Factors influencing the quality of life among Indonesian elderly: a nursing home-based cross-sectional survey

Purpose – In spite of being a significant public health concern, quality of life (QoL) amongst elderly in nursing home (NH) settings is rarely analyzed. The purpose of this paper is to examine the level of QoL and factors influencing QoL amongst elderly NH residents in Indonesia. Design/methodology/approach – A survey was conducted amongst 181 elderly at three NHs in three districts in Yogyakarta province, Indonesia. Purposive sampling was used for the study site and sample collection. Face-to-face interviews were performed using the WHO Quality of Life BREF Indonesian version questionnaire to assess elderly QoL. Multivariate linear regression was performed to determine the factors influencing the QoL amongst elderly NH residents. Findings – The response rate was 66.3 percent. A total of 64.1 percent of elderly had a fair level of QoL, whereas 16.6 percent still had a poor level of QoL. A total of 86.7 percent of elderly lived in an NH due to compulsion, and more than half (53.6 percent) perceived inadequacy of care. The QoL was significantly low amongst those who live in NHs due to compulsion, no social support resources, not receiving any kind of support, having three chronic diseases and perceived inadequacy of care ( po0.05). Multivariate analysis revealed that perceived adequacy of care reasons for living in NH was associated with QoL ( po0.001). Originality/value – Perceived adequacy of care and reason for living in an NHwere highlighted as predictors of QoL amongst elderly NH residents. Improving adequate healthcare services and developing treatment strategies to enforce the adaptation process is required in order to maintain the QoL in elderly NH residents.


Introduction
The proportion of elderly aged 60 years or above is the fastest growing population globally [1]. Additionally, the increment of life expectancy and decline of birth rates have concurrently resulted in a drastic increase of the ageing population in Indonesia [2]. The ageing population in Indonesia was 8.6 percent of the total population in 2015, this number is predicted to increase to 14.1 percent in 2030 and will be almost triple in 2050 [3]. This particular condition places Indonesia as the eight largest elderly population globally and ranks the country third among 25 Asia-Pacific countries [3,4]. 95%  amongst the urban elderly community in India revealed a higher QoL mean SD score when compared to our study finding [22]. A much higher (58.77) overall QoL mean score was found in another study which was conducted in the NH study setting [9]. The different findings of mean scores in QoL might be due to the difference of observed associated factors which influence QoL in different settings. The different instrument tools used to assess the QoL could also be responsible for this variation in results.
In terms of the QoL domains, our study found that physical and psychological domains were the lowest two mean scores compared to social relationship and environment domain mean scores (Table I). These findings are similar to a study in Macao that revealed that older adults had significantly poorer scores on physical and psychological domains of QoL compared to the normative data from the Hong Kong Chinese population [23]. These lower mean scores on physical and psychological domains might be related to the socio-characteristics of respondents. The majority of elderly NH residents were female with no partner. Gender and marital status seemed to affect perceived psychological well-being amongst the geriatric population. Han et al. [24] found that there are gender differences in the correlation between QoL and marriage. Married men had a higher QoL compared to single men. On the contrary, single women had a higher QoL than married, separated or divorced women. Another factor which might influence those findings is related to the educational background of respondents. The elderly in our study had no or low educational backgrounds. This factor could probably influence their perceived mental health well-being. It was observed that a lower educational background was a predictor for a poor psychological QoL amongst the elderly, both in the community and in the NH study settings [25].
In the univariate analysis, we found that QoL was significantly low among those who experienced ⩾3 chronic diseases. It seems that the elderly who presented with chronic diseases tend to have a lower QoL. This finding is in line with a study in Turkey suggesting that the presence of a chronic disease declined the QoL amongst the geriatric population [26]. Lower QoL was also a significant factor among those who had no social support resources and had not received any kind of support. Unsar et al. [27] emphasized the importance of family support or living with spouse. Moving in to an NH could become a stressful situation for the elderly. It might be due to the change of environmental conditions and adaptation processes, including having to live apart from their family or spouse. Although social support may not eliminate the stressful situation, it would make the elderly more optimistic. Therefore, social support is an important resource for better health-related QoL among elderly NH residents [28]. Social support helps the elderly to overcome difficult situations, create new solutions and reduce their despair [29,30].
QoL was also found to be significantly low among those who live in an NH due to compulsion in the univariate analysis. The majority of our respondents lived in an NH due to compulsion due to reasons such as poverty, living alone at home, no one to take care of 95%  them and homelessness and living as a beggar, resulting in the municipal public order agency (Satuan polisi pamong praja) picking them up from the street. All those situations classified this group of elderly into "neglected" or "almost neglected" in the households [20]. The national census stated that about 12 percent of the elderly in Yogyakarta were living alone. This percentage was ranked third compared to all provinces in Indonesia [20]. These forced conditions might affect their psychological domain of QoL, and influence the whole QoL. This assumption is relevant when we look at the psychological domain mean score of QoL in this study finding which was lowest when compared to physical, social relationship and the environment domain of QoL (Table I).
Perceived adequacy of care was also revealed as the associated factor for QoL among elderly NH residents in the multivariate analysis. More than half the respondents perceived inadequacy of care from NH staff who should be responsible for taking care of the elderly. In a geriatric population, most diseases are chronic. A lower QoL was also significantly associated with those who have ⩾3 chronic diseases in our study finding. Apart from medical treatments, psychological and social support might definitely be required through a wide range of services, home care or long-term care in certain units [28]. Once the elderly are admitted to hospitals or NHs, their interactions with family or surrounding people are severely limited [31]. This sudden change of environmental situations might lead to stressful conditions for the elderly. Therefore, when the elderly were not be able to take care of themselves, both healthcare professions and social workers were needed to assist them in activities which cannot be performed, to provide health education and emotional support as well [32]. Providing adequate care should be given equal importance to providing adequate social support toward elderly NH residents. Both professionals in healthcare and social workers should recognize that social support is associated with health-related QoL and pay attention to the importance of social support for the residents in daily practice [27].

Conclusion
The majority of elderly NH residents in Yogyakarta province, Indonesia, had a fair level of QoL. Furthermore, perceived adequacy of care and reason for living in an NH were highlighted as predictors of QoL amongst those elderly NH residents. Hence, improving adequate healthcare services and developing treatment strategies by the professionals in healthcare and social workers is required in order to support the QoL maintenance in elderly NH residents.

Strength and limitation
The strength of our study is the widely validated and reliable WHOQOL-BREF questionnaire used to assess the QoL in various study settings. The limitation of this study is that the sampling technique was purposive since we only selected government NHs for the study sites. Therefore, the findings may not be generalized to the private NHs due to the genuine background, facilities and service availability differences among elderly NH residents. Regardless, this study finding could be beneficial as baseline information in making the future policy strategies relating to the QoL amongst elderly in NHs, particularly in underlining the important roles of social workers and health-related professions on QoL in elderly NH residents. 3. Teerawichitchainan B, Knodel J. Data mapping on ageing in Asia and the pacific: analytical report.

Introduction
The purpose of this study was to determine the relationship between early initiation to breastfeeding with stunting in children 6-59 months. Stunting in children under five years is a global problem, mainly experienced in developing countries.  [3][4][5]. Jambi Province has a stunting problem greater than the national average by 37.9 percent [5]. Stunting creates a barrier to the forming and development of organs and has a short term impact on mortality, morbidity and disability. Long term consequences include negative impacts on adult body size, intellectual ability, economic productivity, reproductive ability and metabolic and blood vessel disease [1,6,7].
Diarrhea, cough, cold, fever, shortness of breath, are associated with stunting in children under five [8] and is a proven risk of respiratory tract infection by four times [9]. The risk of stunting increases based on an unhealthy source of drinking water [10] and poor sanitation facilities [11,12]. Low participation in basic community heath checks is another risk factor of stunting in urban areas [13].
Early initiation to breastfeeding, exclusive breastfeeding, duration of breastfeeding, and vitamin A administration are associated with preventing stunting in children under five years of age. Previous research indicates an exclusive breastfeeding relationship with stunting [14,15] and a relationship between supplementation to breast feeding during the first four months ( p ¼ 0.04) with stunting in children [16].

