Association between difficulty initiating sleep in older adults and the combination of leisure-time physical activity and consumption of milk and milk products: a cross-sectional study

A total of 437 community-dwelling, older Japanese individuals participated in the survey. The participants, aged 65 years and older, were recruited in Tsuchiura and Kasama Cities in Ibaraki prefecture, Japan in 2013. The Tsuchiura City participants consisted of 160 community dwellers (mean age 75.6 ± 6.2 years, male 34.0%) recruited through local advertisements and flyers or invited by municipal employees from June to November 2013. For Kasama City, we used data from 277 randomly selected individuals (mean age 74.5 ± 5.0 years, male 49.6%) who participated in the “Kasama study” [15] from July to August 2013. We started with 437 participants, but we rejected 16 individuals due to incomplete data. There were 421 final participants (mean age 74.9 ± 5.5 years, male 43.7%) in the data analysis. All participants provided a signed informed consent. The Ethics committee of the University of Tsukuba approved this study.

To assess the amount of LTPA, we used the Japanese version of the Physical Activity Scale for the Elderly (PASE) [16]. A previous study deemed the PASE to be reliable and valid [16]. LTPA contains five activities: walking (for both recreation and transportation); light-, moderate-, and vigorous-intensity recreational activities; and muscle strength training. We measured a participant’s frequency (day/week) and duration (minute/day) of each activity during the previous 7 days and calculated a total LTPA score using frequency, duration and weight for each activity used in the analysis.

We assessed the consumption of milk, yogurt and cheese during the previous couple months using a self-reported questionnaire and the following questions: “How many days a week do you drink milk (or eat yogurt/cheese) [day (s)/week]?” and “How much milk (or yogurt/cheese) do you usually consume each day [milliliters or grams/day]?” Additionally, we supplied examples of portion size (e.g. a glass of milk = 200 ml) to help participants answer the questions. Using these two questions, we calculated consumption of each food item, and the sum of the three items equaled a participant’s total dairy consumption used in the analysis. To estimate total dairy consumption, units of milk intake were converted from milliliters to grams using the following equivalency: 100 ml = 103.2 gm” [17]. No participants had a milk allergy. Previous studies have used self-reported questionnaires to evaluate the frequency and the portion size of daily milk and milk product consumption; those authors confirmed a moderate correlation (r = 0.4–0.6) with the weighed dietary recording considered the gold standard for assessing dietary methods [18, 19].

We gathered information on total sleep time and sleep latency during the previous month and evaluated these data by referencing the Pittsburgh Sleep Quality Index (PSQI) [20]. PSQI is used to identify sleep disorders in both clinical and research fields throughout the world, and validity and reliability have been confirmed [20].

We referenced previous studies to determine minimum required potential confounders for our study: age, gender, body mass index (BMI), hypnotic use for sleep [21], engaging in LTPA [22] and dairy consumption [23, 24]. We obtained age, gender and hypnotic use with a self-report questionnaire. We measured height and weight with a stadiometer or a digital weight scale and calculated BMI calculated as weight in kg divided by height in meters squared.

LTPA participation was divided into 3 groups based on median level of engagement or no engagement at all: 1) nothing, 2) low level and 3) high level. Total dairy consumption was divided into 3 groups using tertiles: 1) low level, 2) moderate level and 3) high level. A variable composed of the following 4 groups combining LTPA and intake of dairy products was created: 1) LTPA with large dairy intake, 2) LTPA with small dairy intake, 3) no LTPA with large dairy intake and 4) no LTPA with small dairy intake. By referencing previous studies [25], people who took more than 30 min to fall asleep were defined as having DIS.

We analyzed group differences (non-DIS vs. DIS), using independent t-test and chi-square test. We used logistic regression analysis to examine the association between participation in LTPA, consumption of dairy products or a combination of those two variables (dependent variables) with DIS (independent variable) adjusting for age, gender, hypnotic use, BMI and those LTPAs or dairy products which were not used as dependent variables. In addition to the odds ratios, we also calculated P for trend. Statistical analysis was performed by IBM SPSS Statistics version 19.0 with the level of significance set at P < 0.05.

The characteristics of study participants are shown in Table 1. The mean age was 74.9 ± 5.5 years, and the subjects’ usual sleep time was 405.9 ± 73.4 min/night. Out of 421 participants, 135 individuals met the definition of DIS (32.1%) and 80 people (19.0%) habitually used sleeping pills. Compared with participants who did not have DIS, individuals defined as having DIS were more likely to be women, experienced shorter total sleep duration and long sleep latency, engaged less in LTPA and were more likely to use hypnotics (P < 0.05).

An association between LTPA and DIS is presented in Table 2. After adjusting for age, gender, hypnotic use, BMI and consumption of milk and milk products, a high level of LTPA was associated with decreased prevalence of DIS compared to no engagement in LTPA (OR = 0.48, 95% CI = 0.25-0.93).

As shown in Table 3, participants were less likely to perceive DIS when they consumed a high level of milk (OR = 0.47, 95% CI = 0.26-0.84) even after adjusting for age, gender, hypnotic use, BMI, LTPA engagement and yogurt and cheese intake. Moreover, a significant dose–response relationship was observed between milk intake and prevalence of DIS (Trend P < 0.05). There were no significant associations between consumption of yogurt, cheese or total dairy and DIS.

