Chronobiology International The Journal of Biological and Medical Rhythm Research ISSN: 0742-0528 (Print) 1525-6073 (Online) Journal homepage: http://www.tandfonline.com/loi/icbi20 Seasonal effects on the sleep–wake cycle, the rest–activity rhythm and quality of life for Japanese and Thai older people Yu Kume, Sachiko Makabe, Naruemol Singha-Dong, Patama Vajamun, Hataichanok Apikomonkon & Jiranan Griffiths To cite this article: Yu Kume, Sachiko Makabe, Naruemol Singha-Dong, Patama Vajamun, Hataichanok Apikomonkon & Jiranan Griffiths (2017): Seasonal effects on the sleep–wake cycle, the rest–activity rhythm and quality of life for Japanese and Thai older people, Chronobiology International, DOI: 10.1080/07420528.2017.1372468 To link to this article: http://dx.doi.org/10.1080/07420528.2017.1372468 Published online: 24 Oct 2017. Submit your article to this journal View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icbi20 Download by: [UAE University] Date: 25 October 2017, At: 14:23 CHRONOBIOLOGY INTERNATIONAL https://doi.org/10.1080/07420528.2017.1372468 Seasonal effects on the sleep–wake cycle, the rest–activity rhythm and quality of life for Japanese and Thai older people Yu Kume a, Sachiko Makabeb, Naruemol Singha-Dongc, Patama Vajamunc, Hataichanok Apikomonkond, and Jiranan Griffithsd Department of Occupational Therapy, Akita University, Graduate School of Health Sciences, Akita, Japan; bDepartment of Clinical Nursing, Akita University, Graduate School of Health Sciences, Akita, Japan; cInstitute of Nursing, Suranaree University of Technology, Nakhon Ratchasima, Thailand; dDepartment of Occupational Therapy, Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand Downloaded by [UAE University] at 14:23 25 October 2017 a ABSTRACT ARTICLE HISTORY Background: The sleep-wake cycle and the rest–activity rhythm are known to change with aging, and such changes have been implicated in higher levels of depression as well as an increased incidence of dementia. However, information supporting seasonal changes in the sleep–wake cycle, the rest–activity rhythm and quality of life in older community-dwelling people remains insufficient. The aim of the present study was to prospectively investigate seasonal effects on the sleep–wake cycle, the rest–activity rhythm and quality of life among older people living in areas of Japan or Thailand with different climate classifications. Method: The survey was conducted from March 2016 to May 2017, and 109 participants were recruited from Japan and Thailand: 47 older people living in Akita prefecture, Japan, and 62 older people living in Chiang Mai or Nakhon Ratchasima, Thailand. According to the Köppen–Geiger classification of Asian climates comprising tropical, desert, steppe, temperate and subarctic climates, Akita prefecture, which is located in northern Japan, is classified as a humid subarctic climate, while the Thai study areas are classified as tropical savanna. To monitor parameters of the sleep–wake cycle during nighttime (e.g. total sleep time, sleep latency, sleep efficiency, awaking time and frequency of sleep interruptions) and to calculate parameters of the rest–activity rhythm over the 24 h profile (e.g., interdaily stability, intradaily variability, relative amplitude, mean of least active 5 h period and mean of most active 10 h period), all the participants from both countries wore an Actiwatch 2 device on their nondominant wrist continuously for 7 days during each local season. The World Health Organization Quality of Life Questionnaire-BREF (WHOQOL-BREF) was also assessed during each local season. Results: The final sample size was 37 older people living in Akita prefecture, Japan, and 44 older people living in Thailand; these subjects completed the data collections during each local season. The dropout rates were 21% in Japan and 29% in Thailand. The results for the Japanese subjects showed a significantly shorter sleep time with higher levels of activity during the nighttime on summer (p < 0.001) and a fragmented rest–activity rhythm over the 24 h profile on winter (p < 0.001). The older Thai participants exhibited a poor state of night sleeping year-round, and a significant relationship was observed between seasonal variations in motor activity and the social domain of WHOQOL-BREF for each Thai season (|r| = 0.4, p < 0.01). Conclusion: These findings provide new and important information regarding seasonal effects on the sleep–wake cycle, the rest–activity rhythm and quality of life in older community-dwelling people living in two different Asian climates. Consequently, clinical preventions targeting such seasonal variations might be useful for improving the quality of life of older Japanese and Thai individuals. Received 18 June 2017 Revised 22 August 2017 Accepted 23 August 2017 Introduction The incidence of sleep disorders is higher among older people than among younger adults (AncoliIsrael 2009; Cooke and Ancoli-Israel 2011), and sleep disorders have been implicated in higher levels of depression and poorer cognitive function, in lifestyle and social environment factors, and in an KEYWORDS Rest–activity rhythm; sleep–wake cycle; seasonal effect; quality of life; actigraphy increased incidence of dementia (Cross et al. 2015; Kume et al. 2016; Luik et al. 2013). External stimuli, including light exposure and physical and social activities, have substantial effects on the sleep–wake cycle in older people (Grandin et al. 2006; Gumenyuk et al. 2012; Mauvieux et al. 2007). One such environmental factor is the inexorable passage of the seasons. Seasonal changes are a considerable CONTACT Kume Yu email@example.com Graduate School of Medicine, Doctorial Course in Health Sciences, Department of Occupational Therapy, Akita University, Assistant Professor, Ph. D., Hondo 1-1-1 Akita, JAPAN 010-8543. © 2017 Taylor & Francis Group, LLC Downloaded by [UAE University] at 14:23 25 October 2017 2 Y. KUME ET AL. factor to manage the health of older