Factors associated with quality of life among elderly in urban Vietnam

JOURNAL OF MEDICAL RESEARCH<br /> <br /> FACTORS ASSOCIATED WITH QUALITY OF LIFE<br /> AMONG ELDERLY IN URBAN VIETNAM<br /> Dao Thi Minh An¹, Vu Toan Thinh¹, Dunne P Michael²<br /> ¹Institute for Preventive Medicine and Public Health, Hanoi Medical University<br /> ²School of Public Health, Queensland University of Technology, Australia.<br /> Quality of life (QoL) among the elderly is a big problem in Vietnam due to a growing proportion of the<br /> elderly in Vietnam while many conditions, including policies, social facilities, culture and other factors are not ready to support for QoL among elderly. This cross-sectional study was conducted to<br /> explore QoL and factors associated with QoL among the elderly in Trung Tu ward, Ha Noi, Viet Nam.<br /> The findings showed that the four domains of QoL the among elderly fluctuated around 50. Mean<br /> scores of social and psychological QoL were higher than those in the physical and environmental<br /> domains. A statistically significant difference in mean scores of QoL by socio-demographics was recorded (age profile, educational attainment, and occupation). All four domains of QoL were positively<br /> correlated with each other. Furthermore, age, psychological, social and environmental domains collectively contributed to 47.59% of the physical domain; while the physical, social, and environmental<br /> domains accounted for 56.13% of the psychological domain. We also found that occupation (worker), as well as physical, psychological, and environmental metrics, accounted for 34.19% of the social domain. Moreover, physical, psychological, social domains and occupation (home-wife) collectively accounted for 45.92% of the transformation of environmental domain. Our study suggests that<br /> it is essential to evaluate overall QoL to have a comprehensive view of its effects in the long run.<br /> <br /> Keywords: Quality of Life, Elderly, Hanoi, WHO QoL-Bref<br /> <br /> I. INTRODUCTION<br /> Vietnam’s population structure is in a period of dramatic change, presenting a number of public health benefits as well as challenges. Today, one of the most prominent<br /> issues is how to address a rapidly growing<br /> Corresponding author: Vu Toan Thinh, Institute for<br /> Preventive Medicine and Public Health, Hanoi Medical<br /> University<br /> Email: vutoanthinhdhy@gmail.com<br /> Received: 05 June 2017<br /> Accepted: 16 November 2017<br /> <br /> 114<br /> <br /> elderly population. Statistics from the Living<br /> Standard Survey of Households in Vietnam<br /> showed that the number of elderly people<br /> (defined as men and women aged 60 years<br /> plus) grew from 3.71 million people in 1979<br /> (6.9% of the total population) to 7.72 million<br /> in 2009 (9% of the total population). At this<br /> rate, by 2020, it is estimated Vietnam’s elderly population will be greater than 12 million [1].<br /> With this in mind, quality of life (QoL)<br /> among the elderly is the most pressing isJMR 111 E2 (2) - 2018<br /> <br /> JOURNAL OF MEDICAL RESEARCH<br /> sue. QoL is a multi-dimensional, highly<br /> subjective concept and, as recommended<br /> by the World Health Organization (WHO),<br /> is measured using four major domains, including physical, psychological, social, and<br /> environmental [2]. Within these categories,<br /> QoL has its own characteristics according to<br /> different economic and socio-cultural levels,<br /> producing trend where an overall negatively<br /> asociates with age QoL [3].<br /> Within Vietnam’s cultural context of<br /> multiple generations living together in the<br /> same household, as well as the impact of<br /> urbanization on a rapidly aging population,<br /> QoL and mental disorders among the elderly need to be paid more attention. A recent study conducted in 8 provinces on the<br /> health status of Vietnam’s elderly population<br /> showed that about 95% of the participants<br /> were infected with at least one disease. On<br /> average an elderly person suffers from 2.6<br /> diseases. With this in mind, about 23% of<br /> the elderly people have difficulties in their<br /> daily life, of which more than 90% need supports from other people [4]. According to the<br /> statistic of the National Institute of Gerontology, 9.2% of the Vietnamese population<br /> suffer from depression, one third of which<br /> were elderly and largely retired populations<br /> in major cities [5; 6].