Urban-Rural Dimension of Fall and its Associated Risk Factors Amongst Community Dwelling Older Adult in Indonesia

Background: Falls are common problem among older people, WHO (2014) estimated 424,000 fatal cases of falls occur every year, making fall as the one of the causes of death in the older adult due to accidental injury. This study aimed to assess the risk of falling among community dwelling older adult in rural and urban area. Method: Using a proportional sample size we determine 2 areas that describe rural and urban area in West Java Prefecture. The incidence of fall in the last 12 month is selected as outcome variable. Sociodemographic background, chronic medical condition, fear of falling, visual and hearing impairment, Activity Daily Living (ADL) Barthel index, the physical performance (Short Performance Physical Battery (SPPB)), and living environment were assessed to identify the factor contribute to the incidence of falling. Result: a total of 611 older people participated this study consist of 381 (62%)older persons living in rural area and 230 (38%) person are living in urban area, 70 percent of them at the range of 60-69 years old, and 73% of them are female. The prevalence of fall in rural area (16.5%), is signicantly higher than urban area (10.7%). The multiple logistic regression analysis showed, male gender (OR= 0.29;95%CI (0.095 - .882)), chronic illness (OR=3.25 95%CI (1.236 - 8.533) and visual impairment (OR=3.6; 95%CI (1.517 - 8.544)) are associated with risk of fall in among older adult living in urban area. On the other hand, Visual impairment (OR=1.81; 95%CI (1.027– 3.179)) and living environmental (OR=3.36; 95% CI (1.137-9.921)) are signicantly associated with fall in among older adult living in rural area. Conclusion: We identify the different risk factors for falls in urban and rural area. The results can be concluded that a different approach is needed to reduce the risk of falling for the older adult in urban and rural areas based on risk factors.


Introduction
Ageing process signi cantly in uence the physical and functional capacity of the older adult that implicate weaknesses, limitations and disabilities. This condition can affect the balance that puts the older adult at high risk of falling. Fall-related injury in older adults has been recognized as a major public health issue. Older adult has decreased functional body motor as well as cognitive that increase the risk of falling. [1][2][3][4] Falls are common problem among older people, it is estimated 424,000 fatal cases of falls occur every year, 5 this situation enacting fall as the one of the causes of mortality in the older adult due to accidental injury. More than 80% of fall related deaths in low-and middle-income countries, accounting for more than two-thirds of the highest deaths among adults over the age of 60. 5 The Indonesian national health survey (2018) reported that the proportion of injuries caused by falls among the age group of 55 years or more is 25%, and 67% of falls occur in the home environment. 6 In 2019, the percentage of older people population reached 9.60 percent or around 25.64 million people. 7 This condition shows that Indonesia is transitioning towards aging population because the percentage of the population aged over 60 years reaches above 7 percent of the total population and will become a country with an aging population structure if it is accounted more than 10 percent. Considering such a large population of older people, it is very important to explore factors that may cause accidental falls that deteriorate older people's health. Various effects of falls that can occur in the older adult, including post-fall anxiety syndrome, injury to either soft tissue or broken bones, hospital treatment, disability (decreased mobility), decreased functional status / decreased independency, increased use of health care facilities, and can even result in mortality. [8][9][10][11] The adverse effect of fall might signi cantly in uence older adult mobility, hence fall experience will be a risk of decreased quality of life up to 70%. 12 Fall risk for community-dwelling older people differs somewhat from assessment in a hospital or longterm care facilities setting. Older people living in community are usually more active, and more physically independent than those living in long term care facilities. Nevertheless, fall assessment needs to be multifactorial, because of the complex nature of falls. Multiple factors have been shown to increase fall risk in community-dwelling adults over the age of 65, these factors can be divided into 3 categories including: 1) Demographic factors (Age, gender), Intrinsic factors include health condition that can be de ned as the existence of chronic illness 13,14 , history of falling. [15][16][17] functional capacity indicate by the level of dependency in performing activity daily living and Physical capacity as indicate by gait speed and balance examination, 18,1920-22 sensory impairment such as visual impairment and hearing impairment, 23-25 and depression. 3,8,9,26 Furthermore, another risk factor for falling that come from outside the older adult also been identi ed as the extrinsic risk factor for falling such as living environment, home hazard and neighborhood accessibility 21,27 and medication consumed by the older adult. 28,29 Given the large longstanding disparities in rural and urban areas, investigating the rural-urban dimension of fall among Indonesian older population is a major concern of this study. The difference between rural and urban living environment will have implications for the types of fall risk faced by the elderly. The division of rural and urban areas is differentiated based on certain criteria that meet administrative requirements in terms of population density, percentage of agricultural households and the presence or access of urban facilities such as roads, formal education facilities, public health facilities, etc. 30 Indonesian older population is spread both in urban and rural areas, where the elderly who live in urban areas are higher than in rural areas (52.80 percent compared to 47.20 percent). 7 The living environments for rural and urban older people are also dissimilar. For example, most rural residents live in wider onestory houses whereas urban residents live in densely populated housing environments and tend to have houses that are more than one story high. Access to public services is easier for urban communities to reach good transportation facilities, while rural communities tend to require heavier efforts such as walking through areas with hilly topography. Signi cantly diverse environmental factors could be a reason for dissimilar fall incidence rates among elders residing in rural and urban area. The cultural tradition, leisure and physical activities among rural and urban older population are also dissimilar.
Taking into account the rural and urban differences presented in the previous literature, this study hypothesizes that the rural-urban disparities discussed above may result in varying patterns and determinants of fall between rural and urban older people.
Although numerous studies have identi ed factors of falling, few studies have focused on examining the rural and urban dimension for the risk of fall among Indonesian older people. Additionally, no study has contrasted incidence and risk factor of falls among urban and rural older people in Indonesia. To address the above concerns, this study analyzes community-dwelling individuals aged 60 and over to reveal the incidence of falls among older people in urban and rural and to reveal the contrast environmental and personal risk factors for falls among rural and urban Indonesian older people. The study contributes to the literature by improving our understanding on risk factors of falls among Indonesian rural and urban elders. A clearer understanding of how personal characteristics, health condition, behavioural and lifestyle factors in uence fall rates in Indonesian populations is essential for elucidating fall prevention strategies. Findings from Indonesian study also provide useful implications for other countries with similar characteristics.