adults throughout the year, and some investigators have proposed a correlation between seasonal variations and sleep duration (Yetish et al. 2015), the quality of the sleep–wake cycle (Honma et al. 1992; OkamotoMizuno and Tsuzuki 2010) and changes in physical activity in older people (Goodwin et al. 2001). Since sleep disorders in older people have been investigated using actigraphy, the use of actigraphy is suitable for large population-based studies or long-term epidemiological surveys and also allows the evaluation of two different rhythms: the sleep–wake cycle from applying algorithms on rest–activity data (Ancoli-Israel et al., 2003) and the rest–activity rhythm by nonparametric analysis (Luik et al. 2013). In contrast, polysomnography examinations are typically performed over the course of one night while the subject sleeps in a laboratory (Jean-Louis et al. 2001; Kushida et al. 2001). Interestingly, in a nonparametric analysis using actigraphy data obtained over multiple days, Luik et al. (2013) have reported that a more fragmented distribution of activity and inactivity over 24 h periods was a characteristic of the rest–activity rhythm in older persons and was significantly correlated with the severity of depressive symptoms. In a previous report, we showed that the fragmentation of the rest–activity rhythm by nonparametric analysis over multiple days (e.g. for 7 days) can be characterized for older people with dementia in a nursing home (Kume et al. 2016). In addition, a nonparametric method developed by Van Someren et al. (1999) has a high sensitivity for the detection of changes in the rest–activity rhythm based on motor activity in older adults (Huang et al. 2002) and is currently being promoted as a research methodology for a wide range of diseases including Alzheimer disease (Harper et al. 2004), schizophrenia (Berle et al., 2010) and brain stroke (Goncalves et al. 2014). However, as few studies have examined seasonal changes in the rest–activity rhythm in older people, the seasonal effects on the rest–activity rhythm and quality of life (QOL) in older people are not fully understood. Thus, we hypothesized that the summer and winter seasons in Japan and the rainy season in Thailand might have characteristic effects on measurements of the sleep–wake cycle, the rest–activity rhythm and QOL. The objective of the present study was to investigate seasonal changes in the sleep–wake cycle, the rest–activity rhythm and QOL of older people prospectively over the course of an entire year. The seasonal effects on the sleep–wake cycle, the rest–activity rhythm and QOL were examined prospectively in older volunteers living in Northern Japan and Thailand, which were identified as areas representative of humid subarctic and tropical savanna climates, respectively, according to the Köppen–Geiger climate classification (Kottek et al. 2006; Rubel and Kottek 2010). The confirmation of international differences in the sleep–wake cycle, the rest–activity rhythm and QOL among older people living in Japan and Thailand could potentially suggest educational or rehabilitative targets specific to different Asian environments (e.g. differences in air temperature, humidity and snowfall depth). Method Participants For participants in the present study, sample size calculations were based on the G*Power (Faul et al. 2009, 2007). According to the G*Power for repeated analysis of variance measurements, a sample of 36 participants per group to detect a clinically relevant effect with α = 0.05, power = 95% and effect size = 0.25 was estimated in Japan, and the sample size in Thailand was 43 participants per group. To allow for a dropout of approximately 25%, the sample size was 47 people in Akita prefecture, Japan and 62 people in Thailand which include 32 people in Chiang Mai province and 30 people in Nakhon Ratchasima Province, Thailand. For each local area, the inclusion criteria for participants were as follow: (1) a fully independent level of activities of daily living, (2) fairly healthy without any organic central nervous system disorder, substance-related disorder or mental retardation, (3) no use of a cane or a wheelchair for transportation because of physical impairments and (4) no use benzodiazepine for sleep disturbances throughout the survey’s period. This survey was approved by the ethics committee of Faculty of Medicine, Akita University (approval number1316) and was performed in accordance with the Declaration of Helsinki II. CHRONOBIOLOGY INTERNATIONAL Downloaded by [UAE University] at 14:23 25 October 2017 Procedure The survey was conducted from March in 2016 to May in 2017. The data collections were performed through four seasons in Akita prefecture, Japan (e.g. spring, summer, autumn and winter), or through three seasons (e.g. rainy, winter and summer) in Chiang Mai and Nakhon Ratchasima, Thailand. The measurements in Japan were carried out in spring from March to May in 2016, in summer from June to August in 2016, in autumn from September to November in 2016 and in winter from December in 2016 to February in 2017. In parallel, the Thai side performed the measurements in rainy season from the middle of May to the middle of October in 2016, in winter season from the middle of October in 2016 to the middle of February in 2017 and in summer season from the middle of February to the middle of May in 2017. As the background of each local region collected data in this study, Akita prefecture is located in the northern Japan (AKITA, latitude 39°43.0´ N and longitude 140°5.9´ E), classified a subarctic zone humid climate (Kottek et al. 2006; Rubel and Kottek 2010). On the other hand, Thailand belongs to a savanna climate based on the Köppen–Geiger climate classification (Kottek et al. 2006; Rubel and Kottek 2010), located in the tropical area between latitude 5° 37´ N to 20°27´ N and longitude 97°22´ E to 105°37´ E. From the meteorological point of view, the climate of Thailand is divided into three