Methods
Data from the National Nutrition Survey in Jambi province in 2015 were studied. Data were collected through Nutrition Status Monitoring Activities using a cross-sectional design including questionnaires, interviews and observations. The status of exposure and disease were observed simultaneously in individuals over the same period between September and October 2015.
The research location was Jambi Province, Indonesia consisting of ten districts/cities. In total, 30 clusters were selected from each district and city. In each cluster ten households were selected as the sample.
The population of this study was all children aged 6-59 months in Jambi province. The sample criteria included the following: child's birth date is known, the child is weighed at birth and during observation, the child's length is measured at birth and during observation, the determinant of stunting questionnaire in the child is complete.
The selection of clusters in districts and cities is systematically randomized based on probability proportional to size. The nutritional status used was collected by looking at birth records, viewing MCH books, interviews with toddlers and field observations. Stunting is measured anthropometrically according to WHO standards 2005.
Data collection was conducted by graduates of the Nursing and Midwifery Academy. There were six people in each District/City and two field supervisors. The Data collector and supervisor team attended three days training on September 18-20, 2015. Training materials comprised: nutrition program policies, sampling methods, interview techniques, anthropometric measurements, questionnaire content and field practice.
Univariate analysis was performed to provide a stunting prevalence picture in children 6-59 months, early initiation to breastfeeding and other variables. The bivariate analysis was conducted to find out the association of each independent variable with stunting variable in children 6-59 months. Multivariate analysis was conducted to see the relationship of early initiation to breastfeeding with stunting in children 6-59 months. Analysis was done by binary logistic regression for survey data.
Early initiation to breastfeeding and child characteristics 6-59 months Most children of 6-59 months do not get the opportunity to breastfeed early (54.1 percent; 95% CI: 51.1-57.1) with irregular visits to community base health services/posyandu (44 percent). The number of boys (52 percent) is almost the same as girls (48 percent).

335
Early initiation to breastfeeding A small number of them do not have good sanitation facilities (14 percent). Some of them are not exclusively breastfed (41 percent). A small number of them have a history of diarrhea in the last one month (15 percent). Over a third of them had a history of cough (38 percent), cold (45 percent), fever (35 percent) in the last one month. A small number of them do not boil drinking water (13 percent). Most of them live in rural areas (74 percent). Father and mother ethnicity is predominantly, non native or "The Outsider" (41 percent) and Mixed (12 percent). Some of them have well water as a source of drinking water (57 percent). A few (3 percent) of their families have an inadequate quality source of drinking water (cloudy/smelly/tasty/colorful). A small number of them have poorly structured homes (3 percent). A small number of them have a large number of household members (18 percent). Fathers education (53 percent) and mothers education (55 percent) are basic. A minority of fathers jobs are "not fixed and others" (15 percent) while a small portion of mothers (21 percent) are working. Few of them weigh less than 2,500 gram (3 percent). More than a quarter of them had a birth length of less than 48 cm (27 percent) ( Table I).
Other results of the analysis showed a significant association between water source and birth weights to stunting (Table II). The most dominant risk factor for stunting was age 24-59 months with a risk 2.5 times higher than the age of 6-11 months ( p ¼ 0.000; AOR ¼ 2.5; 95% CI: 1.8-3.6).

Discussion
Stunting in children 6-59 months Based on the research results, it is known that 27.5 percent (95% CI: 25.2-29.9) of children aged 6-59 months were stunted in Jambi province in 2015. This is lower than 2007 (36.4 percent), 2010 (30.2 percent), 2013 (37.9 percent) but still remains a serious problem that needs further treatment. If the trend of stunting problems continues, an estimated 127 million toddlers will be stunted by 2025 [1,2].
In 2012, through the World Health Assembly Resolution 65.6, a comprehensive effort on the nutritional condition of mothers, infants and children, known as the "six global nutrition targets for 2025" was recommended. This policy concerns the first target of a 40 percent reduction in the number of children under-five who are stunted [2]. Nutritional interventions alone (specific programs) are not sufficient to achieve optimal child development and nutrition, as it is important to strengthen sensitive nutrition development efforts and improve them with evidence-based, nationwide multi-sectoral plans aimed at stunting, combining nutritional interventions with health, family planning, water and sanitation strategies and other factors that influence the stunting risk (sensitive programs). It takes a measurable work plan to develop and oversee the implementation of commitments and targets [1,17].

Early initiation to breastfeeding and stunting
This study shows that children who do not get an early initiation to breastfeeding are 1.3 times more likely to be stunted than those who are breastfed early. These results illustrate early initiation to breastfeeding as a form of maternal care and the best giving of nutrients early in life that can reduce the risk of stunting. Early breastfeeding initiation is one of the entrances to successful breastfeeding at a later time and ensures children receive appropriate nutritional intake [2,7].
Similar results are proven in previous studies. Initiation of breastfeeding after six hours of birth or discarding colostrum due to ignorance of its health benefits as well as breastfeeding supplementation is a risk factor for stunting children under five [18]. This study shows that one of the efforts to overcome stunted problems in children is to provide early initiation to breastfeeding. Early initiation to breastfeeding takes place naturally, where the child after birth is directly placed in the mother's chest to initiate the search for the mother's nipple to feed within the first hour of birth. Early initiation to breastfeeding is 336 JHR 32,5 Children

337
Early initiation to breastfeeding easy, takes a short time and has a long-term effect on the child. This opportunity can be accomplished if the mother of the newborn is aware of the benefit and process of early breastfeeding. Promotion efforts to educate the mother and family should be carried out before birth.
Other results of the analysis showed a significant association between age, water source and birth weights to stunting in 6-59 months of children. These results indicate that stunting is caused by many factors.
Early initiation to breastfeeding ensures that the baby gets colostrum that increases the child's immunity to infection and furthermore, exclusive breastfeeding will keep children away from contaminated water sources for at least the first 6 months. Adequate foods and no infection make the child grow optimally [19][20][21][22] and reduces the risk of stunting. The difference in risk of stunting based on drinking water sources was also revealed by several other researchers [10,23].
The results of this study indicate that there are differences in the growth of children by age. Growth disorders will accumulate with age [24][25][26][27]. Differences in the growth of children by age are also found elsewhere. For two years the age of the child was 9 cm in contrast to the WHO standard in Malawi, then only increased by 1 cm by age 2-3 years from 10 cm difference. Differences of two centimeters (20 percent) occur at birth, 4 cm (40 percent) occur from birth to 12 months of age, and 3 cm (30 percent) occur from 12-24 months. Early cause is a combination of growth disorders in the uterus, inadequate nutrition after birth, marked by low exclusive breastfeeding (28 percent) and low nutritional intake of breast milk due to malnourished mother, and infection after birth [27].

Conclusion
Early initiation to breastfeeding can be associated with preventing stunting in 6-59 month of children after adjusted to all others variables. There is also a significant association between water source and birth weight to stunting. Children with a low birth weight are 2.0 times at higher risk of stunting than normal-born. The most dominant risk factor for stunting was age 24-59 months with a risk 2.5 times higher than the age of 6-11 months. Early initiation to breastfeeding is easy and has a long-term positive health benefit on the child.

