The association between daytime sleep and general obesity risk differs by sleep duration in Iranian adults

Abstract Background Sleep duration and daytime napping and obesity are related to adiposity; however, it is not clear whether the association between daytime napping and adiposity measures can differ by sleep duration. Aim To clarify the association between daytime napping and general and abdominal obesity based on sleep duration of participants. Subjects and methods This cross-sectional study was conducted on 1,683 individuals (837 men and 846 women) aged ≥ 35 years. Height, weight and waist circumference (WC) were measured according to the standard protocols. Body mass index (BMI) was calculated. Self-reported sleep duration (in a 24-hour cycle) was recorded. The odds of general and abdominal obesity were compared between nappers and non-nappers, stratified by their sleep duration (≤ 6 h, 6–8 h, ≥ 8 h). Results The mean (SD) age of participants was 47.48 ± 9.35 years. Nappers with a short sleep duration (≤ 6 h) had greater BMI and higher risk for overweight/obesity compared with counterpart non-nappers after adjustment for potential confounders (OR = 1.61, 95% CI = 1.07–2.41). In subjects with moderate sleep duration (6–8 h), nappers had a tendency towards higher BMI in comparison with non-nappers (28.04 ± 0.25 vs. 26.93 ± 0.51 kg/m2; p = 0.05), however, no significant difference was observed for the risk of obesity. Daytime napping was not related to the risk of obesity in long sleepers. No significant association was observed for abdominal obesity measures. Conclusions Daytime napping is associated with increased risk of overweight/obesity in short sleepers. However, in subjects with longer sleep duration, it is not related to the risk of overweight/obesity.


Introduction
the rising prevalence of obesity is one of the critical health issues worldwide (Basen-engquist and Chang 2011).WHo defines obesity and overweight for adults as a body mass index (Bmi) greater than or equal to 25 (Lartey et al. 2020).obesity is one of the most common metabolic diseases and linked to the increased risk of chronic diseases such as cardiovascular disease, hypertension and diabetes (Basen-engquist and Chang 2011).forty-four per cent of diabetic patients, 33% of ischaemic heart disease patients and 9-56% of cancer cases are attributed to obesity (Kyrgiou et al. 2017;Lartey et al. 2020).the relationship between sleep disorders (sd) and obesity has been well established (Beccuti and Pannain 2011;Hargens et al. 2013).
around one-third of each person's lifespan is allocated to sleep, as an essential need (sforza et al. 2010).short sleep duration and sleep disorders are a common health issue which affect quality-of-life (matricciani et al. 2012;Javadi et al. 2014).approximately one-third of adults have a night-time sleep duration of less than 6 h (Bonnet and arand 1995), which may disrupt their daily activities.elevated levels of ghrelin and insulin resistance and/or decreased levels of leptin may cause obesity in short sleepers (mai and Buysse 2008).the american academy of sleep medicine recommends at least 7 h sleeping per night as an essential need to keep adults in the optimal point of health (Watson et al. 2015).However, there is no specific guide to the best bedtime recommendation (Watson et al. 2015).according to earlier studies, individuals with short sleep at nights (≤ 5-6 h) are significantly more likely to be obese, which might be attributable to the changing in glucose tolerance, energy homeostasis and greater desire for calorie-dense and carbohydrate-rich foods (Gangwisch et al. 2005;Chaput et al. 2007;Xi et al. 2014). in addition to the biological effects, nocturnal sleep also affects the behavioural mechanisms related to eating patterns.for instance, short sleepers have more opportunities to eat and receive more calories (dashti et al. 2015).
short sleep duration can be a predictor of weight gain (Cappuccio et al. 2010) and, therefore, should be regarded in weight loss programs.in addition, daytime napping may be associated with greater waist circumference (WC) and Bmi (tse et al. 2021;yazdanpanah et al. 2021;Xiao et al. 2022).there is also evidence indicating that not only is daytime napping associated with greater risk of abdominal adiposity but also cannot decrease the adverse effect of short nocturnal sleep duration on obesity (yazdanpanah et al. 2021).a recent cross-sectional analysis from iran examined the association between anthropometric measures and sleep duration and daytime napping and revealed that short sleepers (< 8 h) had greater Bmi, WC, hip circumference, and waist-toheight ratios, but reported a positive correlation for these measures with daytime napping (tse et al. 2021).However, this question remains unanswered as to whether the association between daytime napping and adiposity measures can differ by sleep duration.therefore, in this cross-sectional study, we clarify the association between daytime napping and general and abdominal obesity based on sleep duration of participants in a randomly selected sample of iranian adults.

