Next Article in Journal
Spine Alignment in Standing and Maximal Upper Limb Elevation in Baseball Players with Lumbar Spondylolysis and Those without Low Back Pain
Next Article in Special Issue
The Impact of Climate Change on Aeromedical Retrieval Services in Remote Northern Australia: Planning for a Hotter Future
Previous Article in Journal
Maternal Dietary Diversity and Birth Weight in Offspring: Evidence from a Chinese Population-Based Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 20
Issue 4
10.3390/ijerph20043229
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Long-Term Apparent Temperature, Extreme Temperature Exposure, and Depressive Symptoms: A Longitudinal Study in China

by
Jianbo Jin
1,
Zhihu Xu
1,
Ru Cao
1,
Yuxin Wang
1,
Qiang Zeng
2,
Xiaochuan Pan
1,
Jing Huang
1 and
Guoxing Li
1,*
1
Department of Occupational and Environmental Health Sciences, Peking University School of Public Health, Beijing 100191, China
2
Department of Occupational Disease Control and Prevention, Tianjin Centers for Disease Control and Prevention, Tianjin 300011, China
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2023, 20(4), 3229; https://doi.org/10.3390/ijerph20043229
Submission received: 29 December 2022 / Revised: 7 February 2023 / Accepted: 10 February 2023 / Published: 12 February 2023
(This article belongs to the Special Issue Climate Change and Medical Responses)
Browse Figures
Versions Notes

Abstract

:
Temperature is increasingly understood to impact mental health. However, evidence of the long-term effect of temperature exposure on the risk of depressive symptoms is still scarce. Based on the China Health and Retirement Longitudinal Study (CHARLS), this study estimated associations between long-term apparent temperature, extreme temperature, and depressive symptoms in middle-aged and older adults. Results showed that a 1 °C increase or decrease from optimum apparent temperature (12.72 °C) was associated with a 2.7% (95% CI: 1.3%, 4.1%) and 2.3% (95% CI: 1.1%, 3.5%) increased risk of depressive symptoms, respectively. This study also found that each percent increase in annual change in ice days, cool nights, cool days, cold spell durations, and tropical nights was associated with higher risk of depressive symptoms, with HRs (95%CI) of 1.289 (1.114–1.491), 2.064 (1.507–2.825), 1.315 (1.061–1.631), 1.645 (1.306–2.072), and 1.344 (1.127–1.602), respectively. The results also indicated that people living in northern China have attenuated risk of low apparent temperature. Older people were also observed at higher risk relating to more cool nights. Middle-aged people, rural residents, and people with lower household income might have higher related risk of depressive symptoms due to increased tropical nights. Given the dual effect of climate change and global aging, these findings have great significance for policy making and adaptive strategies for long-term temperature and extreme temperature exposure.

1. Introduction

It is generally agreed that a range of factors contribute to the development of depression, including biological, environmental, genetic, and psychological factors [1,2]. One environmental factor that has been reported to have an impact on the incidence of depression is temperature, with both high or cold temperatures having been shown to have acute effects on the occurrence of depression, such as increased admissions to emergency departments and hospitalizations [3,4,5,6,7]. However, the long-term effects of temperature on depression have been found inconsistent in previous studies. While a study in Ireland found no significant association between annual average temperature and the risk of depression [8], studies in Spain and Taiwan have suggested that higher annual average temperature may be related to an increased risk of depression [9,10]. In addition, given the potential impacts of climate change, it is important to investigate the effects of increasing extreme temperature events on mental health [11].
Several limitations have plagued research on the relationship between long-term temperature and depression. First, previous studies have often been conducted in regions with a limited range of climate conditions [9], and few have explored the effects of low temperature in cold climates. Second, few studies have considered the influence of various meteorological factors on mental health outcomes. While ambient temperature is a key factor in human thermal perception, other factors such as relative humidity and wind speed can also play a role. The concept of apparent temperature, which combines air temperature and other meteorological factors, may provide a more comprehensive assessment of thermal feeling [12], and more and more studies focused on the thermal environment [13,14]. Although previous studies have found a relationship between apparent temperature and mental health outcomes [15,16,17], the impact of apparent temperature on depression has not been extensively investigated.
In this study, we first explored the long-term effect of temperature on depressive symptoms for middle-aged and older adults based on the China Health and Retirement Longitudinal Study (CHARLS) [18], which repeatedly measured the depressive disorder of a nationally representative sample of middle-aged and older Chinese adults in 2011, 2013, 2015, and 2018 [19]. We also investigated the impact of the annual change in extreme temperature events on depressive symptoms.

