FACTORS ASSOCIATED WITH INDEPENDENCE FOR ELDERLY PEOPLE IN THEIR ACTIVITIES OF DAILY LIVING Faktor-Faktor yang Berhubungan dengan Kemandirian Lansia dalam Aktivitas Kehidupan Harian

1 Student of Nursing, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, syunigg@gmail.com 2 Department of Medical Surgical Nursing, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, christantie@ugm.ac.id 3 Department of Health Behavior, Environment and Social Medicine, Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, fitrina.m.k@gmail.com, fatwasari@ugm.ac.id Correspondence Author: Fatwa Sari Tetra Dewi, fatwasari@ugm.ac.id, Department of Health Behavior, Environment and Social Medicine, Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Farmako Street, Sekip Utara, Yogyakarta, 5528, Indonesia.

The elderly population (aged 60+) in Sleman District in Yogyakarta Province, Indonesia, is estimated to account for 14% of women and 13% of men (Dewi et al., 2018). This increasing proportion of the population should be able to live independently and be productive both socially and economically to avoid imposing a burden on the community's welfare. An independent life requires elderly people to be able to adjust to the physical and mental changes of the aging process; otherwise, they will lose their role in the community, causing loss of self-esteem and an increase in isolation and loneliness (Kodri & Rahmayati, 2016).
Healthy aging is not merely being free from disease, but also importantly concerns functional abilities (World Health Organization, 2015). Functional capacity is related to the abilities of elderly people to perform their activities of daily living (ADLs) at home independently, including eating, dressing, walking, bathing, and toileting (Bleijenberg, Zuithoff, Smith, de Wit, & Schuurmans, 2017;Oliveira, Nossa, & Mota-Pinto, 2019). Independence in performing ADLs can be measured using the Activities of Daily Living Scale (ADLS) questionnaire (World Health Organization, 2006). Among the elderly Iranian population in 2012, 13.20% of women and 12.60% of men experienced low independence in their ADLs (Tourani et al., 2018). This low independence is associated with many factors, including age, marital status, family type, health, economic status, educational attainment level, social condition, family support, cognitive abilities, motor skills, and individual perceptions (Burman et al., 2019;Kodri & Rahmayati, 2016;Oliveira, Nossa, & Mota-Pinto, 2019).
Measuring functional capacity in the elderly population is important for developing an environment, policies, and interventions that prevent severity of disability and improve quality of life (Buz & Cortés-Rodríguez, 2016). To answer the demands of an aging population, this study aimed to measure ADLs and to identify the factors affecting them among the elderly population in Sleman District.

METHODS
This research was conducted as a crosssectional study using secondary data from the Multidimensional Elderly Care Project (ME Care project), a study project nested in Sleman's Health Demographic and Surveillance System (HDSS) in 2018. The samples for this study were elderly people (aged 55+) with no missing data. In 2018, the total population aged 55+ in Sleman District was 12,872 people. From these, there were 4,512 respondents in the Sleman HDSS were and 578 in the ME Care project. After excluding those with missing data, 549 respondents were selected for the current study (see Figure 1).
The dependent variable in this study was the independence level of the elderly respondents, and the independent variables were demographic factors (age, gender, education, job, socioeconomic status, marital status, residence, living with other people), economic factors, cognitive factors, psychological factors, nutritional status, stroke disease, and disability status.

Independence of Elderly People
In this research, the independence of elderly people was measured in terms of their independence in carrying out ADLs such as bathing, dressing, eating, walking, and toileting. This was measured using the Activities of Daily Living Scale (ADLS) questionnaire. A respondent was categorized as having independence if their score was ≤5 and as being dependent otherwise (World Health Organization, 2006).

Cognitive Factors
The Mini-Mental State Examination (MMSE) questionnaire was used to evaluate five cognitive factors: orientation, registration, attention and calculation, remembering, and language. The level of cognition was classified as normal if the respondent scored ≥10 and as showing dementia if they scored <10 (Ministry of Health RI, 2015).

Psychological Factors
To measure psychological factors for people without dementia (i.e., who showed a normal condition for cognitive function), we used the short-version Geriatric Depression Scale (GDS) questionnaire instrument, which consists of 15 questions. We categorized a respondent as showing no depression if they scored 0-4 and as having depression if they scored ≥5 ( (Hancock & Larner, 2015).

Economic Factors
The Financial Management Behavior Scale (FMBS) questionnaire, consisting of 15 items, was used to measure financial management factors. The questions cover cash management, savings investment, insurance, and credit management. Respondents were classified as having good financial management if they scored >23.19 and as having poor financial management if they scored ≤23.19, according to the mean cut-off point (Dew & Xiao, 2011).

