Work stress, health and wellbeing: evidence from the older adults labor market in India

Older adults are the most vulnerable group in the labor market, given their physical and functional health risks. Moreover, they are also at a greater risk of psychological factors like work stress due to risk of age, ageism and social well-being. Given these challenges and the growing proportion of older adults in the developing world, this paper studies the theoretical connotation between work stress and health related wellbeing outcomes among working elderly in India. Using BKPAI survey this study adopts multiple modeling approach (descriptive analysis, one-way analysis of variance, and a logit regression model) to test the association between work stress and wellbeing and understand this theoretical connotation in elderly labor market. Our results confirm that work stress is found to be significantly associated with socio-economic and health risks like income, gender, education, morbidity, disability, and socio-psychological barriers among working older adults. Females are likely to have more risk for work stress [OR, 1.72; (95% CI 2.89–2.91)] than males. Similarly, more affluent groups (wealthiest) are less likely to have any work stress [OR, 0.38; (95% CI-0.25–0.58)]. Older adults with multi-morbidity are at greater odds of having work stress [OR 2.26; (95% CI-1.74–2.94)]. Work stress is a common issue that enhances at higher ages due to the work, health, and socio-economic vulnerability of older adults. The study finds a significant impact of these factors on work stress and therefore argues for a comprehensive policy approach to address the psychological health risk of the working older adults in the labor market.


Introduction
W ork stress is an important labor market challenge due to poor health and work conditions of workers (Brown, 1980;Peterson, 2018). Work stress considerably affects their well-being, productivity, and work outcomes (Kompier and Cooper, 1999;Park, 2007). Research has identified multiple factors that enhance work stress, such as job strain, overworking, long working hours, place of work, and income received as wages (Calnan et al., 2004;Costa et al., 2006;Ferrucci et al., 2016;Leigh and De Vogli, 2016;Strazdins et al., 2004). While these factors are mainly contextual, individual factors also augment work stress (Houtman and Kompier, 1995;Shirom, 2003). These factors include workers socio-economic, demographic, and health conditions, which impact their work functioning and productivity (Härmä, 2006;Houtman and Kompier, 1995;Strazdins et al., 2004). Work stress also affects their work opportunities, making them vulnerable to jobs and livelihood earnings (Kivimäki et al., 2002;Pugliesi, 1999). Although there is clear evidence of how work stress is affected by the socio-economic and health outcomes among the workers (Fotinatos-Ventouratos and Cooper, 2005;Marcenes and Sheiham, 1992;Melchior et al., 2007), very little has been researched related to the older workers particularly in Indian context (Götz et al., 2018).
Work stress can act as a medium to impact workers productivity and work opportunities, particularly at older ages given the complex relationship between ageing, health, and well-being. Similarly, it may also be affected by the socio-economic and wellbeing conditions of the older adults like education, income and living environment. Few studies have also examined work stress from dynamic context given its casual impact. Thus, understanding work stress in ageing context is a critical issue that needs to be examined. So far studies have examined multiple factors ranging from work productivity to working conditions, however limited studies have examined this phenomena in labor market while exploring the health and wellbeing factors particularly of older adults. Therefore, as a first, we propose to understand and analyze the dynamics of work stress and well-being among the older adults in India. This study apart from its novel academic contribution is expected to deliver important policy outcomes emerging from the perspective of wellbeing in labor market of older adults. The same can then be integrated into a framework for of future research in order to understand the vulnerability of older adults given their increasing proportion in India.
Therefore present study will first formulate the theoretical connotation and then analyses the data from building knowledge based on population ageing in India (BKPAI) survey of older adults in seven Indian states to study examine the above research hypothesis in Indian settings.

Work stress and ageing
Psychological factors like work stress can have a significant impact on the wellbeing of populations at risk. Work stress is enhanced through demand for work and other factors at work place (Cavanaugh et al., 2000;McClenahan et al., 2007). However, research has also connoted the impact of age-related factors and the socio-economic well-being of working adults (Nübling et al., 2010;Vinstrup et al., 2021). Whether working adults face similar challenges at older ages despite having better work experience and enhanced productivity is a critical connotation in the labor market research among older workers (Kompier and Cooper, 1999;Wilke et al., 1985).
Work stress measurement is measured through multiple scales to examine the psychological wellbeing of workers, however studies have also tried to capture it through subjective measures where workers are asked to reveal their psychological state during work. Although a multiple set of dimensions can capture it in detail, but there are individual and organizational outcomes that may likely enhance the work stress among the workers in the labor market.