2. Materials and Methods

2.1. Data Source and Processing

We collected data from the China Health and Retirement Longitudinal Study (CHARLS), which is a national representative longitudinal survey of Chinese adults over the age of 45 (https://charls.charlsdata.com, accessed on 15 March 2021). Using the multi-stage probability sampling method, CHARLS selected 150 county-level units from 450 communities in 125 cities of 28 provinces of China [20]. The survey includes four waves, including the 2011–2012 baseline (W1) survey, with 17 596 individual participants recruited; the second wave (W2) survey, with 15,179 followed in 2013–2014; the third wave (W3) survey, with 13,002 followed in 2015–2016; and the fourth wave (W4) survey, with 11,486 followed in 2017–2018.
We excluded 4952 participants with depressive symptoms and 6878 participants who did not provide depressive symptom information at W1, W2, W3, or W4 or who withdrew from the survey to identify a group of respondents without depressive symptoms at baseline. We excluded 158 participants for lack of covariate information and 8 participants in cities with few cases (Lanzhou and Shenzhen). Our final analysis sample included 5600 individuals who had no depressive symptoms at baseline, and then provided complete data for all study variables in W1, W2, W3, and W4 (Figure S1). The likelihood of non-responses seemed to not correlate with specific characteristics (4555 participants without information of depressive symptoms or in withdrawal group) [20].
CHARLS obtained ethical approved for collecting data on human subjects by Peking University Institutional Review Board and all participants provided written informed consent.

2.2. Variable Design

2.2.1. Depressive Symptom Measurement

We used a binary variable of having depressive symptoms or not. The 10-item short form of the Center for Epidemiologic Studies Depression Scale (CESD-10) was used to measure depressive symptoms. CESD-10 has been commonly used to examine depressive symptoms among Chinese middle-aged and older adults with satisfactory validity and reliability [21]. Scores in the range of 0–30 indicated the degree of depressive symptoms. The cut-off point of 10 was used to generate the binary depressive symptoms variable (1 = yes, 0 = no) [22]. Those who had scores above 10 were classified as having depressive symptoms. Details of the CESD-10 are described in the Appendix.

2.2.2. Temperature Exposure Assessment

In this study, we adopted apparent temperature as the temperature exposure. Apparent temperature, as a comprehensive index of ambient temperature, relative humidity, and wind speed, is more objective to reflect human body perception than the average ambient temperature [12,23]. The daily meteorological information (ambient temperature, relative humidity, wind speed, and sunlight hours) of all selected cities in the same period (2011–2018) was obtained from the China Meteorological Science Data Sharing Service Network: (http://data.cma.cn/data/cdcdetail/dataCode/SURF_CLI_CHN_MUL_DAY.html, accessed on 15 March 2021). We first calculated daily apparent temperature by using the original data of each weather station, and further calculated the monthly averages. The apparent temperature was calculated by the following equation [17]:
A T = T a + 0.33 ∗ e − 0.70 ∗ W S − 4.00
e = R h / 100 * 6.105 ∗ exp ( 17.27 ∗ T a / ( 237.7 + T a ) )
where Ta is the temperature (°C), Rh is the relative humidity (%), and WS is the average wind speed (m/s).
Nearest-neighbor interpolation was applied to estimate the monthly data across the mainland of China at a spatial resolution of a regular grid of 0.1° × 0.1° (10 km × 10 km). Due to confidential reasons, the subjects of CHARLS could only be geocoded to city-level codes. Therefore, we matched the longitude and latitude of the city centroid with the regular grids. We utilized the internal average temperature between two waves of survey as the exposure values.
The tenfold cross validation (CV) method was used to validate the accuracy of estimated temperature. The results of tenfold CV are shown in Figure S2, and the estimates were closely compliant with monthly in situ observations (R2 = 0.95; RMSE = 2.34 °C).
We constructed extreme temperature indices considering previous studies [24,25], using the daily maximum and minimum ambient temperatures in the period of 2010–2018 for each city. We finally chose ten typical extreme indices, as shown in Supplementary Table S1. The annual changes in extreme temperature indices in the period of 2011–2018 were calculated by the following equation:
a n n u a l   c h a n g e = e x t r e m e   t e m p e r a t u r e   i n d i c e i − e x t r e m e   t e m p e r a t u r e   i n d i c e i − 1 e x t e x t r e m e   t e m p e r a t u r e   i n d i c e i − 1
where i indicates the year from 2011–2018.