Recommendations
Promoting breast feeding benefits to mothers should begin before birth. Support from health personnel, especially officers who assist the birth process, should be increased.
It is important to reinforce the importance of early initiation to breastfeeding and clean drinking water sources as a prevention to stunting. Improvement of drinking water sources by replacing contaminated river water with clean pipe water should be practiced. Finally, it is important to introduce additional and more regular educational programs for pregnant women about how to prevent low birth weight in babies together with clear information on the benefits of early breastfeeding.  [4]. In 2011, the WHO has reported that over 29m (29,063,194) people have died from diseases requiring palliative care at the end of life [5]. Most patients with advanced stages of cancer suffer from pain, fatigue, anxiety, depression, dyspnea and other symptoms [6,7]. Moreover, when confronting death at the end of life, patients may suffer certain phobias, i.e. fear of pain, lack of all control, loss of dignity and fear of dying alone [8]. As a result, holistic palliative care plays an important role not only in controlling symptoms but also maintaining or improving patient quality of life (QOL). Palliative care focuses on the relief of suffering and supports for the best possible QOL for patients facing serious life-threatening disease and for their families regardless of whether the final outcome is death or cure of the illness [9,10]. Palliative care is an approach to care for patients with potentially life-threatening disease and their families, focusing on all aspects of holistic health care such as pain treatment, advising family members, referring to social services. The ultimate goal is to provide patients and families with the best QOL during the trajectory of diseases [11]. The data from previous studies showed that many terminally ill patients would prefer to stay and die in their own homes, but unfortunately some of patients may not be able to do so [11][12][13][14]. Evidence has presented many benefits of home care for palliative care patients. These are, for example, relieving the symptom burden, increasing the patients' self-management of the illness, increased receipt of sufficient information to handle an illness emergency, relieving the caregivers' burden, and the positive effects on the satisfaction of care [15]. Despite the fact that many studies have also demonstrated that palliative home-based care by physicians or nurses supports patients to die at home, most were performed in countries where palliative is integrated into mainstream medicine. Regarding the latest report from the Worldwide Palliative Care Alliance in 2014, Thailand is classified as a country where palliative care is only provided in some areas [4].
Palliative care services delivery was the highlighted national policy in Thailand [16]. Palliative care services in Thailand are mainly delivered in hospital settings; integrative palliative care from hospitals has not been effectively linked with community settings [17,18]. Wright et al. [19] classified palliative care activity across the globe, and countries were allocated to one of four categories: no known palliative care activity, capacity building activity, localized care provision, and palliative care services integrated in the health system. Thailand is classified as a nation in category 3 because localized palliative care provision is available in hospital. The improvement of community and home-based palliative care programs is needed to increase the coverage of palliative care services [20]. Moreover, costs of medical care in home programs are less than those in acute services care used in emergency departments and hospital wards [21,22]. No national statistics in Thailand are available on the rate of death at home or are related directly to the place of death. It differs from western countries such as in the UK, where the rate of death at home increased from 18.3 percent in 2004 to 20.8 percent in 2010 [23].
The essence of palliative care is that the holistic care includes the psychological, social, and spiritual needs of the patients [24]. Thus, the needs of patients with advanced stages of cancer become a healthcare issue. However, the literature reveals that most of research studies were conducted to assess needs of palliative care among patient in hospital [25][26][27][28]. The result of this study will inform professional nurses and guide policies on how to provide appropriate health care services for patients with advanced stages of cancer. Moreover, an expected outcome of this study could guide care at home aiming to improve patients' QOL and place when the end of life comes. Therefore, the objective of this study aimed to explore the needs of patients with advanced stages of cancer in home setting.

Design
The concurrent mixed-method design, combining a major quantitative and minor qualitative, was used [29]. First, descriptive design was conducted to explore specific needs and problems among elderly patients. The study populations included 110 patients who enrolled in 343 Needs of patient with advanced stages of cancer palliative care program at Sapprasitthiprasong Hospital from January 2014 to March 2014. Inclusion criteria comprised of patients 60 years who were diagnosed cancer stage 4 by the oncologist and were receiving treatment to alleviate symptoms and promote their comfort rather than cure their illness. Patients were excluded if they had a problem with cognitive impairment or a health status that did not allow them to comprehend interview questions.
Second, semi-structure interview was conducted to explore specific needs and problems among elderly patients focusing on home-based care. The interview was constructed on the bases of the item of the questionnaire choosing those considered less deep in the evidence base. A purposive sample of 8 patients was subsequently selected for the interviews. Participants who enrolled in home-based palliative care for at least 1 month were identified by an oncologist nurse. Interviews were taped on tape recording machines and transcribed verbatim. Each participant was given a code number for identification purposes to maintain their anonymity. New participants were added until saturation of the data was reached [30].

Measurement
A self-report questionnaire was developed by the researcher based on literature reviews aimed to explore patients' needs in the study. The general characteristics were created for the collection of basic demographic and clinical data.
The palliative care needs were developed by the research based on literature reviews consisting of 13 items. It asks about different of needs: information (item 1-3), physical (item 4-7), emotional/spiritual (items 8-10) and social supports (item [11][12][13]. Each item represents one need of cancer patients and scale was rated on a scale from 1 (not importance) to 5 (very importance). Five experts were invited to assess content validity index and the questionnaire was revised according to their comment. Afterwards, the reliability was tested by a pilot test on a convenience sample of 30 patients with advanced stages of cancer at the OPD/IPD and returned to the researcher. The reliability of the questionnaire, as well as the reliability of the questions of each dimension, was measured with Cronbach's α coefficient. Results showed that the Cronbach's α coefficient for the overall questionnaire was 0.852.
The semi-structure interview is presented in Table II. All interviews were conducted at patient's home. Individual oral interviews were conducted in the Thai language.

Data analysis
The characteristics of the participants were presented in terms of frequency and percentage for categorical data and mean with standard deviation for continuous data. The relationship between individual variables was assessed with Spearman's correlation coefficient. Significance was set at po0.05. Regarding the qualitative data, we used it to confirm the quantitative findings by examining the coded units of meaning in each narrative content and then recorded to identify the emerging themes. Content analysis was used to explore the needs of home-based palliative care [31]. This study used data triangulation through data collection to identify major issues. The data that were derived from the interviews reached a saturation point when no new information emerged from the interviews. All the transcriptions were read and analyzed by the researcher. The analysis proceeded by defining codes that were derived from the data. The combination of larger thematic categories was used, which were coded and linked using notes and interview summaries [32].

JHR 32,5
Results of the quantitative study Sample characteristics A total of 111 patients participated in this phase. The majority of the participants were female (64.0 percent), with mean age 66.16 years (SD ¼ 6.73) ranged from 60 to 84 years, living with spouse (71.2 percent), Buddhist (93.7 percent), and attained primary school grade (82.0 percent). The health care costs were mostly covered by universal coverage scheme (94.6 percent). Most of them preferred home as a place for caring (77.5 percent). In this sample, the most common cancers in male were liver (32.5 percent) and lung (27.5 percent). The most common cancers in female were breast (29.6 percent) and cervical (11.3 percent) ( Table I).
Needs of palliative care among patients with advanced stages of cancer The participants had high levels of needs for all 13 items, and the four most important needs among cancer patients for palliative care included information needs (98.2 percent), patients needed more treatment for pain (97.3 percent), needed more information for caregiver (95.5 percent) and health volunteers visit at home (95.5 percent) (Table II). There was no correlation between patients' higher age, income, duration of being diagnosis and patients' needs (Table III).

Results of the qualitative study
Based on content analysis, we found three core needs of patients with advanced stages. First, our interviews revealed the issue of type of information needs, pain management, and social support.
Type of information needs. Type of information, the first core quality of palliative care, was described by participants. The participants were described as being ill-informed regarding a chance to ask about cancer treatment and care. First, patients needed information about diseases and treatment: As the time passes and symptoms get worst, the participants need information about chances of cure and progressive of disease. As participants described: Two essential needs for an effective pain management were identified by cancer patients. The first need was for 24 hours palliative care cover and medication: Providing psychosocial support. Psychosocial support involves psychosocial-spiritual comfort. Patients expressed their need of psychosocial support. The first need was patients needed friends and did not want to be alone. As participants described:

Discussion
The quantitative results revealed that slightly more than half participants required caring at their home. The place of death is a major theme among patients in palliative care delivery research. Most patients at the end of life prefer dying at home [14]. The result demonstrated the greatest needs of a majority of patients that they require better understanding about their diseases, prognosis and medical treatment. This study is similar to the study of Morrison and colleagues [33] which indicated the top five needs concerning the treatment, care, and health information. A 2012 systematic review [34], aimed to explore unmet needs of palliative home care patients and caregivers, indicated that they need more information about diseases, in terms to manage their lives and make decisions. They also wanted a specific information regarding treatment, medication, and side effects. The core information-related need include listening actively and providing a safe space for conversation about peaceful death. Steele and Fitch's [35] study revealed that eight of the top 10 most frequently reported needs were not physical, such as fears about the cancer returning or spreading. Moreover, the study in Denmark [36] indicated that women with pre-diagnosis period required an overview of the treatment process supplemented with information, involvement, and help to prepare themselves for treatment. Also, relatives need to be involved.
Effective pain management is the second greatest need among patients with advanced stages of cancer and the greatest need among caregivers. Most of the evidence indicated that pain was a common and devastating symptom of cancer-affected patients [37][38][39]. The 71 percent of patients in Indonesia and 67 percent of patients in Netherlands suffered from pain and pain-related difficulties with everyday activities [40].
Providing psychosocial support is the last need. Our finding appeared to be consistent with Yi et al. [41] that health professional was the most important needs category; information was next; and psychosocial needs were the most unmet needs. The psychosocial supports had the highest mean score in the unmet needs category. The strongest influencing factor was "no one to talk with." Another aspect which emerged from the analysis of qualitative data is touching. The findings provide some relevant implications for nursing 348 JHR 32,5 practice, education and research. Nurses' roles in community health nursing should be to know the needs of their patient while caring at home setting. With this knowledge, nurses should be able to anticipate patients' needs, and enhance their experiences by providing appropriate interventions.
Regarding the strengths of our study, this is the first to provide an in-depth description of the needs of palliative care at home setting. The inclusion and exclusion criteria for participants contributed to population representatives, and the results may therefore be transferred to other similar contexts.
This study may provide only limited insight into the influence of patient's experience on their palliative care services from one hospital. Since the data were collected in the end stages of patients, the interview was always terminated when patients had pain and others symptoms distress. Only some data related to research objectives were presented. We, therefore, acknowledge that future studies should attempt to know what is a need and barrier of palliative care among cancer patients and their families.