Nutritional Status
Nutritional status was measured using the Mini Nutritional Assessment (MNA) questionnaire, which consists of two parts: screening and assessment. Respondents were classified as having good nutritional status if they scored ≥24, as having a risk of malnutrition if they scored 17-23.5, and as having malnutrition if they scored <17 (Ministry of Health RI, 2015).
In this research, we carried out a descriptive analysis of the data, and then we applied bivariate analysis using the chi-square test, Fisher test, and Poisson test. Finally, we conducted a multivariate analysis using a logistic regression test. The data were analyzed using STATA version-13 software with a significance level of p < 0.05. This study has been approved by the Medical and Health Research Ethics Committee (MHREC), Faculty of Medicine, Public Health and Nursing Universitas Gadjah Mada -Dr. Sardjito General Hospital (No. KE/FK/1074/EC/2019).

RESULTS
The prevalence rate of dependency in carrying out ADLs among this sample population was 14.03%. In terms of the demographic criteria, more than half of the respondents were female (56.10%), and the majority of the respondents were aged 56-65 (52.82%). Most of the respondents had low education levels (elementaryschool level) (40.26%), no occupation (73.77%), were married (96.54%), lived in urban areas (79.78%), and lived together with their family (92.35%). For the economic factors, more than half of the respondents had poor financial management (54.28%). The majority of the respondents were healthy elderly people; however, some of them suffered from dementia (4.92%), depression (14.03%), malnutrition (1.82%), stroke disease (4.55%), and disability (0.91%) (see Table  1).
Bivariate analysis showed that ADLs were significantly associated with age, occupation, financial management, psychological conditions, and the presence of morbidities. The level of dependency increased as age increased. In comparison with the youngest age group (56-65), the 66-75 age group showed a 3.11 times higher risk of being dependent (95% CI = 1.80-1.99), the risk for the 76-87 age group was 4.18 times higher (95% CI = 2.35-7.41), and that for the 85-95 age group was 5.69 times higher (95% CI = 2.47-13.11). The respondents who had no occupation were at significantly higher risk (3x) (95% CI = 2.03-4.56) of being dependent than those who had an occupation. Additionally, those with poor financial management capabilities were at significantly higher risk (2x) of being dependent (95% CI = 1.32-3.35) than those with good financial management capabilities (see Table 2). Some diseases-dementia, depression, malnutrition, and stroke-showed a significant relationship with dependency in performing ADLs. Respondents who had dementia were at significantly higher risk (5.48x) of being dependent (95% CI = 3.77-7.97) than those who did not have dementia, and those with depression were at 2.45 times more risk of being dependent (95% CI = 1.59-3.78) than those who did not have depression. Those who had malnutrition issues were at higher risk (at a highly significant level: 12.59x) (95% CI = 9.03-17.57) than those who did not have malnutrition, and those who had experienced a stroke were at significantly higher risk (4.66x) of being dependent (95% CI = 3.07-7.07) than those who had not experienced a stroke (see Table 2).
Age, nutritional status, and stroke were also shown to be significantly related to dependence after being controlled by other variables. In comparison with the youngest age group (56-65), the 66-75 age group's risk of being dependent was 2.65 times higher (95% CI = 1.56-4.51), that of the 76-87 age group was 2.41 times higher (95% CI = 1.36-4.27), and that of the 85-95 age group was 4.31 times higher (95% CI = 1.91-9.72). Respondents with malnutrition were at significantly higher risk (6.62x) of being dependent (95% CI = 3.79-11.57) than those who did not have malnutrition, and those who had experienced a stroke were at significantly higher risk (3.06x) (95% CI = 2.03-4.61) than those who had not. The results of this multivariate analysis show that age (PR = 4.31; 95% CI = 1.91-9.72), malnutrition (PR = 6.62; 95% CI = 3.79-11.57), and stroke disease (PR = 3.06; 95% CI = 2.03-4.61) accounted for 15% of cases of elderly dependence, whereas the remaining 85% of cases were influenced by other variables (see Table 3).

