Insomnia in Indonesia Older Adults: The Role of Mental Health, Sociodemographic Status, and Physical Function
Article information
Abstract
Background and Objective
Insomnia can cause impairment in physical, mental, and social functioning, which contributes to considerable healthcare and social challenges. This study aims to investigate various potential associated factors of insomnia among older adults in Indonesia.
Methods
Publicly available data of the fifth wave of the Indonesia Family Life Survey (IFLS-5) was taken cross-sectionally from 2014 to 2015 with a multistage stratified sampling method. This data included 4236 Indonesian older adults aged of 60 and older from all over Indonesia. Numerous social demographic and wellbeing variables that were gathered through assessments and surveys were examined. The evaluation of despair and its contributing components was done using multivariate logistic regression.
Results
Older adults with low education (adjusted odds ratio [AOR]: 1.73, 95% confidence interval [CI]: 1.16–2.60), poor subjective economic status (AOR: 1.52, 95% CI: 1.13–2.05), having low life satisfaction (AOR: 1.46, 95% CI: 1.14–1.87), being self-perceived as unhealthy (AOR: 2.26, 95% CI: 1.80–2.84) or lonely (AOR: 1.77, 95% CI: 1.34–2.35), physically dependent measured by Activity Daily Living (AOR: 1.46, 95% CI: 1.01–1.96), having arthritis (AOR: 1.40, 95% CI: 1.04–1.85) and depression (AOR: 4.14, 95% CI: 3.21–5.33) are associated to insomnia.
Conclusions
Depression has the strongest association to insomnia in elderly. Involving older adults in cognitive and social activities, providing supports for lonely individuals, improving their physical function, prompt treatment of arthritis and symptomatic pain may reduce the risk for insomnia. Early detection and providing mental services including social activities to depressed older adults may also decrease the risk of insomnia.
INTRODUCTION
Insomnia is described as a sleep disorder typified by unhappiness with the quality or amount of sleep in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). The indications of insomnia must show at least 3 nights per week for at least 3 months [1]. Insomnia symptoms can be difficulty in initiating sleep, maintaining sleep, or early awakenings with the inability to return to sleep [2]. It affects the quality of life negatively, regardless of age and sex. Insomnia can cause impairment in physical, mental, and social functioning, which contributes to considerable healthcare and social challenges [2].
The prevalence of insomnia varies among studies. Older adults, people aged 60 years and over, who reported insomnia symptoms ranged from 40%–75%, while the prevalence of insomnia ranged from 3%–40% in those over 60 years [2-4]. Aging causes homeostatic and circadian rhythm changes that regulate sleep and wake cycles to diminish, thus increasing the risk of insomnia [2]. Poor sleep quantity or quality in older adults can cause daytime sleepiness and fatigue associated with multiple health, psychological, and social problems [3]. Previous studies showed that older adults with no or little education, with low socioeconomic status, low social capital or engagement, who were living alone, and reporting low spirituality were more likely to report insomnia. Older adults with insomnia also experienced a higher number of chronic health conditions, falls, depression, and cognitive impairment [2,3,5].
Indonesia as the country with the fourth largest population in the world has an estimated 16 million older adults, around 6.1% of the total population. Due to its critical role in sustaining both mental and physical health, sleep has gained particular attention. Yet, the prevalent sleep issues within older persons are made worse by age-related health concerns. This poses many challenges to the country, to maintain healthy aging of the older population and prevent various detrimental effects that result from aging, including insomnia. Poor sleep quality has also been linked to occupational limits in older persons, according to earlier research [6,7]. It is possible that the sleep quality has an impact on cognitive function, which in turn has an impact on one’s capacity to carry out everyday tasks. There have not been much study undertaken on the relationship among poor sleep, Activity of Daily Living (ADL), and emotional stress. The results show that ADL may function as a mediator. This study aims to investigate various potential predictors of insomnia, in the hope of providing information for other researchers, clinicians, psychologists, social workers and other formal stakeholder, and anyone that provides care directly or indirectly to older adults, so that action to modify those predictors can be taken.