Work stress can likely enhance at the older ages given the vulnerability of older adults to various health and wellbeing risks. There is a significant risk for age-related challenges through loss of physical and functional health among the working older adults (Partridge et al., 2018;Wildenbos et al., 2018a). Older people at higher ages also face numerous other challenges like morbidity, disability, and socio-psychological barriers affecting their worklife (Frasca et al., 2017;Gulia and Kumar, 2018). Studies so far have examined the impact of work stress in terms of work outcomes like (productivity, opportunity, and work performance) (Adnan Bataineh, 2019;Burman and Goswami, 2018;Ehsan and Ali, 2019;Foy et al., 2019). But there is also a clear implication of individual level factors that can enhance work stress and result in the vicious circle of poor health and wellbeing particularly at older ages (Hsu, 2009;Mauno et al., 2013). Research have already identified these factors that increase the risk for well-being among older adults, resulting in tremendous work stress among them (Blaug et al., 2007;Rauschenbach et al., 2012). Therefore examining work stress is an important research issue that needs to be addressed among the older adults particularly active in the labor market, Theoretical connotation. Work stress is affected by chronic diseases conditions and self-related health among the older adults at large (Salvagioni et al., 2017). Work stress also increases due to work demands at the workplace. Studies find that family conflicts and societal relations also impact the levels of work stress (Nigatu and Wang, 2018;Sheikh et al., 2018). Researchers also relate it with the work environment and poor work conditions; however, socio-psychological and well-being measures of older workers also contribute to their higher work stress levels (Katić and Nešić, 2020;Sinha et al., 2021;Zhang et al., 2019). Moreover older adults are at greater risk for these factors owing to age-related discrimination they face at upper ages (Burnes et al., 2019).
Ageing is emerging as a challenge, and its implications in labor market are inevitable. While there is a clear focus on active ageing to reduce the ageing burden. Older workers are still vulnerable to physical and psychological factors that impact their well-being and work conditions (Katić and Nešić, 2020). Likewise, these older workers suffer from already prevailing health conditions, which are also critical in affecting their work productivity and work opportunities (Bakker and Demerouti, 2018;Neumark et al., 2019). Moreover it is also a factor that likely enhances their risk for better health and well-being. Theories in this regard have connoted that work stress is due to a combination of demands and controls over the job situation. Therefore, an imbalance between the two results in work stress (Karasek, 1979;Karasek et al., 1981). Similarly, early research also emphasizes on the socio-economic gradient and poor health that impact their wellbeing (Andersen et al., 2004). There is already a clear emphasis on better socio-economic status (SES) that can mediate the work stress among the workers, especially at older ages.
The canonical relationship between work stress and contextual factors have been addressed in literature at large. There has been also emphasis on the well-being factors as mediators in controlling work stress. However, what has been dearth so far in this is the association between work stress and ageing. Despite growing ageing challenges in the global south, very little has been explored so far in Indian settings especially in labor market. Therefore with rapid increase in ageing proportion, work stress can impact the labor market composition, given the vulnerability of older adults being largely uncovered and mostly in the informal sector (Bonnet et al., 2019). Similarly there is a greater risk of burden in terms of health and well-being due to lack of social security coverage and health care benefits (Kühn et al., 2018;Samorodov, 1999). Furthermore, active and healthy ageing is already focusing on this aspect, where importance is given on the long working years to avert the ageing burden (Boudiny, 2013;Jacobsen, 2017). Therefore, why this issue must be addressed in terms of the ageing labor market is because: (a) a significant pace is witnessed in the elderly population, especially in a developing country like India. (b) a large account of the labor market in these countries is primarily informal and uncovered by ageing-related work benefits and incentives; and finally (c) ageing is vulnerable to multiple factors that not only involve the physical and functional health, but mental health is also an essential aspect in this context. Therefore, considering these key questions, this study examines the association between work stress and wellbeing among the working older adults aged 60 and above in India.

Methods
Data. Present study used the data from United Nations population fund for ageing (UNFPA) survey 'Building Knowledge-based on Population ageing in India (BKPAI). The BKPAI Survey was conducted in 2011 in seven major demographically advanced states of India. A detailed report on the survey and its sampling procedure is given in the survey report. This survey provides detailed information on socio-economic, physical and mental health, living arrangements, work employment conditions and various other characteristics of older adults. All the survey is not the latest in the settings, but given its coverage of broad range of issues and also the data it has covered regarding the elderly labor market. Moreover the detailed information is given in the Appendix A of the supplementary file in the paper.