seasons as follows: rainy or southwest monsoon season from mid-May to mid-October, winter or northeast monsoon season from mid-October to mid-February, and summer or pre-monsoon season from mid-February to mid-May (The Thai Meteorological Department 2017). Table 1 lists the meteorological data during the survey’s period reported by the Japan Meteorological Agency (2017) and the Thai Meteorological Department (2017). After obtaining informed consent from all the participants, the demographic data was firstly collected, including age, gender, body mass index (BMI), the duration of education, timed up and go test (TUG), mini mental state examination (MMSE), single residence, occupation, alcohol intake and smoking. Actiwatch 2 (Philips Respironics, Inc., Murrysville, PA, USA) worn on the participants’ nondominant wrists for continuous seven days without removal 3 Table 1. The meteorological data during the survey’s period in Japan and Thailand. Akita Seasonal mean of air temperature (°C) Seasonal mean of precipitation (mm) Seasonal mean of relative humidity (%) Seasonal mean of snowfall depth (cm) Seasonal total sunshine duration (h) Chiang Mai Spring Summer Autumn Winter 10.8 23.2 14.2 1.9 130.3 158.8 152.3 162.7 69.0 77.0 75.7 79.3 9.7 0 1.7 89.7 123.7 54.8 186.4 Seasonal mean of air temperature (°C) Seasonal mean of precipitation (mm) Nakhon Seasonal mean of air Ratchasima temperature (°C) Seasonal mean of precipitation (mm) 197.1 Rainy Winter Summer 27.7 24.8 29.3 177.0 34.7 26.6 28.5 26.3 29.3 207.5 11.0 154.7 The meteorological data in Akita prefecture, Japan, was reported by the Japan Meteorological Agency in May, 2017. The meteorological data in Chiang Mai province (Northern region) and Nakhon Ratchasima province (Northeast region), Thailand, was reported by the Thai Meteorological Department in May 2017. The Thai Meteorological Department has not reported the seasonal relative humidity (%) in 2016 and 2017. The reference data of seasonal relative humidity (%) based on 1981 to 2010 period was 81% in the rainy, 74% in the winter and 63% in the summer of the Northern region, and 80% in the rainy, 69% in the winter and 66% in the summer of the Northeast region. was used to collect objective measurements of the sleep–wake cycle and the rest–activity rhythm for each local season. The WHO Quality of Life-BREF (WHOQOL-BREF) (WHO, 1996) was also used to collect subjective measurements of the QOL during each local season. Outcome Actiwatch 2 was used to estimate the sleep–wake cycle throughout the night with the rest–activity rhythm over the 24 h profile. The sleep–wake cycle during the night was analyzed using Actiware version 6.09 (Philips Respironics, Inc., Murrysville, PA, USA), including the total sleep time (min), the sleep latency (min), the sleep efficiency (percent), the awake time (min) and the frequency of a wake status (time) throughout the night. The nonparametric analysis of the rest–activity rhythm was performed using the data of activity counts (AC) for Actiwatch 2. Variables of the nonparametric method include interdaily stability (IS), intradaily variability (IV) and relative amplitude 4 Y. KUME ET AL. (RA) (Van Someren et al. 1999; Witting et al. 1990). Recently, Goncalves et al. (2014) have reported an update of nonparametric methods. However, since a few reports are available on the updated methods, the present study applied original methods of Van Someren et al. (1999). The IS, which provides information about the rest–activity rhythm synchronization to supposedly stable environmental stimuli, is calculated from the mean 24 h profile as the following formula: Downloaded by [UAE University] at 14:23 25 October 2017 IS ¼ Pp xh xÞ2 h¼1 ð P p ni¼1 ðxi xÞ2 n The IV yields information about the fragmentation in the rest–activity rhythm over the 24 h profile, calculated from hourly raw data. The fragmentation estimated by IV may be derived from a change in the rest–activity rhythm, such as a presence of a diurnal nap or a nocturnal awakening and a failure to maintain levels of an adequate activity over the 24 h profile. The formula for IV is as follows: P n ni¼2 ðxi xi1 Þ2 IV ¼ P ðn 1Þ ni¼1 ðxi xÞ2 In equation of IS or IV, n is the total number of data, p is the number of data entries per day, xh is the hourly average, x is the average of all the data and xi represents each hour of raw data. The IS ranges from 0 for Gaussian noise, which consisted of normal distribution noise, to 1 for a perfect synchronization. The IV measure ranges from 0 to 2, whose higher value indicates a more fragmented rhythm. The RA was calculated using the data of the most active 10 h period (M10) (i.e. the level of activity during the daytime) and the least active 5 h period (L5) (i.e. the level of activity during the nighttime) obtained from the average 24 h profile. The RA measure varies between 0 for a lower amplitude and 1 for a higher amplitude. RA ¼ M10 L5 M10 þ L5 For an analysis of the WHOQOL-BREF, four domains (e.g. physical, psychological, social relations and environment) were then scored and transformed to the 0–100 scale using the follow formula (The World Health Organization (WHO) 1996): Transformed Scale ¼ ðActual raw score Lowest possible raw scoreÞ 100 Possible raw score range Statistical analysis As the statistical analysis, the Friedman test was applied to compare the median of measurements of the sleep–wake cycle, the rest–activity rhythm and the QOL among seasons in Japan and in Thailand using the adjusted p value of a multiple comparison test for post-hoc tests. After a difference (Δ) between maximum and minimum values of each parameter of the rest–activity rhythm throughout Japanese or Thai seasons was calculated, the Mann–Whitney test was carried out to compare seasonal variations between countries. Further, the Spearman correlation analysis was applied to analyze the relationship between the seasonal variations of the rest–activity rhythm and QOL measurements for each season in Japan