DISCUSSION
The results of this study show that the majority of the elderly people in the sample had independence in carrying out ADLs such as eating, dressing, walking, bathing, and toileting. This finding is similar to a study conducted in Central Lampung District, Lampung Province, Indonesia, which showed a 32% level of dependence in ADLs among elderly people (Kodri & Rahmayati, 2016). A number of elderly people in current study with low dependence were living with help from a family member, caregiver, or other people. In the current study, we found that the prevalence of elderly dependence was 14.03%, which is similar to research in the Netherlands conducted by Ćwirlej-Sozańska, Wiśniowska-Szurlej, Wilmowska-Pietruszyńska, & Sozański (2019) that showed a 17.10% prevalence of elderly dependence. Our results also indicate that age, malnutrition, and stroke were related to elderly dependence; this finding is similar to the results of a study in East Delhi, India (Vaish, Patra, & Chhabra, 2020).
Age is an important factor in elderly dependence (Burman et al., 2019). A study based on Chinese Longitudinal Healthy Longevity Survey data reported that age is the most important factor in elderly people's potential dependence in the next two or three years (Zhang, Jia, Li, Liu, & Li, 2016).   The current study results support the evidence that people are more likely to be dependent as their age increases. This is in line with a study conducted in China that showed an association between age and elderly people's dependence in their ADLs, with getting older leading to more dependence (Guo, Chen, & Perez, 2019). A study in Coimbra, Portugal, also reported that people aged 75-79 experienced a decline in their physical, cognitive, and social functions that limited their ability to carry out their ADLs (Oliveira, Nossa, & Mota-Pinto, 2019). People over the age of 60 generally have at least one limitation in carrying out ADLs, and with each further year's increase of age, the limitations will be increased by 8% (Ćwirlej-Sozańska, Wiśniowska-Szurlej, Wilmowska-Pietruszyńska, & Sozański, 2019). Malnutrition has negative effects on the health of elderly people, one of which is increased functional disorders (Shakersain et al., 2016).
Physical function decrease is more significant among elderly people with malnutrition than among those who have good nutrition (Hsu et al., 2019). In a study conducted by Wei et al (2018) regarding nutritional status and physical frailty among elderly people, the authors found that malnutrition caused elderly frailty and vice versa. Malnutrition is related to frailty and sarcopenia, a condition of losing muscle mass and a decline in muscle strength as elderly people get older. There is an increased prevalence of sarcopenia and low muscle mass in elderly people who are unable to walk compared with those who can walk independently (Cruz-Jentoft, Kiesswetter, Drey, & Sieber, 2017;Maeda, Shamoto, Wakabayashi, & Akagi, 2017;Sánchez-Rodríguez et al., 2017). Being malnourished and underweight are associated with mobility limitations, and limited mobility is also related to sarcopenia (Maeda, Shamoto, Wakabayashi, & Akagi, 2017). Nutrition is also very important to brain developmentnutrition delays can decrease cognitive function and increase dementia in elderly people (Shakersain et al., 2016).
Chronic disease decreases functional capacity and can increase functional disorder (Oliveira, Nossa, & Mota-Pinto, 2019). The dependence of elderly people increases due to the interaction between environment barriers and multimorbidity. Chronic diseases such as stroke increase the risk of dependency for elderly people by 4%-7%, and elderly people with chronic diseases have at least one problem in their ADLs (Ćwirlej-Sozańska, Wiśniowska-Szurlej, Wilmowska-Pietruszyńska, & Sozański, 2019). A systematic literature review found two studies emphasizing that stroke is a risk factor for limitations in ADLs among elderly people aged 75 and over (van der Vorst et al., 2016). The number of chronic diseases suffered by an elderly person also increases their inability or dependency (Guido, Perna, Peroni, Guerriero, & Rondanelli, 2015;Rizzuto, Melis, Angleman, Qiu, & Marengoni, 2017). Chronic diseases such as cardiovascular disease, cancer, stroke, and lung disease are common causes of death in the geriatric population.
This study's results imply that providing health services to achieve early detection of risk factors for non-communicable diseases (NCDs), monitoring of such risk factors, and prevention of NCDs are very important among high-risk groups. Hypertension and diabetes mellitus are among the NCDs that can lead to cardiovascular and stroke disease. The elderly health service in Indonesia is supported by the Government of the Republic of Indonesia through Government Regulation Number 43, introduced in 2004, which encourages the provision of Integrated Service Post for Elderly and community-based activities to increase the well-being of elderly people (President of RI, 2004).
A service monitoring the nutritional status of elderly people is also important. Family members acting as main care providers in Indonesia should also be involved and educated to identify early signs of functional limitation and malnutrition in elderly people so they can plan further actions for maintaining functional ability and prevent severity of disability (Ris, Schnepp, & Imhof, 2019).

CONCLUSION
Age, nutritional status, and stroke are significant factors associated with elderly dependence. The aging process decreases elderly people's ability to carrying out ADLs. Elderly people with malnutrition and who have had strokes have increased dependency in carrying out ADLs.
Offering health education for elderly people during routine visits to Integrated Service Post for Elderly is an opportunity to achieve health, independence, and engagement among the elderly population. To help prevent malnutrition in elderly people, it is also important to involve their family members. Future research into providing suitable care to help elderly people to retain their ADLs in their local conditions is needed to transfer these results to the real world.

CONFLICT OF INTEREST
The authors declare that they have no conflict of interest.

AUTHOR CONTRIBUTION
SW carried out the data analysis, drafted the article, and approved the publication. CE supported the data interpretation, revising important content, and approving the publication. FMK led the data collection, data interpretation, and revising of important content, and approved the publication. FSTD designed the study, led the data interpretation, acted as corresponding author, led the revisions process, and approved the publication.

ACKNOWLEDGMENTS
This paper used data from the Multidimensional Elderly Care Project (ME Care Project) funded by the Ministry of Research and Higher Education of Indonesia under the doctoral dissertation scheme, a study project nested in Sleman HDSS in 2018. The Sleman HDSS data collection has been primarily funded by the Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, Yogyakarta, Indonesia.