Study population
this cross-sectional study was conducted on 1,683 individuals (837 men and 846 women) in the framework of the third step of the isfahan cohort study (iCs).the iCs is an ongoing population-based longitudinal cohort study (sarrafzadegan et al. 2011), established in 2001 and included 6,504 adults (3,168 men and 3,336 women) aged ≥ 35 years.it aims to determine the individual and combined impact of various risk factors on the incidence of various cardiovascular events.the iCs was conducted in three districts of central iran and participants were recruited using a stratified cluster random sampling method (2,153 from isfahan, 1,028 from najaf-abad and 3,323 from arak).accordingly, target populations were categorised based on their residence area (rural vs. urban) considering the distribution of populations across the areas.these blocks were then randomly selected with regard to the number of households.finally, around 5-10% of the households and one eligible subject from each household in each cluster were randomly selected.further detailed descriptions about the study design have been presented elsewhere (sarrafzadegan et al. 2011).Briefly, at baseline, information about lifestyle factors, including dietary intake, sleep habits, smoking and physical activity, was collected using face-to-face interviews and all measurements were reassessed in the next surveys with a 6-year interval of follow-up (2007 and 2013).due to the lack of information in terms of sleep habits in the first two phases (2001 and 2007), in the current analysis, data from participants in the third phase (2013) who had complete information on sleep habits, anthropometric measures, and covariates were included.this study was approved by the ethics Committee of the research Council of isfahan Cardiovascular research Centre, a World Health organisation collaborating centre in isfahan, iran.

Data collection
trained health professionals completed a general questionnaire about demographic and socioeconomic variables, lifestyle factors including dietary intakes, smoking status (current, former, or never), and physical activity, and medical history through a 30-minute home interview (sarrafzadegan et al. 2009, 2013).Physical activity was assessed using a validated questionnaire (talaei et al. 2013).Body weight was measured to the nearest 0.5 kg using a seca scale while participants were minimally clothed.Height was measured while participants were barefoot and shoulders were in normal position using a plastic metre and recorded to the nearest 0.5 cm.Bmi was calculated by dividing body weight (kg) by height squared (m 2 ) (Zimmet et al. 2005).WC was measured by a non-elastic tape at the midpoint of the lowest rib and highest margin of the iliac (World Health organization 1995; national Heart, Lung, and Blood institute 1998).Body mass index ≥ 25 kg m 2 was classified as overweight and obese (national Heart, Lung, and Blood institute 1998).according to the adult treatment Panel iii (atP iii), abdominal obesity was defined as WC ≥ 88 cm in women and WC ≥ 102 cm in men (Cleeman et al. 2001).

Physical activity
Physical activity level was assessed using a validated iranian version of the iPaQ.this questionnaire assesses physical activity level in four main domains including leisure time, occupational, household and transportation physical activities.the duration and frequency of various physical activities over a day or week were asked and, based on the metabolic equivalent of the task (met), the physical activity level for each domain was estimated.the sum of figures achieved for all domains was considered the total physical activity level (talaei et al. 2013).

