This study is a RCT aiming to use CBT-I as intervention, compared to usual care as control, to elucidate the effect of CBT-I on glycemic control, sleep quality, psychological outcomes, and cognitive function in Hong Kong Chinese older T2D comorbid with insomnia.

Insomnia is the most common sleep disorders affecting 10% to 30% of the population, imposing a huge burden on healthcare systems. While the amount of deep sleep (slow wave and rapid eye movement) declines with age, older people are known to be at escalated risk of sleep disturbances with 20-50% adults aged more than 65 years suffering from insomnia. Cognitive behavioral therapy (CBT-I) is the recommended first line treatment of chronic insomnia. However, only 40% Hong Kong Chinese adults seek help for insomnia and most of them have treatment from Complementary Alternative Medicine (CAM) or western doctors for hypnotics and none for CBT-I. Furthermore, there are increasing concerns regarding the potential negative effects of hypnotics on glycemic control. From a meta-analysis including 17 studies involving 36,938,981 individuals with insomnia followed up for a mean period of 11.6 years, there was a tendency of increased risk of mortality associated with hypnotic use.

CBT-I is well recognized as an efficacious treatment modality in people with chronic insomnia. From the results of a systemic review and meta-analysis including 20 randomized controlled trials (RCT) involving 1,162 participants with insomnia, intervention with CBT-I, incorporated at least 3 of the following elements: cognitive therapy, stimulus control, sleep restriction, sleep hygiene, and relaxation, CBT-I is demonstrated to improve sleep outcomes namely sleep onset latency (SOL), wake after sleep onset (WASO) and sleep efficiency (SE). From another meta-analysis including 37 studies aiming to examine the effect of CBT-I for insomnia comorbid with psychiatric and/or medical conditions including alcohol dependence, depression, post-traumatic stress disorders, chronic pain, cancer, Parkinson disease, coronary artery disease, pulmonary and renal diseases, CBT-I is also shown to improve sleep parameters with beneficial effects greater in those comorbid with psychiatric than in medical conditions. Although CBT-I is considered the first line treatment for chronic insomnia, the effect of CBT-I on the functional outcomes of the comorbid medical conditions is unclear. To date, there was one RCT in United States including 109 older adults (mean age 64.8 ±6.0 years) randomized to 2-hours group sessions (7-10 subjects per group) weekly over 4 months of CBT-I (n=47), Tai Chi Chih (n=39) and sleep seminar (n=23), which found that participants randomized to CBT-I had a lower risk of having increased biological risk (defined as having ≥4 biomarkers out of a total of 8) at 16-months ie 12 months post-intervention when compared to those in sleep seminar only (odd ratio, OR=0.06, 95% confidence interval, CI 0.005-0.67, p<0.01). However, the effect of CBT-I on changes of glycemia had not been reported in this study. There is another pilot RCT (ClinicalTrials.gov NCT03713996; https://clinicaltrials.gov/ct2/show/NCT03713996), enrolled 28 type 2 diabetes (T2D) with insomnia symptoms and recruitment was completed by 1 June 2019 (for 6-weeks intervention including six 1-hour sessions of CBT-I versus health education), in which insomnia severity was the primary outcome and random glucose by glucometer and glycated haemoglobin (HbA1c) were the other outcomes measures; but the results of this pilot RCT are still pending. Taken together, the effect of CBT-I on glycemic control and other outcomes in people with T2D comorbid with insomnia are under-explored.

