Work Ability of Thai Ageing Workers in Southern Thailand: A Comparison of Formal and Informal Sectors

Thai society has been moving towards an ageing society. The independent elderly persons needed to continue decent work after retirement. This study analyses the association of work ability, individual and work-related factors with Work Ability Index (WAI) of the formal and informal sectors in a southern Thailand community. This cross-sectional study with multi-stage sampling employs in 324 Thai elderly workers, aged between 45 and 70, working in Nakhon Si Thammarat province. Data collection of socio-demographic status, health history, work-related factors questionnaires were done with anthropometric data and WAI instrument between March and September 2019. The descriptive and logistic regression analysis were used for the association analysis.


Background
Thai society has been moving towards an ageing society. The number of elderly aged 60 years and over has seen a slow but sure increase. In 2019, the number of older people reached 17.5 percent of the nations' population, which is considered to be part of 'ageing society', as de ned by the United Nations (UN). The Thai society will become a complete ageing society when its proportion increases to approximately 20 percent in the year 2022 [1]. It was found that the number of elderly workers has grown every year-from 3.10 million people in 2010 to 4.70 million in 2020 [2]. Thailand Elderly Plan No.2 (2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021) supported that 67 percent of the independent elderly persons needed to continue decent work after retirement [3].
Moreover, formal workers who shift their work after retirement to informal sectors such as agriculture, or are self-employed or home-based workers, increased according to the increasing proportion of older people. The evidence suggests that older people will remain in the workforce for a more extended period in response to several social and economic drivers [4][5]. The vast majority of Japanese workers aged 60 and over remain in the labour force to maintain their standard of living, and for economic reasons. Five factors contribute to the different labour force participation rates of older workers in the United States and Japan: 1) economic need; 2) type of employment; 3) cultural values; 4) policy factors; and 5) health [6]. In Thailand, the informal workers also continue to work in their old age until physical deterioration occurs and their work ability declines.
As de ned by the World Health Organization (WHO) or the European countries, the term 'older worker' refers to workers aged 45 years and over [7]. The International Labor Organization (ILO) de nes the nature of workers who are beginning to have di culty in their employment and/or occupation due to their increasing age [8]. Thailand has identi ed elderly workers as those who are 60 years of age and above, who are still working and able to perform their duties [2]. However, if based on the deterioration of the body or physiological changes affecting the working potential, it is estimated that people enter the category of 'older workers' at the age of 45 years old.
Naturally, physical tness and health of a population deteriorates according to increasing age-the muscles have less mass, the joints are affected and the body's mobility goes down. However, elders who exercised daily and properly in their younger years, and still continue to, can still be energetic and perform well; their bodies are tter and mind is alert. Thus, having an active lifestyle and including physical exercise in daily life will increase the potential and opportunities to work e ciently. However, ageing worker's productivity depends on their health and health-related works. The work and ageing conditions are two main factors that threaten their health and work ability-such as disease history, behavioural health, work hazards and health care access-which may cause a decrease in working opportunities, productivity and income stability [9].
Thailand has an integrated Long Term Care (LTC) system to community primary health care. The LTC is the public health and social services that operate to meet dependent people's needs due to chronic illnesses, various disability accidents, and the elderly who cannot help themselves in daily life [10]. However, it mainly focuses on the curative and rehabilitation activities that do not include the community's ageing workers' occupational health and safety services. The increasing trend of the informal ageing workers in the community is then a signi cant issue for worker's health, which needs Page 4/21 response from community health service. Health needs of ageing workers comprise physical function and tness, and working ability that will be utilized to perform any job they are engaged with. Work Ability Index (WAI) is an instrument used in clinical occupational health and research to assess working ability during health examinations and workplace surveys. The index comprises demands of the work and the worker's health status and resources, which is an appropriate tool to apply for the ageing worker's health need assessment [11]. Its assessment domain could signi cantly assist the community occupational health and safety operation to support ageing worker's holistic health.
A number of studies have been conducted among different populations to reveal the relation between individual and work-related factors with WAI, but it has yet to be used among formal and informal ageing workers in Thailand [12][13][14]. The difference of socio-economic and work-related factors, and the health status and WAI have not been studied yet. Moreover, aging worker's health needs will be critical determinants, which need to analyzed and quanti ed by WAI so as to express the readiness of work performance of the ageing workers in both formal and informal sectors. This study explores health needs of ageing workers using working ability dimensions between the formal and informal sectors in a southern Thailand community. We conducted the Nakhon Si Thammarat province survey, which had the highest population among Thailand southern provinces. The study's bene ts are for primary healthcare arrangement to provide and meet the needs in occupational health and safety areas for the active ageing workers.