Conclusion
The current evidence suggests patients with advanced stages of cancer required home as the best place for their end of life. Therefore, home-based care supports patients to relieve suffering from symptom distress, and also helps them in improving QOL in their environment with their loved ones. 8 Dengue is a mosquito-borne disease found mainly in countries with tropical and subtropical climates. The global prevalence of dengue has grown dramatically in recent decades. Currently, about half of the world's population is at risk of infection [1]. Globally, one recent estimate indicates 390m dengue infections per year (95 percent credible interval of 284-528m), of which 96m (67-136m) manifest clinical symptoms [2]. An estimated 500,000 people with severe dengue require hospitalization each year and about 2.5 percent of those infected dies [1].
Dengue fever is a disease caused by the dengue virus. The main cause of dengue virus infection in human is through bites from infected female mosquitoes (Aedes aegypti) [3]. Dengue virus is a carrier disease found in all age groups [3]. The symptoms of dengue fever include high fever, chills, fatigue, rash, nausea, vomiting, headache, sore throat and pain (muscle, back, joint and abdomen areas) [4]. In severe cases, it can be life threatening due to serious bleeding and shock [4]. The most effective intervention is to prevent mosquito bites [5].
Thailand is still suffering from dengue fever nationwide and during all seasons. In 2017, Thailand has reported 52,049 dengue cases from all 77 provinces, including 62 deaths [6]. The Ministry of Public Health has adopted a policy to control dengue hemorrhagic fever in the National Health Development Plan No. 11 (2012-2016) which targeted the reduction of dengue hemorrhagic fever rate to not more than 25 percent of the median in the past five years and the reduction of morbidity rate to not more than 0.02 percent [7].
The Northeastern region of Thailand has the largest land area. Nakhon Ratchasima province has the highest population in the Northeast region and the second highest population in the country. The incidence of dengue hemorrhagic fever in Nakhon Ratchasima in the past five years (2013-2017) were 269.29, 33.58, 274.53, 62.45, 65.38 per 100,000 population, respectively. In 2017, the prevalence of dengue hemorrhagic fever in Nakhon Ratchasima were 1,716 cases with the morbidity rate of 65.38 per 100,000 population and two cases of death by dengue hemorrhagic fever, giving the mortality rate of 0.08 per 100,000 population [6]. The model forecasting of dengue hemorrhagic fever in 2017 showed that Nakhon Ratchasima was the high-risk area to monitor the disease [8].
It is believed that the outbreak of dengue hemorrhagic fever is mainly from mosquitoes and by the general nature of mosquitoes, like laying eggs in containers of water inside and outside the house. This is due to the behaviors of local rural people. Some behaviors that may not be appropriate include disorganized house, inadequate lighting management, hanging dirty clothes in the house, not covering water storage container and leaving wet waste with water. As a result, mosquito breeding becomes widespread [9].
Dengue morbidity can be reduced by applying effective communication that can achieve behavioral outcomes that augment prevention programs [10]. At present, the main method to control or prevent the transmission of dengue virus is to combat vector mosquitoes through preventing mosquitoes from accessing egg-laying habitats by environmental management and modification, active monitoring and surveillance of vectors to determine the effectiveness of control interventions [5,10]. The Health belief model (HBM), developed from the theory of social psychology, describes the behavior of individuals [11]. The HBM believes that people who change their behavior must perceive their susceptibility, perceived severity, perceived benefits of modifying health behaviors and perceived fewer barriers of preventive behaviors, cues to action, modifying factors and health motivation [11,12]. Therefore, from HBM constructs, the researcher expected to apply the theory of HBM to use in the prevention of dengue disease because theory says individuals will seek ways to follow the recommendations for prevention and rehabilitation as long as the disease prevention practice is more positive 353 Effectiveness of the intervention program than the difficulty [11]. By following these theory instructions, a person must feel fear for the disease or feel threatened. In addition, a person must feel to have an ability for disease prevention [13]. Previous studies in Thailand have adopted the HBM to modify dengue prevention behaviors and received good results [14,15]. Therefore, this research is based on the HBM as a theoretical framework for the prevention measures and for designing intervention activities.
It is necessary to encourage people in the community to receive knowledge about the disease, promote risk perception and benefits of disease prevention so that people can find solutions to reduce the barriers to disease prevention then take action to prevent dengue fever. The dengue hemorrhagic disease prevention in the community must be supported by the community and apply the campaign for community awareness of the dengue problem. Previous studies have focused mainly on source reduction of water containers in a household and vector control [16][17][18]. These studies earlier did not investigate the combination of vector control activities, and the behavior changing based on the HBM for dengue fever prevention. For this study, the researchers are interested in examining the effects of the dengue hemorrhagic fever prevention intervention program in rural communities. The data from this study will lead to health promotion planning for dengue hemorrhagic prevention in rural communities. The focus is on promoting knowledge in disease prevention, raising awareness of risk and severity of disease, encouraging the benefits of disease prevention and reducing barriers to disease prevention. The main objective of the community intervention program is for eradicating dengue hemorrhagic fever which is a major public health problem in Thailand.

Research objective
The purpose of this study is to examine the effectiveness of the intervention program for dengue hemorrhagic fever prevention among rural communities in Thailand based on the theory of HBM (Figure 1).

Study design
This is a quasi-experimental study that examined the effect of dengue hemorrhagic fever prevention intervention in rural communities. Participants were randomly divided into two  groups as explained below. The experimental group received a five-week intervention program while the control group continued life as usual. Variables were measured before the intervention, after the five-week intervention, and at three-month follow-up.

Sample size
The sample size was calculated by the following formula [19]: Participant inclusion criteria include: at least 20 years old; Thai nationality; both male and female; has lived in the community for more than six months; able to answer questions, no problem speaking, listening and communicating; no training on educational program for dengue hemorrhagic fever prevention in the previous six months; willing to participate in this study; and able to sign the informed consent form.
The criteria for excluding participants from this step are: participants who have difficulties communicating in Thai; and participants who are not available at the time of data collection (Figure 2).

Participants
Two districts were randomly selected from Nakhon Ratchasima province. Then, two sub-districts from two districts were randomly chosen. Further, two villages were randomly selected from the two sub-districts to be part of the study. To prevent the contamination of data, the entire village was randomly assigned as either experimental group or control group. Individual household from each village is randomly selected to be included in the study. They were excluded if they had lived in the community less than six months, had symptoms or illnesses that limit activity, or had participated in any education program for dengue prevention during the prior six months.

Data collection
Participants were randomly divided into two groups (intervention group, n ¼ 33; control group, n ¼ 33). The intervention group was assigned to a five-week program. The control group received the usual health education from public health personnel. Evaluations by questionnaire were measured three times for both groups. The questionnaire consisted of 52 questions that took approximately 45min for respondents to complete. Written consent forms were obtained from participants prior to data collection. This study was approved by the Committee of Human Ethical Research, Maharat Hospital, Nakhon Ratchasima.

Research instrument
Part 1: demographic information included seven items on gender, age, marital status, highest education, occupation, income and information obtained from the community.