Response variable. Work stress is a key outcome to measure the impact of health and wellbeing at work. Although it is a multidimensional concept which involves not only individuals socioeconomic and health conditions, but work stress can be also related to work place environment and work opportunities. However all these aspects were not possible to be included in this study due to data constraints. Therefore we included a proxy variable for work stress where work stress was captured based on the following question asked to the older adults working in last one year. Do/did you feel any physical or mental strain due to this work? The response to this question was recorded in binary with 1 "yes" and 0 "no." Therefore we included this variable as a proxy measure for work stress to examine its linkages with various socio-economic and wellbeing measures. Although the concept of work stress is multidimensional however given the limitation of the data to cover these issues in the latest data we could include these multidimensional aspects in this study.
Independent variables. A set of independent variables were included, comprising socio-economic and health outcomes. We use a group of socio-demographic and economic variables like age, income, place of residence, region and education of the respondents. We also included a set of health outcomes to measure the proxy for health wellbeing. A detailed description of the health outcomes is given in the Appendix B of the supplementary file of this paper.
Statistical modeling Bivariate analysis. Prevalence of work stress is computed among the working older adults by carrying out the bivariate analysis.
Similarly, differences in work stress prevalence was also captured by socio-economic and health outcomes. The results were calculated by χ 2 test with a significance level of P < 0.05. We also provided the descriptive status of sample characteristics by applying the sampling weights in our study.
Multivariate analysis. We estimated the logit model to study the association between the response variable and several predictors. Since the response variable was binary, the response variable in our model follows the binomial distribution function with the probability function given as follows: yi is the response variable with binary outcomes of 0 and 1. π(xi) is the probability occurrence of the i for yi = 1, meaning those having any kind of work stress, and 1−π(xi) is the failure or probability occurrence of the i for yi = 0, depicting the state where older workers do not have any work-related stress. The final logistic model is computed as follows: N in the model represents the number of covariates. Finally, the above model is transformed to obtain the odds ratios using logistic transformation.
The estimation of β parameter in logistic regression is computed by using maximum likelihood estimation (MLE).
Model set up. We used the binary categorical variable as a dependent response variable to examine the work stress among older working adults. 0 represents the persons who do not have any work stress, whereas 1 denotes work-related stress. Independent variables are a set of socio-economic and health outcome variables, assuming that the older adults are at greater risk for health-related outcomes with the increase in age.
Ethical approval and informed consent. Ethical approval for this study, involving secondary data analyses, has been obtained from the Ethics Committee in the University of Southampton. Informed written consent was obtained from each participant, who were ensured that data would remain confidential and used for research purposes only. All methods were carried out in accordance with relevant guidelines and regulations.

Results
Work conditions contrasts across the states in India, therefore it is likely that work stress will vary across these regions. Figure 1 provides the prevalence of work stress among older adults aged 60 and above in India. Work stress is highest in Maharashtra with 83.86%. Similarly, work stress is lowest in Himachal Pradesh with 28.59%. Work stress prevalence in West Bengal is around 60%, followed by 52.42% in Orissa whereas it is 44.29% in the state of Tamil Nadu. Table 1 presents the prevalence of work stress among older adults aged 60 and above in percentages and the sample description of the overall sample by background characteristics and risk factors. Of the total number of respondents, 48.8% of males are having any kind of work stress compared to 66.8% females. As against 53.5% of respondents in rural area having any work stress, the corresponding percentage in urban area is 51.8%. Older adults with higher education levels have a lower prevalence of work stress (35.8%) versus illiterates (57.2%). Similarly, the prevalence rate of work stress is higher among widows and separated older adults than married ones (51.2%). Our results found that work stress prevalence is higher among the lower affluent groups (Poorest; 64.1%; P < 0.05), whereas it is lower among the better affluent groups (Richest; 33.2%; P < 0.05). Table 1 also provides the work stress prevalence in percentages among the elderly by various health outcomes. Work stress percentage is highest among the older adults having more than one morbidity (Multimorbidity; 61.2%; P < 0.05). Similarly it is also highest among older adults with poor self-related health, (56.9%; P < 0.05). Among older adults with having any disability (56.3%; P < 0.05) and (57.6%; P < 0.05) among older adults who are not socially active at a significance level of 5 percent. Similarly, other results were coherent with a greater prevalence of work stress for being at any risk of wellbeing outcomes. Table 2 shows the associated risk factors of work stress among older adults aged 60 and above in India. The results from the table reveal that females are more likely to have a greater risk for work stress [OR, 1.33; (95% CI 0.97-1.82)] as compared to males. Similarly, more affluent groups (wealthiest) are less likely to have any work stress [OR, 0.54; (95% CI-0.34-0.86)]. We also examine the risk factors related to health and wellbeing of older adults. Older adults with multi-morbidity are at greater odds of having work stress [OR, 1.74; (95% CI-1.30-2.32)]. Similarly, poor selfrelated health is positively associated with work stress, and results were significant at one percent level of significance [OR, 1.47; (95% CI-1.18-1.83)]. We also found a significant and positive association of the work stress with wellbeing outcomes like poor social wellbeing [OR, 1.52; (95% CI-1.16-1.92)] and psychological well-being [OR, 1.72; (95% CI-1.36-2.16)] with significant favorable odds, respectively. The Table 2 results also shows the other well-being factors like disability, cognition, social activity, which are significant and positively associated with the likelihood of having any work stress among the working older adults in India, as shown in Table 2.