Study Design and Setting
This study employed a cross-sectional design based on quantitative approaches. We conducted surveys in rural and sub-urban areas of the two districts located in Nakhon Si Thammarat province, which had the highest population of ageing workers among the southern provinces of Thailand. Participants consisted of Thai elderly workers in formal (public and private) and informal sectors, aged between 45 and 70, working in Nakhon Si Thammarat province for at least two years.

Sample Size Determination and Sampling
The data was obtained from the community health survey of each sub-district area. The study population was 112,117, which was calculated by the nite population formula with p 0.28 (proportion of people aged between 45 and 70 years, of population in Nakhon Si Thammarat Province), and the adjusted extra 5 per cent then yielded the minimum sample size of 324. Community areas and types of occupation employed multistage sampling proportional to size selection. First, the study area was selected; two districts with the highest elderly population were chosen. Second, strati ed random sampling by dividing the elderly workers into two strata of formal and informal sectors was carried out. We considered the topography from the sub-districts and villages accordingly, and then allocated the sub-strata's proportional sample size in each stratum. Finally, we used simple random sampling to create organizational name lists for formal workers and community health centre registries for informal workers.

Data Collection and Measurements
Data was collected through personal, face-to-face interviews by four trained data collectors between March and September 2019. Data quality was controlled in the eld by supervisors from the School of Public Health, Walailak University, and investigators.

Socio-demographic Status, Health History and Work-related Factors Questionnaire
To evaluate study factors, we used a separate questionnaire designed by the researchers. Sociodemographic factors included worker sectors, gender, age, marital status and educational level; healthrelated factors included non-communicable chronic diseases (NCDs), smoking status, alcohol consumption and regular exercise activity. Work-related factors in this study was divided into two categories: physical work environment and psychosocial work-related factors. In addition, we collected data on working practices, occupational health services experience (health education, occupational health risk and working process assessment, and primary diagnosis), health promotion service experience and utilization of primary care (health education and counselling, primary prevention program and NCDs screening). Two measurements assessed the quality of the overall questionnaire. The index of item objective congruence (IOC) yielded a value of 0.9-1.0 for content validity, and the Cronbach's alpha coe cient was 0.86 for reliability.

Measurements
The anthropometric data were measured weight, height and waist circumference (WC) using a standardized digital scale (TANITA UM-070, TANITA Corporation, Japan) and a standard measuring tape, to the nearest 0.1 kg and 0.1 cm, respectively. To record body mass, the participants wore light clothing without shoes. WC was taken on bare skin, in a horizontal plane midway between the inferior margin of the last rib and iliac crest [15]. Obesity was determined through calculation of body mass index (BMI) and WC for Asian; value of BMI > = 23 kg/m 2 was used to de ne overweight and > = 25 kg/m 2 was used to de ne obese and WC levels of > 90 and > 80 cm for men's and women's abdominal obesity, respectively [15][16][17].

Work Ability Index (WAI)
The WAI is an instrument developed by the Finnish Institute of Occupational Health Research. This index is aimed at assessing work ability during health examinations and workplace surveys, and avoiding early retirement and work-related disability [18]. The WAI is calculated by summing up the seven items score (range 7-49). Finally, work ability is classi ed into one of the four categories of poor , moderate (28-36), good (37-43) and excellent (44-49) [11]. We use WAI questionnaire version translated in Thai by Kaewboonchoo O, 2015 for the assessment [19]. However, to analyze the impact of independent variables on different domains of WAI, we combined seven items of WAI into three domains according to the purpose of WAI, and other research has used these classi cation [11,[20][21] which took into consideration the demand of work, as well as the : 1) perception of work ability-including item 1, 2 and 6); 2) workers' health status-including item 3, 4 and 5); and mental resources-including item 7.

Study Variables
In this article, the dependent variables were WAI and each dimension of work ability: perception of work ability, health status, mental resources. The explanatory variables included socio-demographic factors, health factors and work-related factors.