2.2.3. Covariates

Recorded demographic characteristics (age, sex, and marriage) and socioeconomic status covariates (education attainment, urbanicity of residence, and household income per capita) were extracted from CHARLS. The city-level per capita gross domestic product (GDP) was also collected from the statistical yearbook, the National Bureau of Statistics, and China’s national knowledge infrastructure. Age was divided into middle-age and old by cut-off of 65 years. Household income status was divided into binary by average income. Educational attainment was divided into binary of whether participants attained junior school. By the Kunlun–Qinling–Huaihe River line, the cities were divided into northern and southern cities, with or without heating policy. The index table of variates is shown in Supplementary Table S2.

2.3. Statistical Analyses

We assessed the association between temperature and the risk of depressive symptoms using a time-varying Cox proportional hazards model on a year-based time scale. The proportional hazards assumption was tested using Schoenfeld residuals. The visualization of the conceptual causal pathway is shown as a directed acyclic graph (DAG) in Supplementary Figure S3. We adjusted for age (continuous), sex (male vs. female), marriage (married or single), education attainment (higher than primary school vs. primary school or lower), urbanicity of residence (urban vs. rural), household income status (higher than average or lower than average), annual average sunlight hours (log), and city-level GDP (continuous).
Firstly, we used punitive spline regression (df = 3) to analyze the exposure–response curve of long-term exposure to temperature and depressive symptoms. After nonparametric estimation, optimum temperature with lowest risk would be the cut-off to stratify temperatures into high and low temperatures. Secondly, we estimated the association between annual change in extreme temperature and depressive symptoms, additionally controlling for annual average apparent temperature (log) and annual average sunlight hours (log).
We conducted several sensitivity analyses to confirm the robustness of our results. First, we repeated the main analyses by using over 12 scores of CEDS-10 as the cut-off. Second, due to different duration times in 3 years between 2015–2018, we repeated the main analyses by using the moving average apparent temperature of the 2 years before each survey as the temperature exposure. Third, we used annual average ambient temperature as exposure to explore the exposure–response curve.
We evaluated the modification in the association between long-term temperature and depressive symptoms by stratifying age (≥65 years and <65 years), sex (male or female), urbanicity of residence (rural or urban), household income status (higher than average or lower than average), education attainment (primary school or below, or junior high school or above), and geographic location (living in northern or southern cities). Effect modification analysis was performed by adding an interaction term between temperature and the testing variable in the model.
Data arrangement, cleaning, and all statistical analyses were conducted using R (version 4.0.2) with packages dplyr, survival, and smoothHR. Statistical significance was defined as p < 0.05, two sides.

3. Results

3.1. Descriptive Statistics

We finally included 5600 participants without depressive symptoms at baseline (mean age of 57.18 yr; 52.3% male; 92.7% married; 64.6% with education higher than primary school; 39.7% in urban areas) (Table 1 and Supplementary Table S3). In cities of CHARLS, the average annual apparent temperature between 2011 and 2018 ranged from −6.89–25.94 °C (median: 14.94 °C). During the follow-up period, 2457 of participants were followed with depressive symptoms. Participants were detected to have depressive symptoms in 5.61 years on average. Average annual apparent temperature and incidence rate of depressive symptoms in 2011–2018 are shown in Figure 1. It was obvious that apparent temperature was significantly higher in the south than in the north. The incidences of depressive symptoms in each city in China were between 0.71% and 9.09%, and the mean annual incidence was 2.85%.