Discussion
Work stress is an emerging challenge in the elderly labor market given the vulnerability of older adults which involve risk for poor health and well-being at later stages of life. With increasing age, there is a greater likelihood of older workers to be at risk of work stress due to poor physical and psychological health (Sinha et al., 2021). This study therefore tries to understand the work stress and its associated risk factors among the older adults in India aged 60 and above.
The results from our study confirm that age-related socioeconomic and health challenges enhance the risk of work stress among older adults in India. Where socio-economic outcomes likely impact the well-being of the elderly and their work outcomes (Grundy and Sloggett, 2003), the older adults active in the labor market with health-related challenges are also more vulnerable to socio-psychological health conditions (Sinervo, 2000). Thus socio-demographic and well-being factors significantly impact the psychological health like work stress among the working older adults, apart from the work-related outcomes like work productivity, job conditions, and wages as found earlier studies (Alam et al., 2015).
Factors, such as gender and income were significantly related to work stress, with women being at greater risk. Similarly, working older adults from low income level households are at greater risk for work stress, as shown by our results. This clearly reflects the ample association between socio-economic characteristics and work stress among older adults in India. In addition, we also observe a significant statistical association between other sociodemographic covariates with work stress. The results corroborate with earlier studies in this context, where there is a clear association of work stress and workers socio-economic conditions, especially at older ages (Duxbury and Higgins, 1991;Fotinatos-Ventouratos and Cooper, 2005;Matud, 2004).
We also studied the work stress scenario across various states in India. Our results clearly showed the variation in work stress across these states with Maharashtra having the highest prevalence of work stress among the older workers and Himachal Pradesh with the lowest prevalence. This is likely due to the nature of economies in these two states. While economy in Maharashtra is mainly dominated by industrial sectors, with greater concentration of urban population, In Himachal Pradesh, the economy is mainly agricultural and concentrated in rural areas. Moreover there is also the greater proportion of older adults falling in the working population in Maharashtra which might be the likely factor for greater proportion of work stress among the older adults in the labor market. Similarly the working   Authors own computation from BKPAI survey of India P < 0.001 significant at 1%, and P < 0.005 **significant at 5%. condition in these states are entirely different with Maharashtra mainly comprised of industrial workforce with large working hours and poor work conditions, however working conditions in Himachal Pradesh are mainly concentrated in small and medium scale manufacturing units which provide relatively better work and employment oppourtunities. Health and ageing-related challenges are interrelated in many ways in the labor market (Ucieklak-Jeż and Bem, 2021). While poor health can reduce the labor market participation and income of older workers (Harris et al., 2021). Poor employment conditions can also affect their physical and psychological health (García-Gómez et al., 2010). Similarly, poor health likely impacts their work conditions and functioning in the labor market (García-Gómez et al., 2010;Schuring et al., 2007). Therefore examining the various health outcomes like self-related health, multi-morbidity and other physical health conditions of the older adults to understand their impact on work stress was one of the prime objectives of this study. The findings are clearly in line with the earlier research that older workers with any morbidity or poor self-related health are likely to have a greater risk for work stress at their upper ages (Ganster and Rosen, 2013;Sparks et al., 2001). The results argue for better health among older adults owing to its significance in work-related behavior. Similarly, good health being a key measure of better work outcomes can also lower the risk for work stress among working older adults in later life.