or Thailand. SPSS version 21.0 for Windows (SPSS Inc., Chicago, IL, USA) was used for the analysis, and the level of significance was set at p = 0.05. Results The final sample size completed data collections by Actiwatch 2 through all seasons was 37 older people in Akita prefecture, Japan, and 20 older people in Chiang Mai and 24 older people in Nakhon Ratchasima, Thailand. Age, gender and BMI were not significantly different between Japanese and Thai groups, but the duration of education (Japanese/ Thai: 12.0 ± 2.4 years/7.1 ± 4.3 years), TUG (6.9 ± 1.4 s/12.8 ± 2.9 s) and MMSE (27.1 ± 2.1 scores/22.2 ± 4.7 scores) were significantly different between the groups (p < 0.0001) (see Table 2). As shown in Table 3, Japanese participants had a significant difference of total sleep time throughout the night (χ2(3) = 37.2, p < 0.001), the fragmented rhythm over the 24 h profile as the IV indexed (χ2(3) = 14.2, p = 0.003), the relative amplitude between rest and active status as indexed according to the RA (χ2(3) = 19.4, p < 0.001) and the L5 values (χ2(3) = 25.8, p < 0.001) among the seasons. CHRONOBIOLOGY INTERNATIONAL Table 2. Demographics of participants in Japan and Thailand (mean ± SD). Downloaded by [UAE University] at 14:23 25 October 2017 N Age (years) Gender (% female) Height (m) Weight (kg) BMI (kg/m2) TUG (s) MMSE (score) Education (years) Single residence (% yes) Occupation (% yes) Alcohol intake (% yes) Smoking (% yes) Japan 37 74.3 ± 6.3 59.5% 1.56 ± 0.08 60.1 ± 7.5 24.8 ± 3.0 6.9 ± 1.4 27.1 ± 2.1 12.0 ± 2.4 21.6% 13.5% 56.8% 8.1% Thailand t p Value 44 72.9 ± 6.2 1.0 0.34 75.0% 0.14 1.54 ± 0.08 0.8 0.42 55.7 ± 10.0 2.2 0.03 23.3 ± 4.0 2.0 0.05 12.8 ± 2.9 −11.7 <0.0001 22.2 ± 4.7 6.2 <0.0001 7.1 ± 4.3 6.5 <0.0001 0% 0.001 54.5% 0.002 6.8% <0.0001 2.3% 0.33 Age, height, weight, BMI, TUG, MMSE and education show the p value of the unpaired t test between groups. Gender, single residence, occupation, alcohol intake and smoking show the p value of the χ2 test. BMI: body mass index; TUG: timed up and go test; MMSE: mini mental state examination. According to a multiple comparison test, the summer season had significantly the shortest sleep times and the highest L5 values in all of Japanese seasons, and the winter season had significant higher IV values than the other seasons. Additionally, the psychological domain of the WHOQOL-BREF differed significantly among the seasons with having no significance according a multiple comparison test (see Table 4). Table 5 lists results of the sleep–wake cycle throughout the night and the rest–activity rhythm over the 24 h profile in Thailand. Thai participants had a significance on total sleep time (χ2(2) = 6.7, p = 0.04). According to a multiple comparison test, the summer season had a shorter sleep time during the night than the winter season in Thailand. Table 6 also shows that the psychological domain of the QOL 5 measurement had a significant difference among Thai seasons with no differences regarding a multiple comparison test. Figure 1 illustrates a box plot of the comparison of the Δ rest–activity rhythm parameters between Japan and Thailand, and included a significant difference for the nonparametric unpaired t test. The Mann– Whitney test showed that a variation of fragmentation (e.g. ΔIV) in Japanese participants was significantly higher than that in Thailand (p < 0.05), whereas ΔL5 in Japanese individuals was significantly lower than that in Thailand (p < 0.05). Table 7 displays the Spearman correlation analysis of the Δ rest–activity rhythm parameters (e.g. ΔIS, ΔIV, ΔRA, ΔL5 and ΔM10) and the QOL measurements for each season in Japan and Thailand. In univariate analysis, the ΔRA in Japanese participants was correlated with physical domain in the spring (r = 0.35, p < 0.05), social domain in the summer (r = 0.35, p < 0.05) and environmental domain in the winter (r = 0.41, p < 0.05) and the ΔRA in Thai side was correlated with psychological domain in the winter (r = −0.40, p < 0.05). Additionally, the ΔM10 in Thai side was significantly correlated with social domain of all seasons (rainy; r = −0.42, p < 0.01, winter; r = −0.45, p < 0.01, summer; r = −0.41, p < 0.01), having a significant correlation with psychological (r = −0.40, p < 0.05) and environmental domain (r = −0.32, p < 0.05) in the rainy. Discussion In this survey, we examined seasonal effects on the sleep–wake cycle, the rest–activity rhythm and QOL of Table 3. The comparison of the sleep–wake cycle throughout the night and the rest–activity rhythm over the 24 h profile in the participants (N = 37) among Japanese four seasons. Spring Summer Autumn Winter Median Quartile Median Quartile Median Quartile Median Quartile Total sleep Time (min) 434.4 79.1** 371.4 65.5** †† ‡‡‡ 414.4 97.8†† § 435.2 122.5‡‡‡ § Sleep latency (min) 10.8 21.8 11.7 15.8 7.2 18.1 10.7 18.5 Sleep efficiency (%) 83.9 9.9 83.0 11.8 83.9 7.8 84.0 9.9 Awaking time (min) 54.0 35.6 53.3 41.6 50.0 36.5 57.0 37.2 Awaking frequency (time) 24.8 16.2 24.2 14.4 23.3 14.2 25.8 14.5 IS 0.61 0.14 0.59 0.15 0.60 0.12 0.61 0.13 0.87 0.33‡‡ 0.98 0.31* † ‡‡ IV 0.87 0.31* 0.89 0.18† RA 0.91 0.07** 0.85 0.08** 0.91 0.08 0.90 0.09 852.9 718.8‡ 925.4 636.3†† L5 (counts) 796.6 633.9*** 1242.2 626.9*** †† ‡ M10 (counts) 18215.7 8007.1 17143.7 8472.6 19058.0 7272.3 17605.0 10256.3 χ2 (3) 37.2 4.8 3.45 1.1 3.1 1.3 14.2 19.4 25.8 1.5 p Value <0.001 0.19 0.33 0.78 0.38 0.72 0.003 <0.001 <0.001 0.67 *p < 0.05, **p < 0.01, ***p < 0.001; †p < 0.05, ††p < 0.01; ‡p < 0.05, ‡‡p < 0.01, ‡‡‡p < 0.001; §p < 0.05; the adjusted p value of a multiple comparison test for post-hoc tests. IS: interdaily stability; IV: intradaily variability; RA: relative amplitude; L5: the least active 5 h period obtained from the average 24 h profile; M10: the most active 10 h period obtained from the average 24 h profile. Table 3 indicates χ2 (degree of freedom) and p value of the Friedman test among seasons. 