METHODS
Study Designs and Participants
Publicly available data of the fifth wave of the Indonesia Family Life Survey (IFLS-5) were used. IFLS-5 was taken cross-sectionally from 2014 to 2015 with a multistage stratified sampling method, involving individuals, families, households, and communities. This data included 4236 Indonesian older adults aged 60 and older from different areas in Indonesia.
Ethical Approval
The IFLS surveys and protocols underwent scrutiny and endorsement by Institutional Review Boards (IRBs) in both the United States at RAND Corporation and in Indonesia at the University of Gadjah Mada (UGM) for IFLS-3, IFLS-4, and IFLS-5. The IRBs ensured that all prerequisites for consent concerning older adults were fulfilled and sanctioned prior to commencing the work (available at https://www.rand.org/well-being/social-and-behavioral-policy/data/FLS/IFLS.html).
Measurements
We followed method used by Pengpid et al. [8] and Peng et al. [9].
Insomnia
Insomnia was assessed using 5 items from the Patient-Reported Outcomes Measurement Information System (PROMIS) on sleep disturbance and sleep impairment measures with a total score of 40 [10]. Insomnia was assigned to participants with a score higher or equal to 21.
Sociodemographic characteristics
Age, sex, relationship status, education, place of residence, region, and subjective financial standing were sociodemographic characteristics. Age was classified into 1) 60–69 years, 2) 70–79 years, and 3) ≥80 years. Marital status was classified into 1) married or coinhabiting and 2) never married, separated, widowed, or divorced. Educational status was classified into high educational status if the participant finished education at least on the high school level and low educational status if they finished below that level. The region was classified into 1) Sumatra, 2) Java, and 3) other regions. Subjective economic status was assessed using the question “Please picture a six-step ladder with the lowest people at the bottom and the wealthiest ones at the top. Which financial level are you now at?” with options ranging from poorest (step 1) to richest (step 6), which was classified into poor (step 1 and 2), medium (step 3 and 4), and rich (step 5 and 6). Tobacco use was assessed using questions “Have you ever smoked cigar/cigarettes, smoked self-enrolled cigarettes, smoked a pipe, or chewed tobacco?” with options 1) no and 2) yes, and “Do you still have the habit or have you quit?” with option 1) quit and 2) still have. Tobacco use was classified into 1) never or former and 2) current smoker.
Anthropometric measurements
Heights and bodyweight measurements were obtained using conventional equipment to compute body mass index (BMI), which was then categorised using the Asia Pacific Classification into 1) underweight, 2) normal, and 3) overweight/obese [11].
Sense of well-being and mental health
Life satisfaction was assessed using the question “Please, think about your life as a whole. How satisfied are you with it?” with options ranging from completely satisfied (1) to not at all satisfied (5). This variable was classified into 1) high life satisfaction (option 1 to 3) and 2) low life satisfaction (option 4 to 5). Subjective-health status was assessed using the question “Please, think about your health. How satisfied are you with it?” with options ranging from very healthy (1) to unhealthy (4), which was classified into 1) healthy (option 1 and 2) and 2) unhealthy (option 3 and 4). Components from the preceding 12 months were used to measure social capital, including participating neighborhood events, volunteering, visiting community meetings, and engaging in religious activities. High social capital was assigned to the participant with participation in at least one of these activities. Loneliness was assessed using a question from the Center for Epidemiologic Studies Depression Scale, 10-item version (CES-D-10), which was “How often did you feel lonely in the past week?” with options ranging from rarely or none (1) to most of the time (4) [12]. Loneliness was classified into 1) no (option 1 to 2) and 2) yes (option 3 to 4). Depression was assessed using items from CES-D-10 with a score higher than 10 was classified into clinical depression [12].