<br /> This is an important point to understand<br /> in an age of rapid urbanization. The proportion of elderly in urban areas is quickly rising<br /> and becoming a far more difficult problem<br /> to properly address. Compared to the elderly living in rural areas, the elderly in urban<br /> zones have distinct lifestyles such as extensive free time, more available information<br /> <br /> JMR 111 E2 (2) - 2018<br /> <br /> relating to health problems, but most of all,<br /> less integrated neighborhood relationships<br /> compared to those in rural areas. Further,<br /> after retirement, may confront psychological<br /> loneliness, emptiness, and even abandonment by their children and neighbors, this<br /> would put the elderly in isolated situations.<br /> Hanoi is the capital of Viet Nam where<br /> there is a rapidly developing economy and<br /> growing population in which many Vietnamese households have 2 to 3 generations live<br /> together [7]. QoL of the elderly in Hanoi after retirement is often influenced by many<br /> factors such as home economics, relationship with their spouse and children, social<br /> issue, physical and mental health, and the<br /> medical system [8 - 10]. However, few studies have specifically analyzed the extent<br /> that these factors impact QoL among the<br /> elderly, especially among those living in urban wards in Hanoi. In Vietnam, there were<br /> some studies conducted on QoL among the<br /> elderly [11]; however, none focused on the<br /> population living in major cities.<br /> Therefore, this study aims to analyze the<br /> quality of life based on the four main domains among the elderly population living in<br /> Hanoi’s Trung Tu ward.<br /> <br /> II. SUBJECTS AND METHODS<br /> 1. Subjects<br /> Target population is the elderly living in<br /> urban areas in Hanoi city. Particularly, the<br /> study population is defined as the elderly<br /> living in Trung Tu ward, Hanoi. Participants<br /> who were recruited into this study if they met<br /> the following criteria 1) People who living in<br /> Trung Tu ward, Hanoi for at least 1 year; 2)<br /> 115<br /> <br /> JOURNAL OF MEDICAL RESEARCH<br /> Aged ≥ 60 years old (according to the ordinance of the elderly, issued by the President of the National Assembly on 28th April<br /> 2000, the elderly are defined as citizens of<br /> the Socialist Republic of Vietnam from 60<br /> years old or more [12]); and 3) Willing to<br /> participate in this study after giving informed<br /> consent. Individuals were excluded if they<br /> were living in Hanoi temporarily, refused to<br /> participate, or had difficulties in understanding or completing the questionnaire.<br /> 2. Methods<br /> Research site<br /> This cross-sectional study was conducted in Trung Tu ward, Hanoi, which is located in Northern Viet Nam. This ward has one<br /> of the densest populations in Hanoi and is<br /> mainly comprised of government officers<br /> that live in 62 dormitories and 2 residential<br /> districts with convenient transportation and<br /> close proximity to entertainment venues,<br /> national hospitals, and schools. Until 2012,<br /> there were 1,593 elderly people in Trung Tu,<br /> accounting for 11.78% of the total population of the ward.<br /> Sample size and data collection<br /> This is a pilot study, so we decided on<br /> a convenience sample of 2% (or 299) of<br /> Trung Tu ward’s total elderly population,<br /> who volunteered for the study. The first step<br /> of recruiting participants was effectively<br /> announcing the study. Ten health collaborators of Trung Tu’s health center wrote an<br /> introduction about the study and announced<br /> the recruitment on the boards at dwelling<br /> areas that they are in charge of. The announcement ordered those who wanted to<br /> voluntarily participate in the study to call a<br /> 116<br /> <br /> toll-free number for registration. After being<br /> contacted by potential subjects, the second step was to screen them for eligibility<br /> using a questionnaire that assessed each<br /> participant’s recruiting criteria. They were<br /> then recruited