Heat stress, health and well-being: findings from a large national cohort of Thai adults

Introduction

Over the last decade interest has grown in the impact of global warming on human health.¹ Increasing heat stress has substantial adverse effects on population mortality and morbidity.^2–5 This information is from developed and temperate countries³ and leaves unanswered questions for tropical and developing countries where high temperature and humidity are common. Furthermore, heat stress in tropical cities is increasing due to urban heat island effects caused by industrial development and urbanisation in developing countries.⁶

Heat stress can have a major influence on daily human activities. The body absorbs external heat due to high air temperature and humidity, low air movement and high solar radiation; as well, some physical activities generate heat internally.⁷ Excess heat exposure during normal daily activities creates a high risk of recurrent dehydration and can cause other effects on physical health (eg, exhaustion, heat cramps, heat stroke or death).⁷ Heat stress affects mood, increases psychological distress and mental health problems,^8–10 and also reduces key human psychological performance variables.¹¹

Other heat stress impacts may arise from increased mistakes in daily activities and accidental injuries. As well, disturbed sleep and degraded physical performance from heat exhaustion reduce work capacity and lead to loss of income.^{8 12 13} Populations at risk of heat stress are not only the elderly but also young people and adults who are more likely to carry out heavy labour outdoors or work indoors without air conditioning or other effective cooling systems during the hot season.^{8 12 14}

In tropical Thailand, hot and humid conditions are common, especially in the hot season (March–June). The monthly maximum, mean and minimum temperatures averaged from 1999 to 2008 were around 33°C, 27°C and 22°C, respectively, with the averaged relative humidity at 75%. The monthly maximum temperatures averaged during 10 years varied little by region (32–33°C) and were highest in the North region during April (40°C) and lowest in the same region during December (24°C).¹⁵

Global warming (or ‘global heating’ may be a better description in relation to Thailand) is now causing increasing alarm in many tropical areas. For example, from 1951 to 2003, the monthly mean maximum temperature in Thailand increased by 0.56°C and the monthly mean minimum temperature increased even more at 1.44°C.¹⁶ Heat stress is already a concern in Thailand and the observed trends indicate further increase in air temperature.¹⁷ A recent study of occupational heat stress in Thailand by Langkulsen et al¹⁸ revealed a very serious problem (‘extreme caution’ or ‘danger’) in an array of work settings (they tested a pottery factory, a power plant, a knife manufacture site, a construction site and an agricultural site).

Heat stress in Thailand, its effects and pathways to exposure have been reported for two cities ^{19 20} and for workers.^{10 21} However, there is no available information on how much heat interferes with normal daily activities and heat stress effects on health and well-being in the general Thai population. Here we report an investigation of association between heat stress interference with daily activities and health and well-being in a large national cohort of young and middle-aged Thai adults.

Methods

Results

We first compared the 2005–2009 cohort to those who dropped out in 2009 (data not shown). The two groups were similar for age, sex ratio, employment, income and health outcomes studied here (energy levels, emotional problems and life satisfaction). Sociodemographic and health characteristics of the 60 569 cohort members followed up in 2009 are presented in table 1. There were slightly more women (54.8%), 70% were aged less than 40 years and 55.3% were married. Nearly 20% reported household monthly income of less than 10 000 Baht (300 $US) per month, 73.2% reported doing paid work and 56% resided in urban areas. Health-risk behaviours—regular smoking or regular alcohol drinking—were reported by 7.7% and 13.7%, respectively. By Asian standards,²⁷ half the cohort members were in the normal weight range, 9.5% were underweight, 18.8% were overweight and 22.1% were obese.

We noted that prevalence of ‘often’ heat interference for each daily activity are not much different in different regions of Thailand (33–42% for daily travel, 29–38% for work, 26–32% for housework, 23–29% for sleeping and 22–28% for exercise). Daily activities and heat interference frequency categories are summarised in table 2. Heat interference ‘often’ was reported (in order of frequency) by 37.5% for daily travel, 34.5% for work, 29.9% for housework, 27.4% for sleeping and 25.9% for exercise. Health and well-being frequency outcomes are reported in table 3: 37.6% reported being very satisfied with their life, around 15% reported having very much energy in the past 4 weeks and close to 11% reported no emotional problems in the past 4 weeks.

Daily activities show a clear trend connecting increasing heat interference with worse health and well-being (table 4). For example, cohort members who experienced heat interference ‘often’ while sleeping reported ‘extreme’ emotional problems (38.9%) much more frequently than ‘no’ emotional problems (16.4%). A similar pattern for ‘little or none’ energy levels was found for those reporting heat interference ‘often’ while sleeping (36.1% vs 22%) and the same trend was observed for life satisfaction (39.8% vs 22.2%). Daily travel and work have also shown strong gradients connecting frequent heat interference and worse health outcomes.

The multinomial logistic regression, adjusting for a wide array of potential confounders (see footnote in table 5), supported the descriptive results. For all three health outcomes, when each of the three graded-adverse outcome categories is compared with the optimal outcome, the relative odds ranged from 1.10 to 4.81. Furthermore, most ORs show a dose–response (for each health outcome, more heat interference associates more strongly with a given grade of abnormality). And 95% CIs for all ORs indicated statistical significance. So heat stress interfering with normal daily activities (sleep, housework, travel, work and exercise) associates with adverse outcomes for all three holistic measures of health. For example, reporting heat interference ‘often’ while sleeping was strongly associated with ‘little or none’ energy (OR=2.23, 95% CI 2.02 to 2.46), ‘extreme’ emotional problems (OR=4.81, 95% CI 4.32 to 5.36) and ‘low’ life satisfaction (OR=2.49, 95% CI 2.28 to 2.71). At work, reporting heat interference ‘often’ was associated with ‘little or none’ energy (OR=2.45, 95% CI 2.22 to 2.71) and ‘extreme’ emotional problems (OR=3.64, 95% CI 3.31 to 4.00). Similar results were found during daily travel and doing housework. A statistically significant association was also found for heat inference during exercise but the magnitude of the effect was lower than for other activities.