Statistical Method
Both descriptive and inferential statistics were carried out using R 3.2.1 for Windows (CRAN, 2016). The proportions of the variable of interest by socio-demographic, health factors and work sectors were calculated. Independent t-test and Pearson's chi-squared were used to examine the different dimensions of work ability, total work ability index score, and the domain according to the purpose of WAI among workers sectors. We categorized WAI into two levels: WAI score lower or equal to 36 is a 'poor and moderate' and above 36 is a 'good and excellent'. Pearson's chi-squared were used for comparing the rates. The potential variables were selected with a p-value less than or equal to 0.2 from bivariate analysis to perform the nal model. Simple logistic regression modelling was performed to detect the association between socio-demographic, health and work-related factors with work ability index, and each dimension in a multivariate analysis. Statistical signi cance level was set at 0.05. Table 1 describes the distribution of individual health status and behaviour, and occupational health variables within this sample. Participants (n = 324) were predominantly general labour (23.8%), female (70.7%), age group 60-70 (37.4%) with average aged 56.2 (S.D. = 7.4), married (74.7%) and primary school (60.8%). Nearly half of them were NCDs patient (48.2%). Smoking and alcohol consumption was more than 70% and 'did not exercise regularly' was 59 %. Current working duration of fewer than 10 years (42.6%) and income was less than 5,000 Thai-Baht (38.9%). Obesity was detected at 63.6% by waist circumference as overweight (65.4%) when detected by BMI. The work environment exposures were an equal proportion (50%) for mild and moderate physical environment groups, and nearly 60%of them fell in the category of good occupational behaviour. Most of the participants had received occupational health services sometimes (51.9%), and received the same frequently for health promotion accessibility (78.1%). Working Ability of Ageing Workers: Dimension and Class Table 2 shows participant's total WAI stood on average at 40.6 (S.D. = 4.6) and 37.5 (S.D. = 5.0) for formal and informal sectors. The ageing workers in formal sector tend to have a higher WAI score than informal workers in every dimension. Independence t-test showed a statistically signi cant difference in the WAI score for each dimension between the formal and informal sectors (p < 0.05). We used the Pearson's chi-square examined work ability classes into two categories; poor to moderate and good to excellent WAI. There was a rate difference of WAI classes between formal and informal workers (p = 0.002). Most of the workers had a high proportion of good to excellent WAI, which are 81.7% and 62.5% for formal and informal workers. We explored WAI of worker sectors according to three purpose categories. The analysis found rate differences between WAI and workers group in the perception and working ability as well as health status categories (p < 0.001). The formal workers had the highest proportion of strong perception of working ability (64.8%) while the informal worker had the highest proportion of bad health status (65.6%). Moreover, the formal worker had higher good health than bad health (56.3% vs 43.7%), which contrasted with the informal worker group (34.4% vs 65.6%). However, there was no difference of WAI for mental resources in both worker groups.  Table 3 shows the bivariable analyses of the rate presenting a so-called 'poor to moderate' and 'good to excellent' WAI categories based on individual and occupational health factors. The analysis found that statistically signi cant association of individual factors and WAI categories are workers sector, age group, presence of NCDs, WC, and overall occupational behaviours (p < 0.05). However, several other factors did not have an association with WAI: residential area, gender, marital status, smoking status, alcohol consumption, exercise activity, BMI, overall occupational health services, overall health promotion services, and physical and psychosocial work-related factors.