Diabetes is one of the most prevalent non-communicable diseases affecting up to 463 million people worldwide and this figure is estimated to rise to 700 million by 2045, accounting for 10.9% of world population. From the United Kingdom Prospective Diabetes Study (UKPDS), the landmark trial of T2D, 0.9% difference in HbA1c between intensive treatment and conventional treatment arms (HbA1c 7% versus 7.9%), resulted in 12% reduction of any diabetes related endpoints (p=0.029). Although diabetes is prevalent and there is a wealth of evidence supporting that good glycemic control can prevent diabetes related complications, many people with diabetes across the globe do not reach glycemic goals. Our group has reported that <50% T2D reached glycemic goal of HbA1c <7% in Hong Kong. Behavioral changes, along with lifestyle modifications, are the cornerstones of diabetes management. Healthy diet and regular exercise not only have beneficial effects on glycemic control in T2D, but also improves sleep in T2D with insomnia. Sleep plays a pivotal role in regulating energy metabolism and updated guidelines have included adequate sleep, on average 7 hours per night, as one of the key elements of lifestyle therapy in diabetes management. Apart from quantity, quality of sleep is also essential for well-being of an individual. Optimal treatment of insomnia in diabetes may potentially help to improve glycemia although the importance of management of insomnia has not received adequate clinical attention. Of note, there are higher percentage of people with chronic insomnia suffering from diabetes than those without insomnia (13.1% versus 5%). Our team has reported that, among 3,753 Hong Kong Chinese with T2D, around one out of ten T2D had insomnia and insomnia is more commonly found in women than in men (11.4% versus 8.2%, p=0.01). Interestingly, we found that T2D men with comorbid insomnia have worse glycemic control, in terms of higher fasting plasma glucose (FPG) and HbA1c, than their counterpart without insomnia whereas such difference was not observed in women. Although insomnia is more common in women, men with insomnia is at higher risk for insomnia related glycemic problems. Other researchers have reported that people with T2D and insomnia symptoms have worse self-care than their counterparts without insomnia symptoms and the suboptimal self-care behavior in people with T2D comorbid insomnia is potentially the hurdle to achieve glycemic goals.

Not only insomnia may worsen glycemic control of T2D, insomnia may also increase the risk of development of cognitive impairment. From 981 Hong Kong Chinese T2D aged ≥60 years, without cognitive impairment at baseline (58.1% men, mean age: 62.5±2.5 years, mean HbA1c 7.4±2.2%), 3.1% developed cognitive impairment after a median follow-up of 6.8 years (interquartile range: 1.7 years), and T2D comorbid with insomnia has a 4-fold risk of development of cognitive impairment compared to their counterpart without insomnia [unpublished data, presented as poster in European Association for the Study of Diabetes (EASD) in Barcelona in 2019, https://www.easd.org/annual-meeting/easd-2019.html\]. While CBT-I improves sleep in people suffering from chronic insomnia, there is a lack of evidence regarding the impact of CBT-I on cognitive function.

Against this background and based on a multidisciplinary team experienced in CBT-I, including endocrinologist, geriatrician, psychiatrist, clinical psychologist, behavioural therapist and statistician, we propose to conduct a RCT with CBT-I as intervention, compared to usual care as control, to elucidate the effect of CBT-I on glycemic control, sleep quality, psychological outcomes, and cognitive function in Hong Kong Chinese older T2D men comorbid with insomnia.

Inclusion Criteria:

1. Aged ≥60 years;
2. T2D with suboptimal glycemic control defined as HbA1c 7.1-10%;
3. Chinese ethnicity;
4. Comorbid with insomnia having ISI score >10;
5. Able and willing to give informed written consent

Exclusion criteria:

1. Cognitive impairment (HK-MoCA score of ≤22);
2. Depression (GDS-15 score >8) and other psychiatric disorders e.g. generalized anxiety disorder, restless leg syndrome, schizophrenia to be elucidated by history from research nurse and questionnaires;
3. Benign prostatic hypertrophy by history and/or elevated PSA and any other chronic medical condition that is likely to affect sleep e.g. chronic pain;
4. Concurrent use of hypnotic drugs, psychotic medications and any drugs that are known to affect sleep;
5. Cerebrovascular accidents, vascular dementia, or any condition that are known to affect cognitive function;
6. Shift workers;
7. Haemoglobinopathies, renal failure, need of regular blood transfusion or any other conditions which will affect the validity of HbA1c in measurement of glycemia;
8. Any condition, as judged by the investigators, as ineligible to participate in this study.