Effectiveness of the intervention program
Part 2: knowledge about dengue hemorrhagic fever included 20 items that assessed participant's knowledge of the causes of dengue fever, signs and symptoms, treatment and prevention. The score of 1 point was given for each correct answer and 0 point for each wrong answer. Part 3: perception of dengue hemorrhagic fever prevention included 20 items which derived from four main constructs of HBM: perceived susceptibility, perceived severity, perceived benefits, and perceived barriers to prevention. Perception measurement utilized a three-point Likert scale that included disagree (1), neither agree nor disagree (2) and agree (3).
Part 4: dengue fever prevention practices were measured using five items derived from principle for dengue hemorrhagic fever prevention practices from the Department of Disease Control, Ministry of Public Health, Thailand [20].

Intervention
Participants in the control group did not receive any interventions. Participants in the experimental group were enrolled in the intervention program based on the HBM theory. The HBM has four constructs representing the perceived threat and net benefits: perceived susceptibility, perceived severity, perceived benefits and perceived barriers [12]. The specific intervention program included four main activities based on HBM theory.  Activity 1: providing knowledge of dengue hemorrhagic fever Implementation timeframe: Weeks 1-5.
The objective of this activity was to provide the knowledge of dengue hemorrhagic fever through the daily community broadcast. The messages were based on HBM constructs as described below: (1) Raise the perceived susceptibility: "dengue is a problem in community for all age group." (2) Raise the perceived severity: "it is a little one but it is the one that can kill you." (3) Raise the perceived benefits: "take care of the house environment, then the safe environment will protect your family from dengue hemorrhagic fever." (4) Reduce the perceived barriers: "just few minutes for easy clean-up can reduce many risks from dengue." Activity 2: dengue hemorrhagic fever campaign Implementation timeframe: Week 2.
The emphasis is on individuals to receive information through campaign activities: (1) The individuals who took part in the campaign consisted of the head of the village, assistant head of the village, village health volunteer, adults in the village and students in the village. Mosquito mascot and cartoons were also part of the campaign parade.
(2) Poster boards were used during campaign parade to raise awareness of dengue hemorrhagic fever such as principle for dengue hemorrhagic fever prevention practices from the Department of Disease Control [20] and mosquito life cycle.
(3) Campaign announcements about practices to eradicate larvae using temephos or abate sand and using mosquito repellent.
(4) Distribution of leaflet and messages on dengue hemorrhagic fever, such as signs and symptoms of dengue hemorrhagic fever.
(5) All participants in the parade campaign went to all the houses in the village to identify mosquito breeding sites. If an open water container was found, the participants removed all the water from it. Activity 3: a contest for safety dengue house Implementation timeframe: Weeks 3-4 and announcement of the winner during Week 5. The objective of this activity was to find a model house that is safe from dengue and to encourage villagers to see the importance of environmental management and create good examples for their neighbors: (1) Defined the attributes of the house that is safe from dengue on the basis of the hygienic and clean house assessment from the Department of Disease Control, Thailand [20] then selected major topics in the assessment. An award-winning house has to show the important features of all attributes.
(2) Assigned the committee to evaluate the houses that participate in community contest. The committee consisted of seven representatives of the community,

357
Effectiveness of the intervention program including the village head (one), the assistant village heads (two) and village health volunteers ( four).
(3) Public announcement to recruit people in the community to the contest and to inform the rules of the contest.
(5) The announcement of the award for the winner and the second place during the group activity at the village hall.
(6) The winners received a large certificate to display at the front of their houses to be examples to neighbors and motivate them to be safe from dengue.
Activity 4: group education Implementation timeframe: Week 5. The objective of this activity was to provide knowledge about dengue hemorrhagic fever. Activity was set in the community hall with the following activities: (1) knowledge exhibition about dengue, mosquito's repellent and methods for eradicating mosquitos; and (2) stage play and role play that reflected the susceptibility for risks, the dengue severity, the benefits of prevention, and reduction in barriers to prevent dengue hemorrhagic fever.

Validity and reliability of research instruments
The intervention program and research instrument were adapted from the literature review based on the HBM theory and were reviewed by experts in the field. Content validity: a panel of three experts evaluated the content validity of the intervention program and research instrument. For content validity testing, CVI was analyzed and found to be 0.89.
Reliability: the questionnaire was tested for reliability with 30 people with similar characteristics to the samples. For internal consistency reliability testing, Cronbach's α coefficient was analyzed. All scales had good levels of internal consistency of more than 0.70.

Data analyses
Statistical Packet for the Social Sciences 23.0 was used for data analyses. Descriptive statistics were calculated to describe demographic characteristics and other backgrounds of the participants. To compare the data between the two groups, t-test and χ 2 test were used. Two-way repeated measures ANOVA was used to analyze the difference in the total scores of six scales for dengue hemorrhagic fever prevention between the two groups across times of measure. The χ 2 test was used for analyzing the difference in the number of house index (HI) between the two groups.

Results
A total of 64 of the initial 66 participants completed the study questionnaires at three timeframes. Thus, data analyses were performed using 64 subjects. At baseline, there were no significant differences in general characteristics between the intervention group and control group. However, there was a significant difference in gender (see Table I).
358 JHR 32,5 Table II showed HI decreased in both groups at the end of the intervention program and revealed a significant difference between groups for both the post-intervention and the follow-up.
Comparison of the groups before starting intervention revealed no significant difference in the scores of knowledge, perceived susceptibility, perceived severity, perceived benefits, perceived barriers and preventive action (see Table III).
The intervention group had increases in the scores of knowledge, perceived susceptibility, perceived severity, perceived benefits, perceived barriers and preventive action after intervention. Only preventive action showed the interaction between group and time (F ¼ 11.19, p-value o0.001) (Table IV ).

Discussion
A quasi-experimental study was designed to assess the effectiveness of the intervention program based on HBM to prevent dengue hemorrhagic fever.

Effectiveness of the intervention program
Results demonstrated that this intervention program significantly increased the knowledge, perceived susceptibility, perceived severity, perceived benefits, perceived barriers and preventive action. This finding supported the assumption that applying the HBM to the intervention in rural communities can promote the preventive actions and may be beneficial in the primary care of people with a high risk of dengue hemorrhagic fever. The intervention program was created to meet the HBM theory because protective behavior is related to the knowledge, perceived susceptibility, perceived severity, perceived benefits and perceived barriers. Although education campaigns have increased people's awareness of dengue, it remains unclear to what extent this knowledge is put into practice, and to what extent this practice actually reduces mosquito populations. In this study, the intervention encouraged the practice of participants in many ways. The knowledge was provided via daily broadcast to the community, dengue campaign and group education at the village hall. Cues-to-action was implemented via the contest for a model house for safety from dengue, which was in line with an earlier research suggestion [13] that there is a greater likelihood of positive outcomes for preventive actions when the participants are supported and encouraged by the good model in the community.
After the intervention program, the total scores on the six scales for dengue hemorrhagic fever prevention were significantly higher in the intervention group than those in the control group. These results support the effectiveness of the intervention program directly. However, in the follow-up phase, some scores were decreased; this may be because the community engagement tends to be insufficient. Thus, the approach toward enhancing community involvement is important.
The results of this study showed that the HI for the intervention group decreased for both post-intervention and follow-up. The effectiveness of the program on preventive action is consistent with the earlier studied that showed the direct link between knowledge of dengue preventive measures and container protection practice [21]. In order to decrease the breeding site of mosquitoes, it is necessary for people in the community to change the behaviors for the dengue hemorrhagic fever prevention.

Conclusions
Dengue hemorrhagic fever is pervasive among rural communities. In the present study, we conducted a quasi-experimental study to investigate the effectiveness of the intervention program. The effects were measured by the knowledge, perceived susceptibility, perceived severity, perceived benefits, perceived barriers and preventive action. After participation in the intervention program, the five scales showed a significant increase; therefore, the intervention program may be beneficial in primary care in such a rural community.