Physical functioning
Functional dependency is assessed by 6 items from the Katz ADL index, which includes items such as feeding, continence, transference, toileting, dressing, and bathing [13]. The Lawton Instrumental Activity of Daily Living (IADL) has 6 components which are handling finance, taking medications, laundry, housekeeping, food preparation, and shopping. ADL and IADL were classified into independent and dependent with having difficulty in at least one of the items indicating dependency [14].
Physical activity was assessed using items from the International Physical Activity Questionnaire short version (IPAQ-S7S) from the question “Do you regularly engage in physical activity?” Classification of physical activity was done based on IPAQ protocol into low, moderate, and high [15]. Falls were assessed using the question “In the last two years, have you ever fallen?” with the classification of 1) no and 2) yes.
Chronic conditions
Chronic conditions were assessed by asking participants if they had received a diagnosis from a healthcare professional for various conditions such as hypertension, diabetes, tuberculosis, chronic pulmonary disease, heart disease, stroke, kidney disease, arthritis, cancer, or high cholesterol. These conditions were then grouped into either 1) “none” or 2) “one or more chronic illnesses.” The study did not take into account the treatment status of the participants. Dementia was assessed using items from Telephone Survey of Cognitive Status (TICS) [8]. The items include awareness of date and day of the week, self-reported memory question, subtraction of 7s from 100, immediate and delayed word recall of 10 nouns with a total score of 34. A score below 8 was considered possible dementia.
Statistical Analysis
Frequency analyses were used for descriptive analyses of the sample. We performed binary logistic regression for bivariate analysis to examine the relationship between the dependent variable (insomnia, categorical variable) and each independent variable separately (categorical variable). Subsequently, we carried out multivariate analysis using multivariate logistic regression, focusing on the significant findings (p < 0.2) from the bivariate analysis to calculate the adjusted odds ratio. The results were presented with p values and 95% confidence interval (CI), and a p-value lower than 0.05 was considered significant. The analyses were done using IBM SPSS Statistics, version 22 (IBM Corp., Armonk, NY, USA).
RESULTS
Characteristics of Sample
A total sample of 4236 older adults was collected (Tables 1 and 2). Insomnia was found in 10.5% of participants. Most participants were in the 60–69 years old group (70.3%), followed by 70–79 years old (25.3%) and 4.4% of older adults with an age of 80 and older. The distribution of sex was equal, with slightly more females (50.3%) than males (49.7%). Most of the participants were married or coinhabiting (66.6%). Only 15% of older adults had high education. There were slightly more participants from urban areas (50.8%) and most of them lived on Java (78.2%), followed by Sumatra (12.6%) and other regions outside Sumatra and Java (9.1%). Participants with medium and poor subjective economic status comprised 39.9% and 33.7% of the total population, respectively. Dependent participants measured by ADL and IADL contributed 12.7% and 25.6% of the total population respectively, 33% were overweight/obese and 26% were underweight, 17.8% has low life satisfaction, 34.6% perceived themselves as unhealthy, and 17.7% had low social capital. Loneliness was reported in 11.8% of older adults, 33.5% were smokers, and 16.3% had depression. Almost half of the total participants have one or more chronic conditions (44.9%) (Tables 2 and 3), including arthritis (12.1%), hypertension (27.6%), diabetes (6.7%), tuberculosis (1.0%), chronic pulmonary disease (2.2%), cardiac disease (4.1%), stroke (2.4%), dementia (40.7%), kidney disease (1.5%), cancer (0.6%), and dyslipidemia (6.9%). A total of 47.2% participants had low physical activities, and 12% had experienced falls.
Characteristics of Indonesian participants aged 60 or older
Prevalence of insomnia in Indonesian older adults
Prevalence of insomnia based on chronic diseases
Factors Associated with Insomnia
Bivariate analyses showed that insomnia was significantly associated with being female, having low education, poor self-perceived socioeconomic status, low life satisfaction, being unhealthy, lonely, or dependent, and having one or more chronic conditions and depression (Table 4). Insomnia was not associated with age, marital status, residential status, living region, BMI, social capital, tobacco usage, physical activities, and falls in bivariate analysis (unadjusted).