6 Y. KUME ET AL. Table 4. The comparison of WHOQOL-BREF measurements in the participants among four Japanese four seasons. Spring Physical domain (score: 0–100) Psychological domain (score: 0–100) Social domain (score: 0–100) Environmental domain (score: 0–100) n 35 37 36 35 Median 67.9 62.5 66.7 59.4 Summer Quartile 14.3 16.6 14.6 15.7 Median 67.9 54.2 66.7 59.4 Autumn Quartile 17.9 12.5 8.4 18.7 Median 64.3 54.2 66.7 56.3 Winter Quartile 14.3 20.9 8.4 15.6 Median 64.3 58.3 66.7 56.3 Quartile 17.9 16.7 8.4 15.6 χ2 (3) 4.6 9.2 2.3 5.4 p Value 0.20 0.03 0.51 0.15 Table 4 indicates χ2 (degree of freedom) and p value of the Friedman test among seasons. Table 5. The comparison of the sleep–wake cycle throughout the night and the rest–activity rhythm over the 24 h profile in the participants (N = 44) among Thai three seasons. Downloaded by [UAE University] at 14:23 25 October 2017 Rainy Total sleep time (min) Sleep latency (min) Sleep efficiency (%) Awaking time (min) Awaking frequency (time) IS IV RA L5 (counts) M10 (counts) Median 371.5 10.3 82.2 50.5 22.8 0.60 0.81 0.85 1459.6 18135.8 Winter Quartile 76.8 12.9 7.9 40.5 11.8 0.18 0.32 0.13 1099.4 6849.6 Median 388.2 9.6 82.2 62.8 26.8 0.60 0.78 0.88 1273.1 19410.3 Summer Quartile 104.6* 11.9 11.7 60.9 10.4 0.21 0.23 0.09 838.9 8032.8 Median 351.4 11.0 82.2 59.7 25.5 0.66 0.78 0.88 1462.2 21960.4 Quartile 62.5* 10.2 9.7 34.4 10.2 0.15 0.27 0.10 1206.0 11706.1 χ2 (2) 6.7 0.1 1.0 4.9 2.2 0.7 0.2 3.3 2.0 5.1 p Value 0.04 0.98 0.62 0.09 0.33 0.72 0.89 0.20 0.38 0.08 *p < 0.05; the adjusted p value of a multiple comparison test for post-hoc tests. IS: interdaily stability; IV: intradaily variability; RA: relative amplitude; L5: the least active 5 h period obtained from the average 24 h profile. Table 5 indicates χ2 (degree of freedom) and p value of the Friedman test among seasons. Table 6. The comparison of WHOQOL-BREF measurements in the participants among Thai three seasons. Rainy Physical domain (score: 0–100) Psychological domain (score: 0–100) Social domain (score: 0–100) Environmental domain (score: 0–100) n 40 40 40 40 Median 58.9 54.2 58.3 61.0 Quartile 14.3 12.5 31.2 14.8 Winter Median 53.6 54.2 66.7 56.3 Quartile 10.7 8.3 25.0 12.5 Summer Median 55.4 58.3 66.7 57.9 Quartile 10.7 8.3 25.0 12.5 χ2 (2) 6.0 6.4 0.8 4.7 p Value 0.05 0.04 0.67 0.09 Table 6 indicates χ2 (degree of freedom) and p value of the Friedman test among seasons. older adults living in specific areas of Japan and Thailand. The results of this study indicated that Japanese older people had significantly shorter total sleep times and higher L5 values over the 24 h profile during the summer and a more fragmented rest– activity rhythm over the 24 h profile, as shown by higher IV values, during the winter, compared with other seasons in Japan. In contrast, the Thai participants hardly showed any differences in the sleep–wake cycle or the rest–activity rhythm parameters according to season, with a significant difference in total sleep time only observed between winter and summer. At first, the results regarding nighttime sleep during summer and winter agreed with those of Okamoto-Mizuno and Tsuzuki’s report (2010), since the older subjects had a significantly shorter total sleep time (mean ± SE: 350.8 ± 15.7 min) in summer than in fall (403.2 ± 16.4 min) or winter (426.5 ± 14.2 min), with a sleep time of under the mean of 7 h a night (Ancoli-Israel 2009). However, no differences on sleep efficiency were observed among seasons in the results showed in Table 3, and this result disagreed with Okamoto-Mizuno’s report that sleep efficiency significantly decreased in the summer (81.4 ± 2.9%) compared with in the fall (90.2 ± 1.2%) or winter (91.6 ± 1.3%) (Okamoto-Mizuno and Tsuzuki 2010). In addition, our results for sleep efficiency suggest relatively low values throughout the seasons compared with the 85% cut-off value for poor sleepers (Lacks and Morin 1992). We believe that among older subjects living in Northern Japan the total sleep time during the night might be shorter during summer than during other seasons. Our results of the rest–activity rhythm also support the hypothesis that older people are easy to have a fragmented rest–activity rhythm over the 24 h profile Downloaded by [UAE University] at 14:23 25 October 2017 CHRONOBIOLOGY INTERNATIONAL 7 Figure 1. The comparison of a difference (Δ) of the rest–activity rhythm parameters between Japan and Thailand. during winter in Japan. For the participants who were monitored throughout the Japanese seasons, the fragmentation score during winter (e.g. median IV ± interquartile range (IQR): 0.98 ± 0.31) was under the 1.10 cut-off value for IV (Kodama et al. 2016). Although few reports are available regarding seasonal characteristics examined using nonparametric methods, one possibility is that the fragmented rest–activity rhythm over the 24 h profile might be higher during the winter, as possibly supported by the Nakanojo study in which older adults exhibited relatively low step counts and physical activity amounts under or over 3 metabolic equivalents (METs) as measured using an accelerometer during winter, with values peaking in spring or autumn (Yasunaga et al. 2008). As shown in Figure 1, the greater seasonal variation (i.e. ΔIV) in IV values in Japanese participants also suggests that the fragmented rest–activity rhythm over the 24 h profile should be carefully observed in Japanese older people during winter. Furthermore, since the L5 values over the 24 h profile generally mean the levels of activity during the nighttime, our results suggest that the levels of activity during the nighttime are higher during summer, and the higher values of L5 on summer may be characterized in Japanese older people living in a community. Though a paucity of material is available on our findings, some studies have reported that an increase in L5 values may express a marked trait of the sleep–wake cycle in older adults (Huang et al. 2002) and a state of night sleeping with frequent sleep interruptions (Goncalves et al. 2015). Thus, these findings suggest the importance of monitoring the levels of activity during nighttime as higher values of L5 were indicated in Japanese older people, with parameters of the sleep–wake cycle. In disagreement with our hypothesis, only seasonal variations in total sleep time were observed in the 8 Y. KUME ET AL. Table 7. The correlation between seasonal variations (Δ) of the rest–activity rhythm parameters and WHOQOL-BREF measurements among Japanese and Thai seasons. Japan (N = 35) Physical domain Psychological domain Social domain Downloaded by [UAE University] at 14:23 25 October 2017 Environmental domain Season Spring Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn Winter ΔIS .033 .199 .143 .029 .092 −.174 −.004 −.061 −.057 −.071 −.075 −.278 −.049 −.150 −.083 −.012 ΔIV −.036 .074 .156 .014 .151 −.020 .196 .134 .002 −.116 .124 −.049 .313 −.011 .113 .319 ΔRA .353* .184 .095 .316 .235 .152 .176 .138 .218 .346* .240 .258 .294 .212 .212 .414* Thailand (N = 40) ΔL5 .195 −.072 .158 .120 .101 .098 .038 .064 .226 .159 .248 .160 .071 .022 .126 .183 ΔM10 .017 .132 .165 .123 .284 .278 .177 .255 −.115 −.013 −.012 −.126 .178 −.120 .119 .022 Season Rainy Winter Summer ΔIS .017 −.067 −.085 ΔIV −.102 .011 −.071 ΔRA .178 −.118 −.148 ΔL5 .127 .185 −.013 ΔM10 −.023 .016 −.070 Rainy Winter Summer −.085 −.114 −.136 −.027 −.033 −.162 .255 −.397* −.247 .144 −.231 .033 −.361* −.145 −.267 Rainy Winter Summer −.020 −.304 −.272 −.260 −.197 −.299 .049 −.304 −.179 −.015 −.096 −.073 −.423** −.447** −.410** Rainy Winter Summer −.162 −.208 −.142 −.053 .066 .105 .089 −.257 −.183 .003 −.047 .074 −.321* −.241 −.249 *p < 0.05, **p < 0.01; statistics represent Spearman r correlations for each domain of QOL measurements and a difference of the rest–activity rhythm parameters. ΔIS: a difference between maximum and minimum of interdaily stability values among seasons; ΔIV: a difference between maximum and minimum of intradaily variability values among seasons; ΔRA: a difference between maximum and minimum of relative amplitude values among seasons; ΔL5: a difference between maximum and minimum of the most active 10 h period’s values among seasons; ΔM10: a difference between maximum and minimum of the least active 5 h period’s values among seasons. Thai participants, including a shorter sleep time during summer (median ± IQR: 351.4 ± 62.5 min) than in winter (388.2 ± 104.6 min). The total sleep times of the Thai participants during all the seasons were also under the cut-off value of a mean of 7 h per night (Ancoli-Israel 2009), resulting in a relatively shorter sleep time than that of the Japanese participants. Furthermore, our findings for sleep efficiency in the Thai participants agreed with the prevalence of insomnia (Assantachai et al. 2011; Sukying et al. 2003) and the appearance of poor sleep quality (Sunuttra et al. 2008) in a questionnaire survey for older Thai population, with a low rate of sleep efficiency below the border range (e.g. under 85% cut-off value) for insomnia sufferers reported by Lacks and Morin (1992). However, no direct comparisons of sleep–wake cycles in older Thai adults have been made. Therefore, our findings suggest that older Thai people experience relatively short sleep times and low sleep efficiency year-round. Regarding the results for the rest–activity rhythm parameters in Thai participants, although seasonal variations were not well documented in this study, the L5 values of the Thai participants were relatively higher than those of the Japanese subjects. The increase in L5 values in Thai participants may express higher levels of activity during the nighttime and a sleep state with frequent sleep interruptions (Goncalves et al. 2015; Van Someren et al. 1999). Considering the increasing evidence that a high movement-activity level during the nighttime are apparent in older Thai adults, as shown in Figure 1, a greater variation in L5 values across the Thai seasons suggests that higher levels of activity during the night can be characterized as a seasonal effect in older Thai adults. Finally, significant differences in QOL measurements across the seasons were observed for a psychological domain in the Japanese (see Table 4) and Thai participants (see Table 6), although the results of a multiple comparison test were not significant. Our results partially support the view that the seasonal environment (e.g. increasing or decreasing air temperatures and light exposures) is associated with changes in mood (Palinkas et al. 2000) and physical or psychological fatigue (Fujii et al. 2015). According to the correlation analyses shown in Table 7, seasonal variations in the rest–activity rhythm parameters across Japanese or Thai seasons were significantly associated with several domains of QOL measurements in the participants in each country. Most interestingly, the ΔM10 of the rest–activity rhythm parameters was negatively correlated with a social domain for each season among the Thai participants (e.g. rainy: r = −0.42, p < 0.01; winter: r = −0.45, p < 0.01; summer: r = −0.41, p < 0.01), indicating seasonal variations in motor activity and motor Downloaded by [UAE University] at 14:23 25 October 2017 CHRONOBIOLOGY INTERNATIONAL capacity. To date, available information supports the hypothesis that lower M10 values are likely to be correlated with motor difficulty or a reduction in exercise (Carvalho-Bos et al. 2007), while higher M10 values are likely to be associated with better QOL (Goncalves et al. 2015). These findings suggest that seasonal variations in motor capacity (e.g. ΔM10) are potentially associated with the quality of social activities year-round in older Thai people. In conclusion, this study provides new and important information concerning seasonal effects on the sleep–wake cycle, the rest–activity rhythm and QOL in older community-dwelling people living in Japan or Thailand. Our findings suggest that for older people living in Japan short sleep times