Discussion

Our study shows that climate-related heat stress in tropical Thailand associated with self-reported health and well-being if the heat interfered with daily activities such as sleep, housework, travel, work and exercise. The large study group included young and middle-age Thai adults, mostly doing paid work, with a little over half residing in urban areas. These cohort members are active and over 20% report often experiencing heat interference for daily activities during the hot season. Daily travel and work were sources of heat stress more often than other activities, probably because they involve time spent in traffic or outdoors during hot periods. Other activities such as housework have less heat stress than daily travel and work, perhaps because these activities are home based where air-conditioning or other ventilation is available.

We found those who report higher levels of heat stress interference with daily activities tend to also be the ones who have adverse health and well-being outcomes (low life satisfaction, low energy level and worse emotional problems). ORs of heat stress effects across all daily activities for emotional problems are between 1.55 and 4.81 and in general are worse than energy level effects (between 1.31 and 2.91) and life satisfaction effects (between 1.10 and 2.49). The worst health outcomes were for heat stress while sleeping followed by heat stress for daily travel, work, housework and exercise.

Our data are based on self-report by educated Thais and we note that questions on heat stress and health outcomes were in different parts of the questionnaire. Findings show strong and highly consistent trends especially for adverse health effects of frequent heat interference during sleep, daily travel and work. Elsewhere we have completed detailed analyses of associations between heat stress and self-reported health outcomes in the cohort using the questions from SF8.¹⁰ Our studied outcomes in this report were holistic fundamental measures of health. We can expect that those who had abnormal findings would also (already or eventually) manifest other more specific chronic diseases such as depression, obesity, hypertension and kidney disease. If so, the eventual burden of heat-related disease will be higher than currently recognised.^28–30

Our findings add to some previous reports on working in hot environments which found that heat stress significantly reduced people's motivation to do their work. Lan et al³¹ assessed office workers’ perceptions of thermal environment, emotions, well-being and motivation to work, and found that participants had lower motivation to work and experienced more negative moods in hot environments. Anderson found that the prolonged, continuous repetitive actions required to maintain performance at work and achieve target goals (such as getting a job finished) can lead to hypertension.³² And when more effort was required to complete a task in hot conditions loss of motivation was experienced leading to lower productivity and increased injury risk. The impact of heat stress on psychological performance variables¹¹ is a likely factor in these work-related impacts of heat.

Psychological effects of heat stress have been noted in other settings as well. Nitschke et al³³ reported a positive association between high-ambient temperature and hospital admissions for mental and behavioural disorders in Adelaide, Australia. Specific illnesses for which admissions increased included anxiety, symptomatic mood disorders and psychological development disorders among elderly people when temperature exceeded 26.7°C.³⁴ Moreover, excessive heat stress exposure may also increase violence.^32–35 Increasing heat stress had been associated with higher rates of aggressive behaviour,³⁶ and higher violent suicide rates.³⁷ In a meta-analysis, Bouchama et al³⁸ concluded that pre-existing mental health problems tripled the risk of all-cause mortality during a heat wave. A related issue is the physical and psychological exhaustion caused by extreme heat stress.⁷

In our study, we found that heat stress in Thailand is not only a problem at work but also heat stress interferes with other daily activities including sleeping, daily travel, housework and exercise. The results of our study complement other Thai research about adverse effects of heat. One recent report shows that heat stress in Thailand is a very serious problem in a wide variety of work settings.¹⁸ McMichael et al¹⁹ and Guo et al³⁹ found a temperature–mortality association and Pudpong and Hajat²⁰ found heat-related excess hospital admissions. Worker studies in Thailand related occupational heat stress, kidney disease and psychological distress.^{10 21}

One limitation of this study is that it could not directly establish that health and well-being outcomes arose as a result of heat stress. Interpreting causality between heat stress exposure and health and well-being outcomes is complex in a cross-sectional study as we cannot be completely sure that heat stress preceded their health condition and well-being. Also, the source of the heat stress was not reported and we could not make direct measurements of heat stress exposure and health and well-being outcomes. Another limitation of this study arose because people answered the questionnaire at different times of the year (but most in March–July—the hot period). The questions on physical and emotional health assessed the previous 4 weeks so most (almost all) were answering for the hot period.

The strength of this study is its large scale with participation from a national group of adults embedded in the socioeconomic mainstream of Thai society. Other strengths include the comprehensive questionnaire which captures a detailed assessment of health and an array of geodemographic, environmental and social attributes. Also, the cohort has been set up for future longitudinal analysis which will provide better insight into causal pathways between heat stress and subsequent health outcomes in the long run.

We conclude that Thai populations are at high risk of heat stress during daily activities. Also, in Thailand an anticipated increase in temperature from climate change plus the ageing and urbanisation of the population could significantly increase heat impacts on health and well-being. There is a need for improvements in public health surveillance and public awareness regarding the risks of heat stress which hitherto have been considered unremarkable in such a tropical environment.

Supplementary Material