Dietary assessment
the habitual dietary intake of participants over the preceding year was assessed using a validated 48-item food frequency questionnaire (ffQ) (mohammadifard et al. 2015).as the present study is an intervention programme like the Cindi programme, the ffQ was adapted from the Cindi programme questionnaire (Leparski and nüssel 1987).overall diet quality was measured by the global dietary index (Gdi), consisting of 29 food items categorised into seven main food groups.food groups were (1) fast foods, (2) fruit and vegetables, (3) beans, soy, chicken and fish, (4) sweets, (5) hydrogenated oil, ghee, animal fats or butter, (6) meat, egg, high fat dairy products and (7) non-hydrogenated oil, olive oil.Based on the average frequency of consumption, each food group obtained a score of 0, 1 or 2. Higher frequencies of healthy food groups were given a higher score and vice versa.Higher values represent an unhealthier dietary pattern (mohammadifard et al. 2009).

Sleep habits
self-reported sleep duration (in a 24-hour cycle) was determined by trained interviewers.usual siesta or nap duration was also recorded.Participants were asked 'how many hours do you normally sleep in a 24-hour cycle at night and how many hours do you normally sleep in a 24-hour cycle during the daytime?' sleep duration less than or equal to 6 hours per 24 hours was defined as short sleep.Long sleep was defined as more than 8 hours/24 hours, and any time between 6-8 hours/24 hours was defined as standard daily sleep duration (fernandez-mendoza et al. 2019).individuals who reported only night-time sleep were considered non-nappers and individuals who reported both daytime (bedtime between 6 am and 8 Pm) and nocturnal sleep were considered nappers.night-time sleep was defined as at least 1 hour and a maximum of 15 hours and daytime nap/sleep from 30 minutes up to 5 hours (Wang et al. 2022).the sleep durations outside our defined range, were excluded (n = 47).a total of 158 subjects had missing data on sleep duration.

Statistical analysis method
Participants were categorised into three main groups based on their sleep duration per 24 hours: ≤ 6 hours, 6-8 hours and ≥ 8 hours.each category was also divided into two different sub-categories based on having daytime napping/ sleep.Continuous variables were reported as mean ± standard deviation and categorical variables as number and percentage.the normality of data was checked by Kolmogorov-smirnov test and Q-Q diagram.Quantitative variables and normal distribution were examined by using the independent t-student test.the continuous variables with abnormal distribution were analysed using the non-parametric mann-Whitney test.the distribution of categorical variables between nappers and non-nappers based on their sleep duration per 24 hours was compared using Chi-square test.Bmi and WC means were estimated through independent sample t-test in the crude model and anCoVa in the adjusted models.
the odds ratio (or) and 95% confidence interval (Ci) of overweight or obesity in nappers compared with non-nappers for each sleep duration category were estimated through logistic regression.this association was also examined in three adjusted models.in model 1, the effect of age and sex was adjusted.in model 2, the effects of education, physical activity and smoking status were additionally adjusted.further control for diet quality was made in model 3. all tests were performed in three categories of sleep hours, including sleep hours less than or equal to 6 hours, between 6 to 8 hours and more than or equal to 8 hours and sPss 22 software was used for data analysis.p < 0.05 was regarded as the significance level.