Introduction
The mangosteen, Garcinia mangostana (GM), which defines as "queen of fruits" [1], is native plant mainly found in the tropical rainforest areas of South-Asian countries, e.g., Myanmar, Thailand, Indonesia, Malaysia, Philippines, Sri-Lanka, etc. It belongs to the Guttiferae Family. GM has known and being used as traditional medicine to treat various types of diseases such as diarrhea, abdominal pain, dysentery, wound-infections and chronic ulcer [1]. According to various studies, GM extracts possess potent anti-oxidant [2], anti-tumoral [3], anti-allergic [4], anti-inflammatory [5] anti-bacterial [6] and anti-fungal activities [7]. Mangostana is famous for its tasty fruit named as mangosteen fruit. It is a dark purple/reddish color fruit with white soft and delicious consumable pulp. This pulp is little bit acidic and sweet in flavor with charming smell. Products containing mangosteen juice/extract are highly demandable in the beverage market; and in the USA, the sales of mangosteen products exceeded up to $200m in 2008. A variety of secondary metabolites have been isolated from GM, xanthone is one of them [8].
Tang et al. [10], in their randomized, double-blind, placebo-controlled trial, reported that, mangosteen intake for 30 days enhanced human immune responses compared to placebo controls. Udani et al. [11] found that intake of blended mangosteen juice by obese subjects for eight-weeks ameliorated the inflammation compared to placebo group. Another randomized, double-blind, placebo-controlled trial by Kudiganti et al. [12] revealed that consumption of meratrim (contained GM fruit rinds) for 16 weeks by healthy overweight subjects, significantly reduced the body weight, BMI, hip size and serum lipid profile compared to the placebo control. After ingestion of 60 mL mangosteen juice by healthy adult subjects with high-fat breakfast, xanthone were detected in serum approximately 762-4,030 nM/L/h and 0.9-11.1 μM in 24 hours urine indicate the well-absorbtion of xanthone when taken with high-fat-diet [13]. Chang et al. [14] unveiled that acute administration of 250 mL mangosteen juice one hour before cycle ergometer exercise, does not have impact on exercise-mediated physical fatigue. The acute toxicity study of ethyl-acetate fraction shows no toxicity to the experimental rat evaluated by physical changes and mortality [15]. Therefore, GM and its isolated xanthones are safe for people.
Side effect of GM There are few scientific reports published about mangosteen side-effect. Daily consumption of mangosteen juice for a long time may produce type-B lactate acidosis in chronic kidney disease patient and metabolic syndrome. This may be due to the releasing of cytochrome-c, by the α-mangostin, cytosol from the mitochondria and directly impair the mitochondrial electron transport that leads to lactic acidosis [18]. Liu et al. [19] found that α -and γ-mangostin from GM can inhibit platelet aggregation and induce cytolysis, therefore it should be avoided before any surgery.
Biological properties of GM Anti-oxidant properties Many scientists reported that GM and its isolated xanthone exhibited potent anti-oxidant properties [20][21][22][23]. In the cisplatin-induced nephrotoxic rats, the α-mangostin from GM exerted renoprotective effect by reducing cisplatin-induced renal oxidative/nitrosative stress [2]. Tjahjani et al. [20] evaluated the anti-oxidant activity of the different fraction of mangosteen rind, and found that the ethyl-acetate fraction exerted higher anti-oxidant than other fractions. The GM's pericarp extract also inhibited the formation of pentosidine and reduced advanced glycation end-products accumulation in the skin, which suppressed the glycation stress and improve skin conditions [21]. Xie et al. [22] found that drinking 245 mL mangosteen contained beverage by healthy populations enhanced the plasma anti-oxidant capacity (maximum 60 percent after one hour). A clinical study reported that oral administration of polar fraction extract from mangosteen pericarp to human subjects for 24-weeks enhanced the anti-oxidant activity without producing potential side-effect [23].

Health promotion effects of Garcinia mangostana
Anti-allergy and anti-inflammatory properties Anti-allergy and anti-inflammatory properties of GM have been proved by various scientific evidence. The GM's pericarp inhibited the histamine-and serotonin-induced isolated thoracic rabbit aorta contractions, and blocked the histamine and serotonin receptor, respectively [4]. In RBL-2H3 cell line, 40 percent ethanol extract of mangosteen was found to inhibit the histamine release and prostaglandin E2 (PGE2) synthesis activities [24]. The 18 hours treatment of γ-mangostin suppressed the continuous PGE2 release and cyclooxygenase-2 (COX-2) gene expression in the C6 rat glioma cells in concentration-dependent manners [5]. The extracted isogarcinol from GM exhibited anti-inflammatory activity by suppressing the CD4 T-cells regulation in the murine model [25]. The combination of Garcinia with cocoa, coffee and green tea significantly reduced the lipid content from serum and liver in a dose-dependent manner. It also improved homeostasis model by assessing insulin resistance index and pro-inflammatory cytokines (TNF-α, IL-6) [26].
Anti-diabetic properties Different fractions of the GM's pericarp and isolated xanthone showed an anti-diabetic effect by inhibiting α-amylase and α-glucosidase enzyme activities [42,43]. Ethanolic extract of GM showed the anti-diabetic effect against streptozotocin-induced diabetic rats by lowering high blood glucose, biochemical parameters, hepatic architectures and increased the HDL and total protein levels, compared to the control group [44]. While, in high-fat-diet-and streptozotocin-induced diabetic mice, mangosteen vinegar rind from GM, lowered the hyperglycemia, hyperlipidemia, oxidative stress marker, hepatic damage marker and improved the glycogen contents and anti-oxidant markers [45]. In similar model, isolated xanthone from GM demonstrated the nephroprotective effect by reducing the body weight, high glucose level, kidney hypertrophy, kidney damage marker and MDA level of plasma and kidney tissue [46].

Conclusion
Recently, the increasing awareness on the negative impacts of nutrition and diets, such as causing obesity, development of chronic diseases, cardiovascular disease, diabetes, cancer and other disease, was quite alarming as well as obvious especially in the western countries. A large number of bioactive compounds have already been identified in foods and drinks, contain high percentage of polyphenolic compounds (e.g. xanthone, lignans, phenolic acids, etc.).
The GM possesses potential biological properties without producing noticeable side-effect. It boosts up immune systems, reduced metabolic syndromes and its associated diseases, respectively. The GM's xanthone was found to associate with health beneficial effects by lowering oxidative stress, reducing inflammation, managing and controlling obesity and diabetes, inhibiting the growth of cancer cells as well as showing anti-allergic, anti-bacterial, anti-fungal and anti-malarial properties.
However, most of investigations on GM mentioned in this report were conducted either in cell-free, cell-type or animal model and none of them was on humans. Therefore, the further studies of health effects on human subjects are crucial and needed.

Introduction
Musculoskeletal disorders (MSDs) are disorders that occur in the muscle, tendon, ligament, bone, joint, intervertebral disc and skeleton of the whole body. In general, diagnosis of MSDs caused by work is accepted by an agreement in each country. In 2007, the Ministry of Labor in Thailand, classified MSDs as being within the group of occupational diseases. The Bureau of Policy and Strategy in Thailand's Ministry of Public Health reported an increase in the morbidity rate of work-related musculoskeletal disorders (WMSDs) from 121.93 per 100,000 persons per year in 2015 to 135.26 per 100,000 persons per year in 2016. Due to this rapid increase, there is an urgent need for innovation in ergonomic interventions to prevent an increasing health problem among the workforce [1].
WMSDs are one of the most common health problems among healthcare workers (HCWs). Previous studies indicated that HCWs, whose responsibilities include carrying, transferring or relocating patients, regular forward bending of the whole body and prolonged standing, were exposed to a high risk of WMSDs in the neck, lower back and knee regions [2,3]. Moreover, a study by Yasobant and Rajkumar [4] showed that HCWs working in a prolonged sitting, standing, awkward posture or cramped positions put employees at a major risk of WMSD. There are personal factors to WMSDs as well. Mirmohammadi et al. also reported that female HCWs were more prone to develop WMSDs than male HCWs and that the body mass index of the HCWs was also related to WMSDs [2]. Specifically, nurses had the highest prevalence of WMSDs compared to dentists and physical therapists, while laboratory technicians and physicians had a lower prevalence [5]. Also, Manmee et al. reported that Thai nurses had the highest prevalence rate of WMSDs when compared to support staff in hospitals [6]. Although there is research regarding the prevalence and risk factors of WMSDs in HCWs, there are no cross comparisons between each occupation in the healthcare sector.
Therefore, this paper reviews the existing literature on the prevalence, risk factors and prevention of WMSDs among HCWs to find the occupations that have the highest prevalence and related risk factors of WMSDs. The emphasis is on workstation redesign and innovative approaches to reduce or to prevent WMSDs among HCWs according to their job characteristics and postures [7][8][9][10][11][12][13][14][15][16][17].