Multivariate logistic regression analysis of factors associated with insomnia in older adult
Older adults with low educational status had 1.73 times (95% CI: 1.16–2.60) higher prevalence of insomnia (Table 4). Poor subjective economic status increased insomnia prevalence1.52 times (95% CI: 1.13–2.05). Being overweight/obese also increased the prevalence of insomnia by 1.31 times (95% CI: 1.02–1.70). Having low life satisfaction (AOR: 1.46, 95% CI: 1.14–1.87), being self-perceived as unhealthy (AOR: 2.26, 95% CI: 1.80–2.84), lonely (AOR: 1.77, 95% CI: 1.34–2.35), and dependent measured by ADL (AOR: 1.46, 95% CI: 1.01–1.96), all significantly increased prevalence of insomnia. Overall, having one or more chronic condition did not increased the prevalence of insomnia, however, additional analysis showed that older adults who had arthritis were also more likely to report insomnia (AOR: 1.40, 95% CI: 1.04–1.85) (Table 5). Depressed older adult had a 4.14 increased prevalence of insomnia (95% CI: 3.21–5.33) compared to older adults who did not have depression.
Multivariate logistic regression analysis of chronic condition and insomnia in older adult
DISCUSSIONS
Our research showed that the prevalence of insomnia was higher among low educated, unhealthy status, poor, overweight/obese, low self-satisfaction, lonely, dependent, and depressed older adults. Having arthritis was also increase the prevalence of insomnia. Insomnia was found in 10.5% of the total participants. This number was similar to the national survey in India, which was 12.7% in total, 13.2% among women, and 11.9% among men [16].
Age, marital status, and residential status were not found to be risk factors for insomnia in our study, but sex was. Studies by Dangol et al. [17] and Pengpid and Peltzer [16] also showed no significant association between sex and insomnia, however, difficulty in initiating sleep was more prevalent in females compared to males. Other studies showed that the risk of insomnia increased with age, being female, being divorced, widowed, or not married at all [4,5]. Study by Akberzie and Kataria [18] found that, the risk raised to 73% for adults above 65 years old. After menopause, levels of estrogen and progesterone vary, which has a significant impact on how well older women sleep. Women were found to be more at risk due to being more prone to depressive states, stressful events, and psychological distress [19]. A study by Pengpid and Peltzer [16] indicates that older adults who lived in urban areas were protected against insomnia compared to those living in rural areas.
Older adults with low education had raised the probability of insomnia in this research, which is similar to several other studies [4,17]. The possible mechanisms are having a better income which is associated with access to healthcare and medication, having better skills at problem-solving, good networking, and a better position (with more control and less stress) in society, but after adjusting for several of these covariates, low education still independently increased the risk of insomnia [3]. Poor economic status was also an independent risk factor for insomnia in our study, concurrent with the result from Ma et al. [20]. Older adults with low life satisfaction were more likely to report insomnia in our study, which is similar to the Nepali study done by Khagi et al. [21]. However, the study by Pengpid and Peltzer [16] showed that there were no associations between both variables.
Being self-perceived as unhealthy increased older adults’ risk of insomnia in our study, where 68.3% of self-perceived healthy older adults reported having no insomnia compared to only 31.7% of self-perceived unhealthy older adults. Although there is less proof, research have shown that the quality of sleep has a better correlation with medical status than does sleep length [22]. Nevertheless, only 2 research studies tried to evaluate the effects of sleep along with other major lifestyle variables in regards to health status. These investigations encompassed all of the significant modifiable risks found by the WHO, including smoking, alcohol consumption, food, and physical activity [23,24]. Social capital did not show significant association in our study, but loneliness did increase the risk of insomnia independently. This may be explained by the mechanism of insomnia in loneliness not only be related to fewer social engagements, but that other mechanisms may play a significant role, such as psychological perception of loneliness [25]. Multiple studies have also shown that insomnia and loneliness are associated, even after adjusting for anxiety and depressive symptoms, and have a cyclical relationship (e.g., loneliness worsens insomnia and vice versa) [25,26]. Loneliness affected insomnia independently despite adjustment for depression and social capital. Mental health services and an increase in social and family engagements may alleviate the feeling of loneliness, thus decreasing the risk of insomnia [27].