Method Study design
This is a descriptive analytic study that using cross sectional approach. Household survey was conducted during June -September 2020. West Java Provinces selected as one of Indonesian province that entering aging population (have more than 10% older people population). 31 Population and sample Population of this study are community dwelling older adult aged 60 years old or more. The sample size calculation is based the prevalence of fall in community dwelling older adult in Indonesia by using fall prevalence 29 % in the previous year 32 and precision was considered as 95% (±0.05) counted for urban area and each area. A cluster sampling method was used to achieve the required study sample from 2 selected area out of 28 municipalities in West Java administrative area namely Bandung Regency that representing rural area and Bandung City representing urban area. From those selected municipalities, then we randomly selected 4 sub-district from each rural and urban area. From each subdistrict we randomly selected 2 villages for each subdistrict in rural area and 2 urban villages in urban area, a total of 16 urban village and village are selected. The sample size for each village was calculated proportionally and the sampling was carried out randomly in each village.

Variable and Measures
Outcome variable Fall is become outcome on this study. Fall is assessed with following questions; "Have you fallen down in the last twelve months?" and "How many time have you fallen down?" Risk Factor for Fall Sociodemographic factor included age, sex, education, occupation status and residential area. Chronic medical conditions were assessed with the question "Has a doctor/paramedic/nurse/midwife ever told you that you had following diseases?" ("hypertension, diabetes or high blood sugar, heart diseases, stroke, arthritis/rheumatism, and depression") (Yes, No). The existence of one or more chronic medical condition considering as having chronic illness.
Visual impairment refers to any degree of impairment to a person's ability to see that affects his or her daily life 33 . Visual impairment was assessed subjectively with the questions "Do you have any problem with your vision? Such as blurred vision? (Yes/No) any answer will be con rmed with tumbling E chart test to con rm subjective visual impairment. Any correction with glasses will be considered as normal visual.
Hearing impairment refer to a person who is not able to hear as well as someone with normal hearinghearing thresholds of 20 dB or better in both ears 34 . This study assessed the subjective hearing impairment using the question: "Do you have any problem with your hearing function?" (Yes/No. Any subjective answer, to be con rmed by observation and whisper test. Any correction with hearing aids considered as normal hearing. Functional capacity was measured by Barthel index of ADL (Activity Daily Living). This Barthel index measures the likelihood of being able to live at home with a degree of independence 35 . Ten basic activities of daily living (ADL) are captured: bowels, bladder, grooming, toilet use, feeding, transfers, walking, dressing, climbing stairs, and bathing. A dichotomous category selected as mild/severe dependent (score ≤19) and independent (score 20) 35 .
The physical performance measured by SPPB (Short Physical Performance Battery). SPPB is a wellknown and sensitive performance test which combines the results of 3 balance tests, a gait speed test and a chair stand test into one score 36,37 . Hand-clocked durations are used to indicate the performance capacity for individual components which are scaled to a grade between 0 and 4 and add up to a score between 0 and 12). This performance is dichotomized in to poor and good physical performance.
Living environmental risk factor was assessed with the CDC Home Fall Prevention Checklist for Older Adults to identify home hazard that may increase the risk of falling, including room lighting, oor condition, stairs and steps, kitchen, bedroom and safety in the bathroom. Any "Yes" in one of the hazards categorized as unsafe environment. Medication