Methods
Papers published between 1990 and 2017 were selected from respectable databases such as PubMed, Science Direct, Google Scholar and E-Thesis. Keywords used to search for these papers included: WMSDs, healthcare workers, physicians, dentists, physical therapists, pathologists, laboratory technicians, allied health professions, nurses, hospital workers, ergonomic design, workstation, guidelines and scientific tools such as electromyography (EMG), 3-D electromagnetic tracking system, lumbar motion monitor (LMM) and anthropometric measurements. Inclusion criteria included studies that showed prevalence and risk factors of WMSDs in each body part. In total, 11 papers about the prevalence and risk factors of WMSDs among HCWs were selected for review. In total, 12 additional papers were reviewed to explore the guidelines and the innovative interventions that can reduce or prevent the risk factors of WMSDs in each body part. Studies that report a result of the intervention suggested following a self-report questionnaire survey were excluded.

Results
There are many groups of HCWs in the hospital setting including physicians, dentists, nurses, physical therapists, laboratory technicians or allied health professions. Almost all HCWs had reported problems that are categorized under WMSDs. The common WMSDs among HCWs included the upper and lower back, neck, shoulder and hand/wrist. Risk factors were repetitiveness, awkward postures, working in the same position for long periods and bending/twisting of the back for physicians. Eye care physicians experienced a higher prevalence of WMSDs in all body parts compared to family medicine physicians [18]. Dentists were also found to work with prolonged periods of time sitting in inadequate ergonomic working environments. That is why dentists have the highest prevalence of overall WMSDs (92 percent) [19]. Additionally, physical therapists and allied health professions who work in laboratories work with prolonged use of medical devices such as microscopes and sonographs. WMSDs among physical therapists were mainly in the lower back (46.5 percent), shoulder (45.2 percent) and neck (44.9 percent) [20], while the overall WMSDs among other allied health professions such as microscope workers, pathologists and radiologic technologists were 62, 76 and 88.9 percent, respectively [21][22][23]. Among nurses, the overall prevalence of WMSDs was about 55.6-91.7 percent. When focusing on the body part, nurses have the highest prevalence in the shoulder (12.6-64.4 [3,24,25]. Table I shows the prevalence and risk factors of WMSDs among HCWs in each body part from the selected papers.
Regarding interventions, the redesign of workstations and guidelines were used to prevent risk of WMSDs. These interventions vary according to the HCW's working postures and job characteristics. From previous research, the scientific tools used to measure between, before and after intervention included: EMG, 3-D electromagnetic tracking system, LMM and anthropometric measurements. For the subjective tools, they used modified WMSD questionnaires, satisfaction forms and comfort scales. Results show that the interventions not only reduce muscle activity, but HCWs were also satisfied when they used the new workstation. For example, an ergonomic dentist chair with an arm rest and thoracic support was used to reduce WMSDs in the upper limb [7]. Microscope workstations with arm support also reduced WMSDs in allied health professions who work in laboratories when analyzed by EMG [9,10], while the self-report questionnaires showed satisfaction with the new workstation [10]. Since nurses show the highest prevalence of WMSDs when classified by body region, there were many interventions recommended. However, from previous research, interventions cannot prevent all risks of WMSDs in nurses [11]. Table II shows the workstation and guidelines that previous research recommended to use for the prevention of WMSDs among HCWs. Table I shows that WMSDs among HCWs occurred primarily in the upper body part including lower and upper back, neck and shoulder, while Table II shows the interventions and guidelines that help to prevent WMSDs among HCWs. It must be noted that all interventions and guidelines were designed by considering the job characteristics that include risk factors of WMSDs.

Discussion
A previous study [18] found that physicians were exposed to patients in the same working posture resulting in repetitive work, awkward positions and bending or twisting the back. Similarly, dentists, worked with repetitive head rotation, spine rotation and prolonged leg bending resulting in dentists complaining about WMSDs [26]. Furthermore, bad working habits and uncomfortable physical postures are the causes of MSDs, discomfort and fatigue among dentists [27]. Regarding these working postures, Haddad et al. [7] found that the ordinary dentist chair was a risk factor that caused WMSDs in the trapezius muscle among dentists so they designed an ergonomic dentist chair with an arm rest and thoracic support. The result showed that the ergonomic dentist chair reduced EMG activity in the trapezius.
Among laboratory technicians, Fritzsche et al. [21] found that the prevalence of WMSDs in pathologists who are microscope users was 76 percent, whilst pathologists with visual problems, mainly myopia, was at 90 percent . Jain and Shetty [22] indicated that 94 percent of microscope users in the laboratory reported some visual problems during microscope use, while WMSDs of microscope workers were reported in the neck, back, shoulder and wrist and hand regions, respectively. There are many interventions and guidelines to prevent WMSDs in laboratory technicians, i.e., the guidelines of Mitchell et al. [14], and the Centers for Disease Control and Prevention [15]. Also, the Occupational Safety and Health Administration guidelines for microscope workers suggested that the standardized microscopes should be ergonomically designed as shown in Table II [16].
There are many studies aimed at reducing the problems of WMSDs in laboratory-based microscope workers by redesigning new ergonomic workstations. For example, in 2002, Kofler et al. [9] designed a new microscope workstation by using a table fitted with unique adjustable slanting "wings," also allowing the forearms to be angled at 90 o at the elbow and to rest on the surface while operating the control knobs and sitting on the ergonomic chair

379
Work-related musculoskeletal disorders with support for the lower back. The result showed that the new design of an ergonomic workstation can reduce EMG activities in the microscope workers' muscles. Another example is the study of Sillanpää et al. [10] who designed a new ergonomic microscope table with forearm support for microscope workers. They found that the new ergonomic microscope table reduced EMG activity of muscles in the neck and shoulder, and that microscope workers in their study were satisfied to work with the new ergonomic microscope table. In 2010, Sillanpää and Nyberg [28] recommended the design of microscope workstations that should have forehead and support of the upper extremities and good sitting postures with chair adjustments and variation in posture. Helander et al. [17] also recommended that using ergonomic microscope workstations and introducing training programs can reduce visual problems and muscular fatigue in microscope workers.
Similarly, workers in the radiology department spent more time in front of a computer monitor. Factors including lighting, temperature and ventilation, special circumstances, noise, personal factors and training were found to be related to repetitive strain injuries. In 2004, Lama [23] reported that the radiology technologists reported a prevalence of musculoskeletal symptoms in the back, right hand/wrist and dominant hand/wrist, right shoulder and dominant shoulder. For guidelines and interventions to reduce WMSDS in workers in the radiology department, Goyal et al. [13] discussed the key features of an ergonomically designed workstation and reporting room in the radiology department that can be used as the guideline to design a good workstation because the working environment and working conditions were also found to be related to WMSDs. Specifically, regarding sonographers in the radiology department, Murphey and Milkowski [8] studied about an adjustment of sonographer scanning postures. They indicated that the EMG activities of the muscle in the shoulder and forearm were decreased when the sonographer changed their working posture to near natural position and used the support under the forearm when they used their instruments.
Among physical therapists who work with patients with physical problems, they were found to work with awkward postures in their physical therapy practice. Weerapong et al. [20] reported that the highest prevalence of WMSDs among Thai physical therapy students was in the lower back followed by shoulder, neck, upper back and wrist/hand. This is because Thai physical therapy students must work in prolonged seating positions during their physical therapy training. Bae and Min [29] reported that WMSDs among the physical therapists were mostly found in the shoulders followed by hand and back, neck, arm, hip and knee, respectively. Moreover, a previous prospective cohort study with one-year follow-up [30] reported that the incidence rate of WMSDs among physical therapists was 20.7 percent.
Similar to physical therapists, nurses have to take care of patients with physical disabilities and patients who cannot help themselves. Nurses in particular are required to take care of all patients in all units. Furthermore, previous studies [5,6] show that nurses were at the highest risk of developing WMSDs when compared with other healthcare workers and support workers. Tinubu et al. [3] indicated that the 12-month period and point prevalence rate of WMSDs in any part of the body among Nigerian nurses were 78 and 66.1 percent, while most WMSDs occurred in the lower back, neck and knees, respectively. Thai nurses also had a high prevalence rate of WMSDs because they were exposed to physical workloads, non-neutral working postures and a psychologically demanding workload. Sinsongsuk et al. [24] reported that the 12-month prevalence of WMSDs among Thai nurses was about 55.6 percent. Common areas of the body that were found to be related to WMSDs were lower back 33.4 percent, shoulders 19.8 percent and upper back 19.4 percent. In 2011, Jin et al. [25] reported that 92 percent of Thai nurses had WMSD symptoms in their body a 12-month period during which time 54 percent of them visited a physician or physical therapist, while common injury areas related to their work were lower back, shoulders and neck. The top 3 major hazards of their workload were nursing management, repositioning patient in bed and preparing and distributing medication, while 380 JHR 32,5 the top 3 major hazardous working postures were neck/shoulder in non-neutral posture, bending and twisting waist in awkward posture and prolonged standing.
Literature reviews [24,25,31] showed that lower back pain was a major cause of WMSDs among nurses. Based on three hazard categories developed by Nelson et al. [32], the author studied patient handling tasks for nurses. They focused on the ten high risk tasks in order to redesign new working postures and used this tool to collect data shown in Table II. The interventions proposed in their study included manual assistive devices, mechanical devices and administrative management. When they applied ten interventions to ten tasks, the results showed that the biomechanical and EMG data in all interventions were statistically different between the intervention and non-intervention groups in joint moments, forces and muscle activity [11]. However, three out of ten interventions cannot prevent all risk of WMSDs including, i.e., bathing patient in bed (under side), making an occupied bed and dressing a patient in bed. Therefore, future research should investigate the tasks of nurses that are the risk factors of developing WMSDs and develop innovative ergonomic workstation designs to effectively prevent WMSDs in nurses.
Weiner et al. [12] studied about repositioning a passive patient upwards in bed by choosing an assistive device including regular cotton sheet, sliding sheet and carrier and predicted the risk for low back disorder based on the LMM torso kinematic inputs. The result showed that assistive devices used by nursing personnel had a significant influence on the back, upper limbs, shoulders and neck loading, and the result from the focus group indicated that nurses were satisfied to use the sliding sheet because it can be kept under the patient for extended periods, while using a carrier required extracting it out after every repositioning of the patient.
From the above discussion, it can be seen that many guidelines and new ergonomic workstations were designed according to the position of work and job characteristics. When they compared the EMG activities between before and after intervention, EMG activities in after intervention were found less than before intervention. However, if the intervention by workstation redesign has some limitations, the guidelines or administrative controls will be used.