into the study based on these<br /> criteria until the target sample size of 299<br /> elderly people was met. In the last step of<br /> sampling, collaborators contacted registered participants at home and provided<br /> them with consent forms. After reading the<br /> consent form, if the elderly agree to participate in the study, they would then receive a<br /> self-administered questionnaire from collaborators. They then allowed at least 2 weeks<br /> for participants to complete their questionnaires and return them to health collaborators in Trung Tu ward, either by themselves<br /> or their relatives. If their relatives delivered<br /> the questionnaire, it would be sealed in<br /> an envelope to ensure confidentiality. The<br /> self-administered questionnaires were immediately screened to check for missing<br /> information to ensure participants could circle responses they missed. If their relatives<br /> delivered their questionnaires, we used the<br /> telephone number which was recorded on<br /> that questionnaire to call the elderly. After<br /> that, the participants' phone number was<br /> deleted to secure their personal information. If the elderly refused to answer, that<br /> questionnaire was considered as ineligible.<br /> Measures<br /> Demographics: Includes 7 questions<br /> about participants’ age, marital status (married vs. unmarried), education level, living<br /> arrangements, and occupation before retirement.<br /> <br /> JMR 111 E2 (2) - 2018<br /> <br /> JOURNAL OF MEDICAL RESEARCH<br /> Quality of Life: WHO QoL-Bref questionnaire is self-assessment that antains<br /> 24 items, each presenting one facet of QoL<br /> and two “benchmark” items in an individual’s overall QoL and general health. The<br /> facets are defined as those aspects of life<br /> that are considered to contribute to a person’s QoL. QoL comprises of four main domains – physical health (7 items relating to<br /> pain and discomfort, dependence on medical treatment, energy and fatigue, mobility,<br /> sleep and rest, activities of daily living, and<br /> working capacity), psychological health (6<br /> items relating to positive feelings, spirituality, religion and personal beliefs, thinking,<br /> learning, memory and concentration, body<br /> image, self-esteem, negative feelings), social relationship (3 items relating to personal<br /> relations, sex life, practical social support),<br /> and environment (8 items relating to physical safety and security, physical environment, financial resources, information and<br /> skills, recreation and leisure, home environment, access to health and social care, and<br /> transportation). These facets were scored<br /> on a Likert scale from 1 to 5 with 1 = Very<br /> poor, 2 = Poor, 3 = Neither poor or good, 4 =<br /> Good, and 5 = Very good; 1 = Very satisfied,<br /> 2 = Dissatisfied, 3 = Neither dissatisfied or<br /> satisfied, 4 = Satisfied, and 5 = Very satisfied; 1 = Not at all, 2 = A little, 3 = A moderate amount, 4 = Very much, and 5 = Extremely; or 1 = Never, 2 = Seldom, 3 = Quite<br /> often, 4 = Very often, and 5 = Always. The<br /> raw score from each domain of QoL include<br /> varying scales; for instance, the physical<br /> domain ranges from 7 to 35 points; psychological domain ranges from 6 to 30 points;<br /> social domain scores ranges from 3 to 15<br /> JMR 111 E2 (2) - 2018<br /> <br /> points; and environmental domain are from<br /> 8 to 40 points. The raw scores of each domain were then converted to a scale of 0 to<br /> 100 to compare with other populations, with<br /> lower scores indicating poor QoL. A domain<br /> was treated as missing when over 20% of<br /> its items were missing. With regard to QoL<br /> scores, they are on a positive scale (higher scores represent better QoL) and there<br /> is no cut-off point to determine a specific<br /> score by which the QoL could be assessed<br /> as “good” or “bad” [13].<br /> Data analysis<br /> Data had been cleaned by checking<br /> missing data before it was entered into the<br /> database. Data was entered and cleaned<br /> for outlier and illogical data using Epidata<br /> software, then converted into file.data to be<br /> analyzed in Stata version 10.