Data analysis
Data analysed using statistical software IBM SPSS statistics 22. A descriptive analytic was perform to describe the sociodemographic pro le of study participants. Independent sample t-test and chi-square analysis selected to compare risk of falling among the older person living in rural and urban area. Binary multiple logistic regression analysis performed to identify the association between fall and the risk factor.
Prior to participation, all older people participated in this study voluntarily signed their informed consent to prove their voluntary contribution on this study. This study has passed the ethical clearance from institutional research bureau of Universitas Indonesia with approval number: Ket-326/UN2.F10.D11/PPM.00.02/2020.

Results
This study conducted in two representative rural and urban area in West Java, Indonesia. A total of 611 sample are satisfy the statistical analysis, older adult participated in this study consisting 230 older adult from urban area and 381 older adult from rural area with 77% response rate. Table 1 showed that 19% of the older adult are having history fall in the last 12 month, and there is no signi cant difference between those living in rural and urban area. Older adult living in rural area are signi cantly older than those living in urban area. Older adult living in rural area are having higher proportion of hearing impairment and visual impairment. Rheumathoid/osteoarthritis reported by 59% of study participants, most of study participants are having unsafe living environment. Older adult living in rural area having signi cantly higher number of unsafe living environment compare to urban area. Older adult living in rural area are signi cantly have poorer functional capacity and physical performance.
There is no signi cant different in the frequency of fall between urban and rural setting. Table 3 describe the result from logistic regression analysis for the variables associated with fall in older adult living in rural and urban area. Gender is signi cantly associated with fall in among older adult living in urban area. Being male likely to become protective factor for fall OR= .289; 95%CI (0.095 -0.882).
Visual impairment is signi cantly associated with the incidence of fall both in rural and urban area whereas older people living in urban area who have visual impairment likely to have 3 times higher risk to fall OR=3.601; 95% CI(1.517 -8.544). On the other hand, older people living in rural area who have visual impairment likely to fall 1,8 times higher compare to normal visual (OR=1.807;95%CI (1.027 -3.179).) Chronic medical condition is only associated with the incidence of fall in elderly living in urban area, older people who have chronic medical condition likely to fall 3 times higher compare to healthy older people OR = 3.247; 95%CI (1.236 -8.533). Living environment is associated with the incidence of fall in older people living in rural area, those who have unsafe environment likely to fall three times higher than those with safe environment OR=3.358 (1.137 -9.921).