Physical and functional health is important for the well-being of workers in the labor market. This becomes increasingly important with increase in age among the working older adults. Thus better physical and functional health conditions among the older adults are key for their work and well-being. Therefore, to examine these factors and understand their association with work stress among the older workers, we studied the risk factors like disability, social well-being and health risk behavior among older adults and their association with work stress. Our results clearly showed the significant impact of these factors on work stress. These results psychiatric disorders may likely enhance chronic work stress and impact overall health and well-being of older adults. Moreover they suggests that workers with any disability will experience greater work stress being more vulnerable due to age as found by earlier studies (Dew et al., 2005;Dewa et al., 2007).
Studies have identified a consistent association between work stress and health risk behavior, like smoking, drinking and obesity. There is a clear cut association between work stress and alcohol consumption found by earlier studies. We also find coherent results suggesting the need for health promotion programs to lower these behavioral risks among working elderly (Siegrist and Rödel, 2006). Work stress considerably impacts work outcomes at older ages and enhance the elderly labor market challenges. Older adults at upper ages are not only at risk of their well-being, but they also face multiple health challenges, which impact their productivity and work opportunity (Rauschenbach et al., 2012). This study finds a strong association of work stress with socio-economic and health factors among the older adults in India. Although we could not focus on the work outcomes despite the connotation being likely casual. However, while examining the role of work stress, we find a strong association of work stress with these risk individual factors at higher ages. Working older adults, being more vulnerable to health risks particularly of work stress therefore, need a significant attention. Age-related psychological challenges for older working adults must be addressed through specific policy measures to avert this labor market challenge and lower work stress among the older adults (Hansson et al., 2001). Similarly increased risk for health outcomes, like morbidity, disability, and functional limitation, will likely exhibit the significant levels of work stress, which can further impact their work opportunities. Therefore, strong emphasis is needed on workrelated health promotion programs to enhance well-being and ensure work opportunities for older adults in India. Moreover, social benefits can likely reduce the burden, which must also be addressed in labor policies and examined in future research.
Limitations of the study. The current study have some limitations such as the study was carried based on a very narrow definition of work stress given the data limitations. The study could not use the labor market outcomes in detail given the emphasis on health and wellbeing outcomes only in the survey. Moreover the possibility of contextual factors as confounding factors could not be considered in the study since they were not covered in the survey. The data is also not latest in the line however this is the only survey in India that covers this aspect. This study was also restricted to 60 and above as the survey covered only this population. Lastly the use of control group would have made the results more elaborative and could have added the greater strength to the final results of this study.

Conclusion and policy recommendations
Work stress augments at higher ages due to work, health and socio-economic conditions of the older adults. While better work opportunities may reflect a better health conditions of the older adults, work stress is likely going to be a challenge to address among older workers with increasing age. Therefore, providing a situational analysis of ageing and work stress among the working older adults in India.
To sum up this study finds significant impact of these underlying risk factors on the work stress, therefore a comprehensive healthcare approach is needed to address these psychological challenges of working older adults in the Indian labor market. Some key policy challenges that can be taken into consideration are as following. Labor policies should aim for improving health and wellbeing factors that can likely reduce the risk for work stress among the older adults in India. Similarly, older working adults should be of particular focus with respect to work stress and casual studies in this aspect are key to be carried out the further analysis.
Psychological health related challenges like work stress are increasing with rapid growth in aging among the countries like India. Therefore urgent need for labor and health related policies must be designed to ensure the protective measures against the stressful health outcomes. Similarly, future research is needed to take into account, how health and socio-economic status will enhance their risk for work stress and impact the labor market opportunities in other developing countries among the working older adults. Also other factors like living arrangements and social Authors own computation from BKPAI survey of India. Confidence interval in parentheses; LL is the lower confidence intervals and UL is the upper confidence limit in the parenthesis .® denotes is reference category of independent variables, respectively. Not having any work stress is the reference category for dependent variable. Significance level: ***significant at 1%, **significant at 5%, *significant at 10%; 95%.
security challenges can also be considered to examine the risk factors of work stress and its impact on work and employment opportunities among the older working adults in further research.

Data availability
The data that supports the findings of this study are available on request. The dataset used in the study is available in the public domain and can be accessed on a request from the Institute for Social and Economic Change (ISEC), Bangalore, India.