Results
overall, 1,683 individuals (49.7% male) entered our study.the mean (sd) age of participants was 47.48 ± 9.35 years with the interquartile range (iQr) of 34-82.sleep duration in 24 hours ranged from 4 to 15 hours, while the corresponding values for daytime napping were from 0.5 to 4 hours.nocturnal sleep duration was between 3.5 and 14 hours.means (sd) of sleep duration, daytime napping, and nocturnal sleep were 7.7 (1.6), 1.3 (0.7), and 6.4 (1.4) hours, respectively.
the demographic characteristics of participants in three groups of sleep duration based on having daytime napping or not are shown in table 1. in all groups, daytime nappers were older (p < 0.05). in groups with a sleep duration of 6 to 8 hours (p = 0.01) or greater than or equal to 8 hours (p = 0.02), daytime nappers were more likely to be low-income.nappers in the long sleep duration group were less likely to be smokers (p = 0.02).Physical activity level ranged from 0.0 to 9024.3 met.h/week, with a mean of 743.0 (611.9).
table 2 shows the mean Bmi and WC in nappers and non-nappers based on sleep duration.in the crude model, in individuals who slept ≤6 hours, Bmi tended to be higher in daytime nappers than non-nappers (28.54 ± 0.29 vs. 27.81 ± 0.27 kg/m 2 ; p = 0.06).this association became significant after adjustment for age and sex, and further control for education, physical activity and smoking and diet quality only slightly changed it (28.59± 0.28 vs. 27.80 ± 0.26 kg/m 2 ; p = 0.04). in individuals with a sleep duration of 6-8 hours a tendency towards higher Bmi in nappers compared with non-nappers was observed in the fully adjusted model (28.04 ± 0.25 vs. 26.93 ± 0.51 kg/m 2 ; p = 0.05).However, in those with long sleep duration (≥8 hours), no significant difference was found between nappers and non-nappers.regarding WC, despite a significant difference between nappers and non-nappers with short sleep duration in the crude model (98.71 ± 0.7 vs. 96.69± 0.65 cm, respectively; p = 0.03), adjustment for potential confounders removed this association.no significant difference in WC was observed between nappers and non-nappers in those who slept more than 6 hours.
the results of multiple logistic regressions for general obesity are shown in table 3. in the crude model in short sleepers, nappers tended to have 44% greater risk for overweight and general obesity compared with non-nappers (95% Ci = 0.99-2.09;p = 0.06).adjustment for age and sex strengthen this association and remained significant even after further control for education, physical activity, smoking and diet quality (or = 1.61, 95% Ci = 1.07-2.41;p = 0.02). in subjects with longer sleep duration (>6 hours), no significant association was observed, though in the fully adjusted model in subjects with a sleep duration of 6-8 hours per 24 hours this association was marginally significant (or = 1.62, 95% Ci = 0.93-2.82;p = 0.09).
table 4 shows the or and 95% Ci for abdominal obesity in nappers compared with non-nappers considering their sleep duration.no significant association was found between abdominal adiposity and daytime napping in any of the sleep duration categories.