<br /> The results were initially analyzed using<br /> means, standard deviations, and frequencies. Mean and standard deviation were<br /> used to assess normal distribution. Subsequently, Man-Whitney tests were employed<br /> to compare means between the four domains of QoL by socio-demographics.<br /> The relationships between each domain<br /> of QoL were identified by conducting Spearman tests, since domains of QoL were not<br /> normally distributed. To analyze the influence of independent variables of each domain of QoL, bivariate and multiple linear<br /> regression analysis were used, in which<br /> dependent variables were transformed into<br /> ranks because of the absence of normal<br /> distribution (physical and social variable<br /> was squared to meet this condition). Some<br /> socio-demographic factors (age, marital<br /> status, gender, occupation, education lev117<br /> <br /> JOURNAL OF MEDICAL RESEARCH<br /> els, and living arrangement) and significant factors in bivariate linear regression or<br /> in literature documents were then put into<br /> multiple linear regression for the full model.<br /> The final model was selected by performing<br /> stepwise linear regression. The significance<br /> level adopted for statistical test was 5%.<br /> Co-efficient, constant, p value, confidence<br /> interval and R-square for each model were<br /> calculated and presented.<br /> The final model was tested for its fitness<br /> by 1) checking its linear predicted value<br /> (_hat) and linear predicted value squared<br /> (_hatsq); 2) check goodness of fit ("predict<br /> resid, r"; 3), by checking for multi-collinearity.<br /> 3. Ethics<br /> The risk of discomfort to participants and<br /> risk of confidentiality loss were marginal.<br /> There were some questions about individual feelings among the elderly about their<br /> happiness with their life, family members,<br /> sex life, and surrounding physical environment, as well as their social connectedness.<br /> To reduce these risks, in the consent form,<br /> participants were advised that they can withdraw at any time and that they can refuse<br /> to answer any question which made them<br /> uncomfortable. They were also advised that<br /> all their refusal or withdrawal will not have<br /> any effect on them in any way. Moreover,<br /> an anonymous self-administered questionnaire was developed and used, in which<br /> can complete by participants without the<br /> survey privately. Additionally, participants<br /> were asked to return their completed questionnaire by themselves to the field workers,<br /> who are outside the participants’ wards. The<br /> 118<br /> <br /> consent form with participants’ agreement<br /> to participate in the study and their administration group numbers was detached from<br /> the main body of the questionnaire and sent<br /> to the principle investigator (PI) to be securely stored. Therefore, all individual information will be separate throughout the data<br /> collection procedure. Our approach was to<br /> ensure that participants feel that they have<br /> control over the proceedings of the survey.<br /> They were clearly advised that all information is anonymous and will only be analyzed<br /> at the group level. In the consent form, the<br /> PI’s contact number was printed and participants were instructed to if they have any<br /> questions. If participants do become distressed during or after filling out the questionnaire, they could also contact the PI for<br /> further counseling.<br /> All survey questionnaires were anonymous (no name and individual address<br /> identified) and securely stored. This study<br /> was submitted and approved by the Ethical<br /> Committee of the School of Public Health<br /> and accepted in May, 2012.<br /> <br /> III. RESULTS<br /> Among the 299 participants, the proportion of males to females was balanced at<br /> 48.8% and 51.2%, respectively. The mean<br /> age of study participants was 70.6 years,<br /> while the mean age of males was higher<br /> than females (p < 0.05). The proportion of<br /> the elderly in the group under 70 years was<br /> 45.5% compared to these age 70 years and<br /> older 54.5%. The majority of participants<br /> (40.6%) were post-graduation, working<br /> as government officers (80.3%), married<br /> (84.6%) and living primarily with their husJMR 111 E2 (2) - 2018<br /> <br />