptoms. As a secondary aim, we measured the association between diabetes self-care behavior and insomnia severity for all the sample. Sixty participants with T2D were divided into 2 groups based on the presence of insomnia symptoms, using the Insomnia Severity Index‎ with scores 10 indicating the presence of insomnia symptoms. We established a standardized composite score to account for the Diabetic Care Profile domains, which measured diabetes self-care behavior. The total Diabetic Care Profile composite score was significantly lower for people with T2D and insomnia symptoms compared to people with T2D without insomnia symptoms. Stepwise linear regression results ‎showed that a one-point increase in Insomnia Severity Index score significantly predicted decreased standardized Diabetic Care Profile composite score by 0.03 after controlling for age, symptoms of pain, depression, and anxiety. The data suggested that ‎people with T2D and insomnia symptoms were more likely to engage in less diabetes self-care behavior compared to people with T2D without insomnia symptoms. However, there was no significant between-group difference in the glycemic control, which might be due the small sample size as the effect size for the between-group difference was large. Whilst suboptimal diabetes self-care behavior is associated with insomnia symptoms in people with T2D, it is imperative to understand the impact of insomnia symptoms on daytime functioning in people with T2D. In Chapter 4, we compared fatigue, daytime sleepiness, and quality of life related to vitality and physical function in people with T2D with and without insomnia symptoms. Insomnia severity index was used to stratify participants into two groups, insomnia group (N=32) and non-insomnia control group (N=28). Daytime functioning including fatigue, quality of life related vitality, and physical function were worse in people with both T2D and insomnia symptoms compared to people with T2D without insomnia symptoms. Depression symptoms may have independently contributed to daytime functioning outcomes in people with T2D and insomnia symptoms. In previous chapters, we found evidence for the negative impact of insomnia symptoms on people with T2D, specifically on sleep efficiency, diabetes self-care, and fatigue severity. Thus, it is imperative to identify effective and safe insomnia intervention(s) to help people with T2D with insomnia symptoms optimize their sleep, diabetes, and health outcomes. In Chapter 5, we established a protocol study to be utilized in Chapter 6 and Chapter 7 based on resources and available funds. In Chapter 6, we examined the effect of cognitive behavioral therapy for insomnia (CBT-I) on insomnia severity in people with T2D and insomnia symptoms. The secondary aim was to explore the effect of CBT-I on other sleep outcomes and concomitant symptoms. Twenty-eight participants with T2D were randomly assigned to the CBT-I group (n=14) or the health education (HE) group (n=14). The data suggested a positive effect of CBT-I on insomnia symptoms, sleep quality, sleep self-efficacy, sleep latency, sleep efficiency, wake after sleep onset, and number of awakenings. In addition, participants in the CBT-I group showed improvements in the severity of depression and anxiety symptoms. Although the HE group received the same amount of face-to-face attention, there were no significant changes in the sleep outcomes or concomitant symptoms for the HE group. These results suggest the importance of using CBT-I with people with T2D and insomnia symptoms to improve sleep and psychological outcomes. We were also interested in the effect of CBT-I on T2D and health outcomes. Thus, in Chapter 7, we explored the effect of 6 CBT-I sessions on glycemic control, diabetes self-care behavior, and fatigue. Similar to Chapter 6, we randomized 28 participants with T2D and insomnia symptoms to the CBT-I group (n=14) or the health education group (HE; n=14). CBT-I participants showed significantly greater improvements in glycemic control, diabetes self-care behavior, and fatigue. We also found that participants in the CBT-I group experienced a clinically meaningful change in glycemic control (i.e., a difference 5%). In addition, CBT-I positively impacted diabetes self-care behavior and fatigue. The results suggested improving insomnia symptoms may have positively impacted diabetes health outcomes. In Chapter 8, we summarized our findings, and we discussed possible mechanisms and future directions for research. In summary, this dissertation project has shown the negative impact of insomnia symptoms on sleep, diabetes, and health outcomes, which improved following CBT-I. Our findings showed worse average sleep efficiency and more variability of sleep efficiency for people with T2D and insomnia symptoms compared to people with T2D without insomnia symptoms. The data also showed a negative relationship between the average sleep efficiency and the variability of sleep efficiency, which suggested improving average sleep efficiency may minimize sleep efficiency variation and vice versa. Because of these findings, clinicians may want to consider screening for insomnia symptoms in people T2D to optimize sleep efficiency. For our other findings, we found poor diabetes self-care behavior and daytime functioning in people with T2D and insomnia symptoms, in which psychological symptoms might be a contributing factor in this difference. Identifying factors that affect diabetes self-care behavior and daytime functioning in this project may encourage diabetes clinicians to screen for insomnia symptoms with people who have suboptimal diabetes self-care behavior and potentially refer them to sleep specialists. However, future research using a larger sample size of people with T2D is needed to identify the causality relationship between insomnia symptoms and diabetes health outcomes. Finally, this dissertation highlighted the effectiveness of CBT-I on sleep, diabetes, and health outcomes. The data suggested clinical meaningful changes in several sleep and diabetes outcomes. The findings showed improvements in psychological symptoms, which might contribute to blood glucose regulation. However, there is a need to understand the long-term effect of CBT-I on sleep, diabetes, and health outcomes for people with T2D