Conclusion
The most common interventions used in HCWs were introducing arm support and height adjustments. WMSDs among HCWs were found to vary based on the position of work and job characteristics. Therefore, in future research, workstation redesign should also be considered according to the position of work and job characteristics of HCWs. This literature review offers evidence that the ergonomically designed workstation is the recommended intervention that can reduce and prevent the risk factors of WMSDs among HCWs in terms of muscle activity, force and job satisfaction. This study provides directions for future research on developing ergonomic intervention or innovation to prevent the risk factors of WMSDs among HCWs.

Introduction
Heterosexual discourse based on patriarchal values has overwhelmed people throughout the world, including Thai society, pushing lesbian, gay, bisexual, transgender and queer (LGBTQ) people to become socially marginalized. Although some international mass media have recognized Thailand as an open place for LGBTQ people, there is still social oppression, such as no right to marriage, the lack of opportunity to be accepted as a worker in various careers, and being bullied both physically and psychologically at school and in the workplace because of their gender identity or sexuality [1,2]. As a result, some LGBTQ people develop psychiatric symptoms, such as depression, suicidal ideation or actually committing suicide, [3][4][5] as well as turning their life toward substances to deal with their psychological pain. Many studies revealed that LGBTQ people take more risks to use illicit drugs; in one study reported that homosexual males take more risks in terms of using amphetamines than heterosexual males [6], which was similar to one study showed that homosexual males and females had higher levels of non-medical drug usage and lower levels of abstinence than heterosexual males or females [7]. In addition, it is not only illegal substances, but also legal substances such as alcohol that LGBTQ people use to remedy psychological suffering. One study shows that homosexual females use alcohol more often than heterosexual females [8]. In conclusion, we can assume that LGBTQ people have more risks to use non-medical substance as well as more difficulty being abstinent from drugs in comparison to non-LGBTQ people. However, there is still not enough literature to describe the in-depth road of life among LGBTQ people and why they have to cope with adversity by using illegal drug or alcohol. This leads to the objective of this study that aims to explore the life narrative of Thai substance dependent LGBTQ people. The results of this study can be the guideline for counselors or clinicians to design an appropriate intervention for Thai LGBTQ clients.

Methods
This study employed a narrative research approach that prioritized the story of marginalized persons based on the constructivist paradigm [9]. The narrative inquiry often tells the story from the beginning to the turning point in the story that changes their direction in life.

Participants and setting
The research team conducted in-depth interviews with ten LGBTQ participants who attended the rehabilitative stage in a substance dependence treatment center in central Thailand. All of the participants were diagnosed as having a substance use disorder. The participants included individuals of various sexual orientations/genders, such as transgender females, tomboys, gays, lesbians, bisexuals and queer people, who identified themselves. The personal profile of the participants was listed in Table I. After being approved by the research ethics committee, the research team went to a substance dependence treatment center to build rapport with both the nurses and ex-addict staff members. The researcher asked them to introduce LGBTQ people who were in the process of substance dependence rehabilitation. The research team contacted the participants, explained the objective of the research, invited them to be interviewees and asked to sign the consent forms before collecting the data. Each participant was interviewed one or two times lasting from 30 to 120 min and used voice recordings. The interview guides were created from related theories and studies, but the researcher used them only in the initial interview phase and allowed the data to emerge mainly from participants by probing. The interview guide included the following questions: "Would you please tell me your story and how you got here?" And "What kind of social oppression have you faced in your lifetime and how this is related to your substance use."

Data analysis
Thematic analysis was used to analyze the data [10]. Both inductive and deductive methods were employed to generate the themes and categories [11]. First, the researcher transcribed the voice recording word for word. Then, the researcher read the transcripts about each participant several times to identify the common and emerging themes. Later, codes and categories were generated to support each theme. The results of this narrative research were written into stories in a chronological sequence [12]; beginning with their life before using substances, followed by significant turning points that resulted in substance use and later, their motivation to remain abstinent from drugs and alcohol.

Trustworthiness
Trustworthiness is a method to ensure that the results from a qualitative study are both reliable and valid [11]. The data and triangulation method were used to ensure credibility by collecting data from various sexual orientations/genders of participants such as gay males, transgender women, tomboys, bisexuals and lesbians, as well as using data from different methods, such as interviewing and observations to make sure of the veracity and credibility of the study. Furthermore, the researcher had several debriefing sessions between the researchers and her supervisor, an expert in qualitative research in data analysis process.
The results of data analysis were sent back to all of the participants to prove the credibility of the study before publication. All of the data would be destroyed after the one-year completion of the study by deleting voice recordings and Microsoft word files from both computers and drives.

Ethical considerations
The ethical considerations were approved by the Research Ethics Committee of the Drug Dependence Treatment Center (No. 021/2560 on June 2, 2017). The name of this Drug Dependence Treatment Center is deliberately not identified to ensure the confidentiality of the participants.

Results
The findings were divided into three themes. Each theme had several categories with quotations, their alias, their sexual orientation/gender and their age. These themes may overlap because the events occurred continuously, as follows.
Theme 1: life before using substance Theme 1: life before substance use had predisposing conditions that can cause participants to use substances in a later period in their life, which was divided into five categories. Category 1: family relationship problem. Most of the participants were raised in broken homes, characterized by the loss of parents through abandonment, divorce or death. Some participants were hurt by a lack of emotional support from their families. They often thought about their love-deprived childhoods and needed someone for emotional support or act as a positive role model in order to deal with adversity: My mother has never listened to me or taken care of me. I had to live with my grandmother instead. She never visited me on Mother's Day, so I had to lie in the empty chair while my friends came with their parents or asked a neighbor to come with me that day. (Poy told her story in tears, lesbian, aged 36) My parents were separated since I was a child. I lived with my mother and stepfather. I was bullied many times by relatives around my home, while my mother worked all day to earn money. I often said to my sister that if we had a father to protect us, we would not be bullied like this. (Ef, bisexual, aged 31) 389 Life narrative to substance use