Discussion
This study sought to identify the difference risk factor of fall among community dwelling older adult in urban and rural area using multifactorial approach. We identi ed signi cant different in the risk factor model in urban and rural setting. The ndings suggest that more falls occurred among older people living in rural than urban area; and the fall incidence rate is higher among females than males. Being male, having visual impairment and the existence of chronic illness are signi cantly associated with the incidence of fall in among older people living in urban area. Meanwhile, for the older people living in rural areas, we identi ed visual impairment and safety of the living environment as risk factors for falls.
Gender has been identi ed as on of risk factor for fall among older people living in urban area. Female older people likely to have higher risk of fall compare to male. This nding is inline with study conducted by Zhang et.al (2019) in among Chinese, that male older people living in urban area are likely to have lesser fall than female. 38 This nding can be assumed that male older people in urban areas are mostly former workers who work outside the home, which allows them to have better muscular endurance than women who are mostly housewives. The remaining muscle strength allows it to be a protective factor for falls. 39 Chronic illness is experienced by majority of the older adult both living in urban and rural area. Chronic illness can be de ned as degenerative diseases that experienced by the older adult as part of their ageing consequence. Although the prevalence of chronic illness in older adult living in rural area are signi cantly higher than those living in urban area, the direct association between chronic illness and fall was only found in urban model. As being widely known that chronic illness has a signi cant effect on the physical and functional capacity of the older adult. Chronic illness may have signi cant impact on the balance and posture stability of the older adult which will increase the risk of falling. 40 Several studies conducted by former researcher identi ed the association between chronic illness and the incidence of fall in older adults .16, 41,42 Compare to the older adult living in rural area, older adult living in urban area with chronic illness are likely to fall 3 times higher.
This study identify visual impairment is signi cantly associated with the incidence of falling in older adult living in rural and urban area. Visual impairment refers to the condition that a person have subjective complain on their vision acuity that affecting daily activities. This nding is in line with former study conducted by Hong at. Al (2014) on their longitudinal study among older people with visual impairment, who identi ed that older people who have visual impairment likely to have two or more fall. 43 Older adult with visual impairment and high mobility are at higher risk of falling. Previous studies have identi ed the effect of visual impairment on falls. Another study conducted on the Shinpey Eye Study community in China showed that, the incidence of falling was signi cantly associated with the visual acuity of Snellen which was best corrected for less than 6/12. Furthermore, a study in Shenzen China showed that impaired visual increased the risk of falling with RR = 1.46, 95% CI 1.03 to 2.08. 4 This study also nds that unsafe living environment likely to increase 3 times higher risk of falling in among older people in rural area. Unsafe living environment refer to the condition of living environment that considered as unsafe for older people to perform their daily activities. Approximately 96% of older adult living in rural area living in unsafe home environment and may increase the risk of falling. There are fundamental differences in the environmental conditions of the elderly in rural and urban areas. Older adult living in urban areas are more likely to occupy modern housing with bathrooms and toilets in the house. Meanwhile, in rural areas, traditionally use shared communal latrines that placed outside their home. Therefore, older people in rural area have to leave the house to access toilets and clean water. The topography of rural areas, which is generally hilly and rocky, increases the risk of imbalance which has implications for falls. This study is in agreement with former study conducted by Zhang et.al (2014) identi ed the signi cant association between living environment and the incidence of fall in among older people in rural area. 38

Conclusion
The main part of this research has concerned whether factors associated with senior falls are different between rural and urban residents. This study identi es different risk factors for falling among older people based on the characteristics of the area of residence. These ndings have implications for the importance of different approaches to fall prevention efforts of older people in urban and rural areas. Implications drawn from this research provide suggestions for the government and local agencies to develop suitable fall prevention strategies using rural and urban approach especially concerning in gender, visual impairment, having existing chronic illness, and living environment.

Study limitation
This limitation confounds the ability to consider covariates effecting outcomes. First, the data are based on a cross-sectional survey. As other cross-sectional studies, there is an inherent weakness of not being able to separate the cause-effect relationships of variables. In addition, recall bias may exist, especially among those elders who have poor memory. Since the data were obtained on the basis of 12 month fall recall and only the most recent fall was queried in detail, it is possible that the frequency of falls was underreported. The measures of environmental factors are also scarce it is based on observation during data collection, the older people behaviour toward living environment were not being assessed. Further study needs to exploring fall risk factors in a more comprehensive manner. Even with the limitations, this study is one of the rst ones study that analyse rural-urban dimension using representative sample size that risk factors of falls among Chinese Indonesian community-dwelling elders in both rural and urban areas. Findings of this research offer an important base for conducting future prospective studies and launching possible fall prevention program for older people.

List Of Abbreviations
OR: odds ratios CI: con dence intervals Declarations Ethics approval and consent to participate The study protocol was approved by the ethics committees of the institutional research bureau of Universitas Indonesia with approval number: Ket-326/UN2.F10.D11/PPM.00.02/2020 and has been conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects by the Indonesian Government.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and analysed during the current study are not publicly available because the Ethical Guidelines prohibit researchers from providing their research data to other third-party individuals.