We found that physical disability measured by ADL showed significant associations with insomnia, which is similar to several other studies [16,28]. The association between insomnia and physical disability may also be modified by the tools used in assessing insomnia. In a study by Chien and Chen [29], insomnia assessed by PSQI (which has wider definitions and criteria) showed a significant association with physical disability, but when stricter criteria based on DSM-IV or DSM-5 were used, the association was not found. A study by Spira et al. [30] showed that physical disabilities measured by ADL did not have a significant association with insomnia, but IADL did. The possible mechanism behind those findings was that disturbed sleep can lead to fatigue and daytime sleepiness, and then cause functional impairment, especially on complex tasks.
Insomnia was found to be more prevalent in older adults with depression and two meta-analyses also had similar results, with early insomnia specifically being more associated with depression [31,32]. In many demographics and medical contexts, medical problems like physical sickness, anxiety, and some serious sleep disorders like restless legs syndrome and obstructive sleep apnea are also linked to or comorbid with sleeplessness [33]. An enhanced activation of REM sleep pathways may provide an explanation for the link among poor sleep quality, depression, and anxiousness [34]. Results imply that ADL had a significant mediation influence on the association among poor sleep quality and emotional distress, which is in line with results from earlier research [35]. According to this research, the more severely older persons’ observed sleep quality was poor, the more severely their bodily functions were restricted, and thus, the more severely their psychological discomfort was severe. Poor sleep has a negative impact on both the emotional and physical aspects of life, which is one plausible reason. ADL limits in older people make them less prone to be capable of taking care of and maintain oneself, which can have an adverse psychological effect. Older people’s mental health was severely impacted by long-term adverse psychological experiences including remorse and worthlessness [36]. The interplay between depression, subjective health status, life satisfaction, and insomnia is complex and often bidirectional, with each factor influencing and being influenced by the others. The strong association between both variables may be related to the fact that difficulty to sleep is part of the depression assessment questions of the CESD-10 we used. Numerous theories have been put forth to describe the data of how anxiety affects sleep, such as the control of inflammatory, the impact of environmental and social variables, and the role of biological and hereditary variables [32]. A study by Cheng et al. [37] found that in this overall population, the structural connectivities among the temporal cortex, angular gyrus, precuneus, cingulate cortex, and the lateral orbitofrontal cortex were linked with depressive problems scores. It is intriguing to investigate potential brain processes that might influence the link among depression and insomnia because there is a biological factor to the link. The anterior cingulate cortex has been identified as an area in which genetic mutations that control circadian rhythms are emotional dysregulation in depressive episodes and in which nerve cells raise their action during bedtime and disconnection from activities. It has also been demonstrated sleep deprivation impacts brain structures that are associated in emotion, such as the amygdala [38]. Arthritis was associated significantly with insomnia, which was similar to other study [39]. This association might be mediated by pain and physical limitations that are often found in people with arthritis.