and higher levels of activity during the nighttime on summer are notable among parameters of the sleep–wake cycle and the rest–activity rhythm, with a fragmented rest–activity rhythm over the 24 h profile observed during winter. On the other hand, older Thai people should pay attention to a state of night sleeping year-round as well as the relationship between seasonal variations in motor activity over the 24 h profile (e.g. ΔM10) and QOL based on social activities. This information obtained through nonparametric analyses of the rest–activity rhythm should help older people to understand their current statuses as they relate to the seasonal rest–activity rhythm and QOL. Professionals also need to provide individual advice based on an individual’s lifestyle and should provide educational or rehabilitative programs to improve the rest–activity rhythm and psychological status. Moreover, information on international differences could help older people worldwide to become more aware of the different environments existing in Asian countries (e.g. before considering migration or a prolonged stay). Limitations of study The absence of detailed information on the amount of daylight and on lifestyles might be a potential limitation of this study since modern light exposure patterns (e.g. exposure to natural and/or electric light) can affect sleep and other daily rhythms in physiology and behavior (Wright et al. 2013). The sample size in this study was also determined based on sample size estimations using the G*Power (Faul et al. 2009, 2007), but the samples were somewhat 9 biased with a female preponderance observed in both countries. In addition, the significant difference in general cognitive function between the countries might have been influenced by potential educational differences. These limitations were considered when interpreting the present study findings. Interestingly, our results indicated that the physical function status of the Japanese participants (mean TUG ± SD: 6.9 ± 1.4 s) was significantly higher than that of the Thai participants (12.8 ± 2.9 s), whereas the M10 values (i.e. motor activity during daytime) for each season in Japan were lower than those in Thailand. The actimetry worn on the participants’ nondominant wrist to measure the rest–activity rhythm detects all wrist movements and records the accumulated ACs at 1 min intervals. These conditions raise the possibility that older Thai people perform physical activities or jobs that involve more upperlimb motions than lower-limb motions compared with older Japanese individuals. However, this point remains a matter of speculation. Further research taking lifestyle differences between countries into consideration is needed. Acknowledgements The authors were supported by the staffs at Akita University, Chiang Mai University and Suranaree University of Technology for their assistance in performing measurements. Mr. Tanaka S and Ms. Tanyalak Y from Bangkok branch office at HOKUTO bank arranged to firstly meet researchers among three universities. We wish to thank all the participants from Daisen area, Itagawa-Shimoabukawa area, Kamiiwakawa area in Akita prefecture, Japan and all the participants in Chiang Mai and Nakhon Ratchasima, Thailand. In addition, for the present study, the user license of the WHOQOL-BREF was approved by the World Health Organization (WHO). Description of authors’ roles All of the authors designed the present study together. We also contributed to data collections, data interpretation, writing and revision of the report. Funding This research was supported by the Grant-in-Aid for Scientific Research (C) (No. 26463444, 2014-2018) in Japan. This sponsor played no role in the research other than funding the research. 10 Y. KUME ET AL. ORCID Yu Kume http://orcid.org/0000-0003-3318-1082 Downloaded by [UAE University] at 14:23 25 October 2017 References Ancoli-Israel S, Cole R, Alessi C, Chambers M, Moorcroft W, Pollak CP. (2003). The role of actigraphy in the study of sleep and circadian rhythms. Sleep. 26: 342–392. Ancoli-Israel S. (2009). Sleep and its disorders in aging populations. Sleep Med. 10(Suppl 1):S7–11. doi:10.1016/j. sleep.2009.07.004. Assantachai P, Aekplakorn W, Pattara-Archachai J, Porapakkham Y. (2011). Factors associated with insomnia in older people with a mild to moderate degree of poor cognitive ability in Thailand. Geriatr Gerontol Int. 11:16– 23. doi:10.1111/j.1447-0594.2010.00627.x. Berle JO, Hauge ER, Oedegaard KJ, Holsten F, Fasmer OB. (2010). Actigraphic registration of motor activity reveals a more structured behavioural pattern in schizophrenia than in major depression. BMC Res Notes. 3:149. doi:10.1186/ 1756-0500-3-149. Carvalho-Bos SS, Riemersma-Van Der Lek RF, Waterhouse J, Reilly T, Van Someren EJ. (2007). Strong association of the rest-activity rhythm with well-being in demented elderly women. Am J Geriatr Psychiatry. 15:92–100. doi:10.1097/ 01.JGP.0000236584.03432.dc. Cooke JR, Ancoli-Israel S. (2011). Normal and abnormal sleep in the elderly. Handb Clin Neurol. 98:653–65. Cross N, Terpening Z, Rogers NL, Duffy SL, Hickie IB, Lewis SJ, Naismith SL. (2015). Napping in older people ‘at risk’ of dementia: relationships with depression, cognition, medical burden and sleep quality. J Sleep Res. 24:494– 502. doi:10.1111/jsr.12313. Faul F, Erdfelder E, Buchner A, Lang AG. (2009). Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods. 41:1149–60. doi:10.3758/BRM.41.4.1149. Faul F, Erdfelder E, Lang AG, Buchner A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 39:175–91. doi:10.3758/BF03193146. Fujii H, Fukuda S, Narumi D, Ihara T, Watanabe Y. (2015). Fatigue and sleep under large summer temperature differences. Environ Res. 138:17–21. doi:10.1016/j. envres.2015.02.006. Goncalves BS, Adamowicz T, Louzada FM, Moreno CR, Araujo JF. (2015). A fresh look at the use of nonparametric analysis in actimetry. Sleep Med Rev. 20:84–91. doi:10.1016/j.smrv.2014.06.002. Goncalves BS, Cavalcanti PR, Tavares GR, Campos TF, Araujo JF. (2014). Nonparametric methods in actigraphy: an update. Sleep Sci. 7:158–64. doi:10.1016/j. slsci.2014.09.013. Goodwin J, Pearce VR, Taylor RS, Read KL, Powers SJ. (2001). Seasonal cold and circadian changes in blood pressure and physical activity in young and elderly people. Age Ageing. 