Discussion
This study has been carried out mainly to explore the difference of WAI between formal and informal sector workers, the relations of socio-demographic, health and work-related factors, and WAI and its dimension. The results show that individual age group, presence of NCDs, occupational health behaviours are signi cantly associated with the WAI and WAI dimensions; perception of working ability, health status and mental resources of the study sample.
Age group and work ability. This study categorized worker's age into two groups; less than 55 and 55 years and above. The effects of age group associated to the WAI and similar to the three dimensions. The workers of age 55 years and above have a higher risk to 'poor to moderate WAI' class. Age was a signi cant factor associated to WAI in many studies. Our results' direction of association is consistent with many studies showing that age is signi cantly and negatively associated with WAI in various occupational sectors [22][23][24][25]. However, the data did not highlight gender and working ability, corresponded to the study of van den Berg et al. in 2009 [26], although there is the probability of low or poor work ability, which was higher in women's WAI measurement [27]. The ageing of workers requires two issues that need consideration [28][29]: the rst one involves worker's health, and the second is job productivity and performance. A larger number of older workers implies, for example, an increasing number of people at work with minor and major health problems that occur more frequently after 55 years of age [30]. About 53% of both sectors in the study were ageing workers, and half were NCDs. Age and the presence of NCDs were the predominant factors of ageing workers' productivity. The results show that the presence of NCDs is a strong and positive association among the 'poor to moderate WAI', especially in a 'health status' dimension (OR = 6.42). Our data explored the presence of NCDs (48.2%), with hypertension in the highest proportion. Complex interactions were proposed between ageing of working population and lifestyle risk factors such as low-level of physical activity, known as the risks of cardiovascular disease, and work-related risk factors [31]. Our data showed approximately 60%of workers had overweight or obesity status, in line with other reports [18,26,[32][33][34]. However, the rate may not be different between the two sectors. Moreover, more than half of the workers (59%) reported a non-regular exercise group. Physical health, which is one of the intermediate determinants of individual lifestyle, directly affects worker's health status and functional activity, which are base factors for working ability [35][36]. Then the declining of physical capability and increasing of risk to NCDs; obesity was the negative factor in uencing working ability of the ageing workers. Other risk factors, including smoking and alcohol consumption, are not associated with WAI. We found the high proportion (more than 70%) of 'non-smoker' and 'non-drinking' workers in this study, especially in the informal sector. This association is not signi cant in bivariable and multivariable models, this result corresponds to Mehdi El Fassi et.al.
(2013) [27] stating that workers' smoking habit was reported as signi cant in a single study only [32].
Working conditions correspond to the fourth oor and consist of work, and all of its dimensions as described by WAI [35][36] In Thailand, informal worker's occupational health services were integrated in the primary healthcare system arranged by the Universal Health Coverage scheme (UCS). The formal workers in the public and private sectors utilize healthcare services by their health insurance scheme, which is Civil Servant Bene t Scheme (CSMBS) and Social Security Scheme (SSS), respectively. Those services promote health, are prevention and cure-based when it comes to general health and diseases, which cannot be separated from occupational problems and diseases, except in the SSS. However, worker's health improvement issue is more complex and requires a holistic approach, especially in the ageing worker. The services have to set in at the early phase of physical and mental deterioration described by the decline of the WAI score. Work ability model is appropriate for an ageing workforce process based on the self-assessment of subjective experiences of personal resources, working context, and work-life interface [35][36]. The structure of work ability changes during a person's life and career, such as the fact that ageing affects the individual's resources [37]. The implementation of work ability assessment in the occupational health and safety program will provide preventive measures and early rehabilitation in the workplace and healthcare centre [38][39][40].
There are several limitations in this study. Firstly, the cross-sectional design by which exposure and outcome were measured concurrently does not certify the causal relationship model. In epidemiological studies, an association of exposure and outcomes is causal only if the study's plausibility was explored.
Secondly, self-reported measurements of the study's variables-including working ability and occupational hazards-may lead to recall and information bias. Workers' awareness of the occupational hazards could affect the measures' correctness. Workers in the formal sector were familiar with their work conditions, and they were able to recognize the occupational hazards in their workplace, while informal workers may not. Lastly, we focused on the working abilities of ageing workers residing in sub-urban and rural communities because of the information access, which may not be representative of workers elsewhere in different settings. Generalization beyond the study population should be used with consideration.

Conclusions And Recommendations
In conclusion, this study determined the association between working ability and the determinants, health-related factors, socio-demographic and work environment of the formal and informal ageing workers. The results presented that age group, presence of NCDs and safety practices associated with work ability of the Thai ageing worker. There is a strong association of between NCDs and health status of the workers. It suggested that promoting healthy behaviour and work environment may be an important strategy to improve ageing worker's work ability and health. The formal worker's organizational health strategy might consider providing the concept of total worker health at the early ageing stage.
Correspondingly, the primary healthcare centre may integrate the concept of worker's holistic health for health promotion and prevention strategy to work well for the informal workers. In addition, the ageing health services should cooperate with occupational health programmes in order to improve work ability and productivity of ageing workers for life-long working. For further study, longitudinal studies in various occupational settings of the ageing worker are required to determine the association better.
Abbreviations BMI Body mass index; NCDs:Non-communicable diseases; OR:Odds ratio; WAI:Work ability index; WC:Waist circumference Declarations