As Table 4 indicates, after controlling for age, gender, hypnotic use, BMI and yogurt and cheese consumption, participants who engaged in LTPA as well as consumed milk were less likely to complain of DIS compared with those who neither engaged in LTPA nor consumed milk (OR = 0.27, 95% CI = 0.10-0.73). Similarly, after adjusting for multiple confounding variables, we found that the combination of LTPA engagement and cheese consumption was associated with decreased prevalence of DIS (OR = 0.34, 95% CI = 0.14-0.85). We did not find a significant association between DIS and the combination of LTPA and yogurt or total dairy consumption.

In the present study, complaints of DIS were reported by 37.1% of women and by 25.5% of men, which is similar to other Japanese studies [1]. Furthermore, since a large body of evidence also indicates there is a higher prevalence of insomnia in older women than in older men [26], the rates of insomnia in the study participants do not appear to be specific to our study population.

In this study, sufficient LTPA was associated with a lower prevalence of DIS, which is in accord with a large amount of research showing that exercise improves sleep latency [10, 11, 27] and physically fit individuals have shorter sleep latency [28]. Although the mechanisms are not clear, earlier studies have suggested that physical activity may improve sleep due to 1) body temperature increases, 2) energy expenditure, 3) antianxiety and antidepressant effects and 4) entrainment of circadian rhythm [22, 29]. The improvement we saw in DIS in our study may be a consequence of these physiological and psychological effects of LTPA.

We found a positive link between decreased prevalence of DIS and the intake of milk, but not yogurt, cheese or total dairy consumption. At present, there are divergent opinions as to the effects of milk on sleep; some researchers found that consuming milk improves sleep quality [30, 31], whereas, others reported no significant effects of milk intake [14, 32]. Yamamura et al. demonstrated a significant improvement in sleep efficiency after ingesting melatonin-rich milk before bedtime [14]. From results of our study, we suggest that daily intake of ordinary commercial milk may also facilitate falling asleep. Our findings confirmed the beneficial effect that commercial milk consumption can have on the prevalence of DIS, which is useful information for the older adult population.

We also found a significant linear trend between drinking milk and a lower prevalence of DIS, that is, DIS became less frequent as the amount of milk consumed increased. To our knowledge, literature concerning the association between sleep and milk/milk products intake in older people is limited. In particular, the dose–response relationship between these factors is still unclear. Hence, our findings contribute to this developing area of research.

To our knowledge, although some studies have investigated the relationship between sleep and either LTPA or milk/milk products, no prior study has evaluated the association between DIS and LTPA in conjunction with dairy intake. Our research produced interesting and novel findings as it shows that older adults who engaged in LTPA and consumed milk or cheese reported DIS less frequently than people who neither engaged in LTPA nor consumed milk or cheese. Additionally, individuals who only engaged in LTPA or only consumed milk or cheese had higher odds ratios than people with both habits. The combination of LTPA and milk or cheese consumption may be an effective tool to help people fall asleep.

In modern society, older adults frequently complain of insomnia symptoms including difficulty falling asleep [1]. They may turn to medication to attain sleep, although medications have a greater risk of side effects e.g. falling [33] and even death [34]. Attele et al. showed that older people are often reluctant to take sleeping pills because of the negative impacts on their health [35], hence, non-pharmacological therapy with less adverse effects is needed. Along with the beneficial effects on sleep, LTPA provides other health benefits such as maintaining physical function [36] and decreased risks of mortality [37] and dementia [38]. Milk and milk products also contribute significant amounts of calcium, which plays an important role in reducing risks of osteoporosis [39] and hypertension [40]. Due to these benefits, participating in LTPA and consuming milk or dairy foods are recommended for well-being in later life. Furthermore, taking into account our findings on sleep improvement, we suggest that habitual LTPA along with milk or cheese consumption are helpful not only for maintaining general health, but also as a non-drug intervention for DIS in older adults.

There are some limitations in our study. First, the generalization of the study’s results is uncertain. Almost all study participants entered the study after receiving an invitation letter or seeing a local advertisement. Thus, the study may underestimate the proportion of older adults with poor living habits (e.g. inactive, low consumption of dairy products or insomnia) or those in poor health. Additionally, sample size was small, specifically with regards to people who did not participate in LTPA and/or consume dairy products; this limits confidence in the results. To verify our results, future investigations with larger sample sizes in unhealthy, older adults are needed. Second, we did not assess depressive symptoms, cognitive function/dementia, chronic illness (e.g. heart disease or lung disease) [41, 42] or a number of medications [43] which could be potential confounders for our results. Third, since LTPA, dairy consumption and sleep were assessed via self-reported questionnaires, there may be recalling/reporting bias in our data. In particular, Silva et al. suggested that sleep latency might be overestimated by a self-reported measure [44], thus, future studies should include objective measures of sleep, such as the use of actigraphy. Fourth, because of the present study’s cross-sectional design, we could not prove a causal relationship. There may be a bidirectional relationship between physical activity and sleep [45, 46]. To reveal a causal relationship, a prospective cohort study and an intervention study are needed. Finally, unfortunately, we did not assess the times of day that participants’ practiced LTPA and consumed milk/dairy products. The effect that exercise has on initiating sleep differs depending on when it was performed [22]. Similarly, ingesting milk and other milk products, which contain several nutrients with potential sleep-promoting properties, may be effective for falling asleep when these food items are consumed before bedtime.

Because earlier studies have revealed that LTPA and dairy intake separately have positive effects on DIS, it is reasonable to conclude from our findings that LTPA in combination with milk or cheese intake may help decrease DIS, even with the above study limitations.