30:311–17. doi:10.1093/ageing/30.4.311. Grandin LD, Alloy LB, Abramson LY. (2006). The social zeitgeber theory, circadian rhythms, and mood disorders: review and evaluation. Clin Psychol Rev. 26:679–94. doi:10.1016/j.cpr.2006.07.001. Gumenyuk V, Roth T, Drake CL. (2012). Circadian phase, sleepiness, and light exposure assessment in night workers with and without shift work disorder. Chronobiol Int. 29:928–36. doi:10.3109/07420528.2012.699356. Harper DG, Stopa EG, McKee AC, Satlin A, Fish D, Volicer L. (2004). Dementia severity and Lewy bodies affect circadian rhythms in Alzheimer disease. Neurobiol Aging. 25:771–81. doi:10.1016/j.neurobiolaging.2003.04.009. Honma K, Honma S, Kohsaka M, Fukuda N. (1992). Seasonal variation in the human circadian rhythm: dissociation between sleep and temperature rhythm. Am J Physiol. 262:R885–891. Huang YL, Liu RY, Wang QS, Van Someren EJ, Xu H, Zhou JN. (2002). Age-associated difference in circadian sleepwake and rest-activity rhythms. Physiol Behav. 76:597–603. doi:10.1016/S0031-9384(02)00733-3. The Japan Meteorological Agency. (2017). Weather, climate & earthquake information, Climate in Japan. Tables of Monthly climate Statistics. [accessed 2017 Apr 3].http:// www.data.jma.go.jp/obd/stats/data/en/smp/index.html. Jean-Louis G, Kripke DF, Cole RJ, Assmus JD, Langer RD. (2001). Sleep detection with an accelerometer actigraph: comparisons with polysomnography. Physiol Behav. 72:21–28. doi:10.1016/S0031-9384(00)00355-3. Kodama A, Kume Y, Tsugaruya M, Ishikawa T. (2016). Deriving the reference value from the circadian motor active patterns in the “non-dementia” population, compared to the “dementia” population: what is the amount of physical activity conducive to the good circadian rhythm. Chronobiol Int. 33:1056–63. doi:10.1080/ 07420528.2016.1196696. Kottek M, Grieser J, Beck C, Rudolf B, Rubel F. (2006). World Map of the Köppen-Geiger climate classification updated. Meteorol Z. 15:259–63. doi:10.1127/0941-2948/ 2006/0130. Kume Y, Kodama A, Sato K, Kurosawa S, Ishikawa T, Ishikawa S. (2016). Sleep/awake status throughout the night and circadian motor activity patterns in older nursing-home residents with or without dementia, and older community-dwelling people without dementia. Int Psychogeriatr. 28:2001–08. doi:10.1017/ S1041610216000910. Kushida CA, Chang A, Gadkary C, Guilleminault C, Carrillo O, Dement WC. (2001). Comparison of actigraphic, polysomnographic, and subjective assessment of sleep parameters in sleep-disordered patients. Sleep Med. 2:389–96. doi:10.1016/S1389-9457(00)00098-8. Downloaded by [UAE University] at 14:23 25 October 2017 CHRONOBIOLOGY INTERNATIONAL Lacks P, Morin CM. (1992). Recent advances in the assessment and treatment of insomnia. J Consult Clin Psychol. 60:586–94. doi:10.1037/0022-006X.60.4.586. Luik AI, Zuurbier LA, Hofman A, Van Someren EJ, Tiemeier H. (2013). Stability and fragmentation of the activity rhythm across the sleep-wake cycle: the importance of age, lifestyle, and mental health. Chronobiol Int. 30:1223– 30. doi:10.3109/07420528.2013.813528. Mauvieux B, Gouthiere L, Sesboue B, Denise P, Davenne D. (2007). Effects of the physical exercise and sports on the circadian rhythm of temperature and waking/sleep pattern of the elderly person. Examples in retired and night workers. Pathol Biol. 55:205–07. doi:10.1016/j.patbio.2006.12.007. Okamoto-Mizuno K, Tsuzuki K. (2010). Effects of season on sleep and skin temperature in the elderly. Int J Biometeorol. 54:401–09. doi:10.1007/s00484-009-0291-7. Palinkas LA, Houseal M, Miller C. (2000). Sleep and mood during a winter in Antarctica. Int J Circumpolar Health. 59:63–73. Rubel F, Kottek M. (2010). Observed and projected climate shifts 1901-2100 depicted by world maps of the KöppenGeiger climate classification. Meteorol Z. 19:135–41. doi:10.1127/0941-2948/2010/0430. Sukying C, Bhokakul V, Udomsubpayakul U. (2003). An epidemiological study on insomnia in an elderly Thai population. J Med Assoc Thai. 86(4):316–24. Sunuttra T, Napatharin P, Wipawee K, Saejew A. (2008). The effects of Tai Chi on sleep quality, well-being and physical performance among older adults. Thai J Nurs Res. 12:1–13. 11 The Thai Meteorological Department. (2017). Climate. Monthly Weather Summary. [accessed 2017 Jun 9]. https://www.tmd.go.th/en/climate.php?FileID=4. Van Someren EJ, Swaab DF, Colenda CC, Cohen W, McCall WV, Rosenquist PB. (1999). Bright light therapy: improved sensitivity to its effects on rest-activity rhythms in Alzheimer patients by application of nonparametric methods. Chronobiol Int. 16:505–18. doi:10.3109/ 07420529908998724. Witting W, Kwa IH, Eikelenboom P, Mirmiran M, Swaab DF. (1990). Alterations in the circadian rest-activity rhythm in aging and Alzheimer’s disease. Biol Psychiatry. 27:563–72. doi:10.1016/0006-3223(90)90523-5. The World Health Organization (WHO). (1996). The World Health Organization Quality of Life (WHOQOL). Mental health. [accessed 2017 Apr 3]. http://www.who.int/mental_ health/media/en/76.pdf;. Wright KP Jr., McHill AW, Birks BR, Griffin BR, Rusterholz T, Chinoy ED. (2013). Entrainment of the human circadian clock to the natural light-dark cycle. Curr Biol. 23:1554–58. doi:10.1016/j.cub.2013.06.039. Yasunaga A, Togo F, Watanabe E, Park H, Park S, Shephard RJ, Aoyagi Y. (2008). Sex, age, season, and habitual physical activity of older Japanese: the Nakanojo study. J Aging Phys Act. 16:3–13. doi:10.1123/japa.16.1.3. Yetish G, Kaplan H, Gurven M, Wood B, Pontzer H, Manger PR, Wilson C, McGregor R, Siegel JM. (2015). Natural sleep and its seasonal variations in three pre-industrial societies. Curr Biol. 25:2862–68. doi:10.1016/j.cub.2015.09.046.