Discussion
our study analysis indicated that daytime napping/sleep compared to non-nappers is associated with higher mean Bmi.moreover, daytime napping/sleep was more probably associated with higher risk of overweight/obesity in short sleepers (≤6 hours per 24 hours).no significant association was observed between daytime napping and abdominal obesity considering sleep duration.
according to earlier studies in different geographical regions, short sleep duration (generally <6 hours per night) is associated with obesity risk in adults and children (Patel and Hu 2008;Leproult and Van Cauter 2010).However, the number of studies comparing night sleep with daytime napping in relation to obesity is small (sayón-orea et al. 2013;Chen et al. 2018).Less than 5 hours of sleep duration per night is significantly related to obesity, especially general obesity.Long daytime napping not only increases the risk of abdominal obesity, but also cannot rectify short night-time sleep (dashti and ordovás 2021).there are some studies which failed to find any significant association between nocturnal sleep duration of 6 hours or longer and risk of any type of obesity (Wu et al. 2014;tse et al. 2021).People who work night shifts and have short night-time sleep duration are at higher risk for obesity.the underlying reasons for this can be related to the effect of night shifts on circadian rhythm, especially, sleep rhythm and sleep disturbances.moreover, the HPa axis (hypothalamic-pituitary-adrenal axis) activation and the autonomic nervous system dysregulation are associated with inflammation which can affect obesity risk per se (akerstedt 1998;Björntorp and rosmond 2000).Light exposure at night is also associated with late bedtime.although the studies' reports are mixed, it seems late bedtime behaviour is a risk factor for obesity, independent of sleep duration (sasaki et al. 2018; sunwoo et al. 2020).according to recent studies, daytime napping can disrupt circadian rhythm, leading to obesity development (Patel et al. 2014;Wang et al. 2019).
increased energy intake in people with sleep disorders, who have short sleep duration, may be the principal reason for obesity (taheri et al. 2004;Gangwisch et al. 2005;schmid et al. 2009;Brondel et al. 2010).one study reported that people with late bedtime received about 250 kcal (mostly at dinner time and after that) more than those who went to bed earlier (Baron et al. 2011).the results of the nHanes study showed that people who slept on average 5 hours a night were 26% more likely to have fizzy drinks and 18% more likely to drink non-carbonated sweeteners compared to individuals with longer sleep duration (Prather et al. 2016).these findings are in line with other studies (Baron et al. 2011;franckle et al. 2015;min et al. 2018).
Circadian clock influences eating behaviour and body weight through some temporal alterations.for instance, shift working increases the risk for obesity by changing usual eating time and disrupting the circadian rhythm.the balance of energy regulatory metabolic hormones is influenced by sleep.short sleep duration can decrease blood leptin levels, which suppresses appetite and increases ghrelin levels, which stimulates appetite (spiegel et al. 2004; taheri et al. 2004). in addition, sleep deprivation is associated with insulin resistance, higher levels of pro-inflammatory cytokines, oxidative stress, and lower levels of melatonin which altogether can promote fat accumulation (Padilha et al. 2011).although we observed that, regardless of sleep duration, daytime nappers had greater WC compared with non-nappers, this difference did not reach a significant level.to date, many studies have examined the relationship between sleep duration and abdominal obesity, but the results are not conclusive.WC and sleep duration are inversely related, while a mean difference of 9 cm was observed between those with a sleep duration of <5 hours and those with a sleep duration of >8 hours (theorell-Haglöw et al. 2010).although most of the studies suggested an inverse association (theorell-Haglöw et al. 2010;Liu et al. 2018;Brum et al. 2020), there is some research reporting a null association (rabanipour et al. 2019). in addition, the association between daytime napping and abdominal obesity has been investigated by several studies. in a Chinese study on middle-aged and elderly people, a longer siesta was associated with greater WC compared with those who did not have regular siesta (tang et al. 2021).the results of a cohort study showed that daytime napping was significantly associated with type 2 diabetes mellitus mediated by the effect of napping on adiposity indicators (Xiao et al. 2022).daytime napping (more than or equal to 30 minutes) also had a strong relation with cardiovascular disease and hypertension incidences (Wang et al. 2022).However, far fewer studies have examined the association of sleep duration and daytime napping with obesity (sayón-orea et al. 2013).this cohort study revealed that taking a siesta for 30 minutes/day is associated with a 33% reduction in obesity compared with those with no siesta (sayón-orea et al. 2013).However, they failed to examine the associations for siesta based on sleep duration.to our knowledge, this is the first study which considered the simultaneous association of daytime napping and sleep duration with obesity.However, it comes with some limitations including self-reported sleep duration, physical activity and dietary habits, and the lack of data on shift work, sleep apnoea and menopausal status.although our study population was large, when they were categorised based on napping and sleep duration, only a few subjects were placed in some categories which may fail to indicate any real association.moreover, the cross-sectional design of the present study does not allow us to draw any causal relationship.earlier studies have mainly defined short sleep duration based on night-time sleep duration, however, due to the imbalanced distribution of participants across different categories of sleep duration and daytime napping, it was not possible for us to define short sleep duration only based on night-time sleep.this limitation also did not allow us to explore this association stratified by sex.although we tried to control the effect of some relevant confounding factors, the effect of some unmeasured or residual factors cannot be ruled out.
in conclusion, the results of this study show that taking a nap during the day increases the risk of overweight/obesity in those with a short sleep duration.therefore, recommendations on not having a nap might be an efficient strategy in weight loss programs.However, prospective cohort studies are required to explore this association in larger sample sizes.

Table 1 .
Characteristics of the study participants by sleep duration and napping habit.
Derived from Kruskal-Wallis == → (mann-Whitney u test) test for continuous and Chi-square test for categorical variables.

Table 2 .
Comparison of mean body mass index and waist circumference based on non-nappers and nappers sleep.a

Table 3 .
Crude and multivariable-adjusted odds ratios and 95% Cis for overweight and obesity in non-nappers compared with nappers.

Table 4 .
Crude and multivariable-adjusted odds ratios and 95% Cis for abdominal obesity in non-nappers compared with nappers.