Being underweight, tobacco use, engaging in physical activities, and experiencing falls did not show any significant association to insomnia after adjusting for other covariates in our study, while being overweight/obese did. Different from our findings, Pengpid and Peltzer [16] and Soysal et al. [40] showed that being underweight or malnutrition in older adults (but not obesity) was associated with insomnia. The meta-analysis by Chan et al. [41] also concluded that obesity did not have a significant odds ratio for insomnia. Regarding physical activities, studies by Endeshaw and Yoo [42] and Chen et al. [43] showed that engaging in physical activities improves sleep and decreased the risk of insomnia, which was different from our findings. The difference of our results might be explained by several factors. A study by Tsunoda et al. [44] showed that the increase of exercise intensity did not necessarily improved sleep quality of older adults. There are subjective and objective sleep measures and the relationship between these measures were only modest [45,46]. A study by Melancon et al. [46] showed that improvement in objectively measured sleep outcomes does not increase the subjective measure of the population of older adults. Over time, the natural shifts in sleep patterns that come with aging could lead older individuals to adjust how they perceive the quality of their sleep, often without acknowledging the disturbances in their sleep patterns [47]. Several mechanisms could play a role in these findings, e.g., 1) engaging in outdoor physical activities acts as photic stimuli for circadian timing system, 2) improves emotional function, and 3) increases the need for sleep due to energy stores depletion [42,43]. It may be that leisure (e.g., moderate exercise) or occupational physical activity (e.g., exhaustion and hard physical labour with poor nutrition) should be distinguished to show different effects on sleep. Chen et al. [48] showed that leisure-time acitivty, but not domestic or work-related physical activity, was associated with decreased risk of insomnia. Currently, it remains uncertain why engaging in physical activity as a recreational pursuit, rather than domestic or work-related tasks, demonstrates a predictive influence on insomnia. Engaging in leisure activities might elevate mood, offering a greater sense of enjoyment, satisfaction, and distraction from stress. Previous studies have suggested that increased participation in leisure activities correlates with enhanced wellbeing, improved quality of life, and reduced symptoms of depression [49]. Consequently, this may contribute to better sleep patterns.
The strength of our study is that we provided analysis largest community-based data of older adults in Indonesia. We hope to provide insights and general characteristics of older adults, which represent Indonesian population, not just one region. The limitations of our study were that most of the data were self-reported data, which can create bias in the study and as cross-sectionally collected data, we could not determine the causality of those predictors. This includes assessment of insomnia, which is subjective measurement. The prevalence of insomnia may vary between studies depending on the tools that the researcher used for assessment. Involving older adults’ cognitive and social activities, providing social support for lonely individuals, improving their physical function, prompt treatment of arthritis with occupational therapy to overcome limitations and reduction of symptomatic pain may reduce the risk for insomnia.
Early detection and provision of mental health services to depressed older adults may also decrease the risk of insomnia. Our findings regarding age, sex, marital status, residential status, living region, BMI, social capital, tobacco use, physical disability measured by IADL, physical activities, and falls may not show any significant results, but the relationship of insomnia and those variables need to be explored further (future directions).
Notes
Availability of Data and Material
Data employed in this study are publicly available by registering request at RAND (https://www.rand.org/well-being/social-and-behavioral-policy/data/FLS/IFLS.html).
Author Contributions
Conceptualization: Yvonne Suzy Handajani, Elisabeth Schroeder-Butterfill. Data curation: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Formal analysis: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Investigation: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Methodology: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Project administration: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Resources: Yvonne Suzy Handajani, Antoninus Hengky. Software: Antoninus Hengky. Supervision: Yvonne Suzy Handajani, Elisabeth Schroeder-Butterfill, Elisabeth Schroeder-Butterfill. Validation: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Visualization: Yvonne Suzy Handajani, Yuda Turana, Antoninus Hengky. Writing—original draft: all authors. Writing—review & editing: all authors.
Conflicts of Interest
Economic and Social Research Council provided funding for this research project (ESRC). All authors affirm that they have no associations with or participation in any group or entity that would have a financial or non-financial stake in the topics or materials covered in this publication.
Funding Statement
None
Acknowledgements
We acknowledge that RAND provided us with access to the IFLS-5 data (https://www.rand.org/well-being/social-and-behavioral-policy/data/FLS/IFLS.html) and that the Economic and Social Research Council (ESRC), UK, provided monetary assistance for this study through its grant funding of the research project for Care Networks in Indonesia (Project ES/S013407/1).