3.2. Main Results

After adjusting for covariates such as age, sex, marriage status, household income, education attainment, urbanicity of residence, city-level GDP, and annual average sunlight hours, the long-term effect of low/high apparent temperature on depression shows to be non-linear (Figure 2). Long-term exposure to both low and high apparent temperature tended to increase the risk of depressive symptoms.
Based on the exposure–response curve of long-term exposure to apparent temperature and depressive symptoms, we stratified average apparent temperatures into high and low temperatures. The optimum apparent temperature (12.72 °C) was chosen as the cut-off. High or low apparent temperature referred to apparent temperature higher or lower than 12.72 °C, respectively. Table 2 showed the risks of high and low apparent temperature, in that the risk of depressive symptoms increased by 3.2% and 2.4% when the apparent temperature was higher or lower than 12.72 °C, respectively (Table 2).
We also observed that each percent increase of annual change in ice days, cool nights, cool days, and cold spell durations was associated with higher risk of depressive symptoms, with HRs (95%CI) of 1.289 (1.114–1.491), 2.064 (1.507–2.825), 1.315 (1.061–1.631), and 1.645 (1.306–2.072), respectively. For the annual change in extreme hot temperatures, the change in tropical nights was associated with depressive symptoms, with an HR (95%CI) of 1.344 (1.127–1.602).

3.3. Modification

We estimated the effect of long-term temperature and extreme temperature in different subgroups and checked the significance of modification. For participants exposed to low apparent temperature, a significant difference was found in the long-term low-apparent-temperature-related risk between participants living in northern (hazard ratio and 95%CI: 1.031 (1.018,1.044)) and southern cities (hazard ratio and 95%CI: 1.115 (1.053,1.181)) (interaction p value = 0.014) (Table 3). In addition, older people seem to have higher risk relating to more cool nights (interaction p value = 0.007) (Supplementary Figure S5). As for more tropical nights, middle-aged people, rural residents, and people with lower household income might have higher related risk of depressive symptoms (interaction p values: 0.007, 0.046, and 0.003, respectively) (Figure 3). No other significance was observed, and the results of stratification analysis are shown in Supplementary Figures S4–S7.

3.4. Sensitivity Analyses

To confirm the robustness of our results, we conducted several sensitivity analyses. The main results are still robust, as shown in Supplementary Table S4. We chose a score of 12 as the cut-off point to classify depressive symptoms and used the moving average apparent temperature of the year before each survey as the temperature exposure. We also found the robust U-shaped exposure–response curve when using annual ambient temperature as exposure (Supplementary Figure S8).

4. Discussion

This study found a U-shaped exposure response curve between depressive symptoms and long-term temperature exposure, and both high and low apparent temperatures increased the risk of depressive symptom incidence. Extreme hot and cold were significantly associated with depressive symptoms. People living in southern China might be more sensitive to low temperature, and older people perhaps suffered more depressive symptom risk confronting more extreme cold events. Middle-aged people, rural residents, and people with low household income might have higher extreme-heat-related risk of depressive symptoms.
A study conducted in Taiwan also found a non-linear exposure–response relationship between long-term temperature and depression, but found no significant effects of low temperature (with a threshold temperature of around 23 °C) [9]. In addition, a study in Spain found a significant association only in regions with the highest level of temperature [10]. However, there might be several reasons for this discrepancy in findings. One is the difference in the depression measurement, as the study in Taiwan measured major depressive disorder and the study in Spain measured depression only by self-reported diagnosed depression and antidepressant treatment, while our study measured depressive symptoms. Additionally, the temperature range in our study was wider than that of Taiwan and Spain (ambient temperature: −1.76–23.38 °C in our study vs. 16.5–25.4 °C in Taiwan vs. 11.4–20.2 °C in Spain). Using Figure S8, we can visually explain the inconsistency over a wider range of temperatures. Annual average ambient temperature in Spain was around optimum temperature, showing that only the highest level of temperature was associated with depression, while temperature in Taiwan was above optimum temperature, showing a significant impact of high temperature but no significant impact of low temperature.

4.1. Explanation of the Potential Mechanism, and Modification

The observed association may be plausible in the biological mechanism. Both heat and cold stress may affect alterations in autonomic function [26,27,28], in which prolonged imbalance may trigger the chronic low-grade inflammatory reaction, a potential cause of depression [29]. Humidity may also play a role in this relationship by affecting the body’s ability to cool down through sweat evaporation, potentially exacerbating the effects of extreme heat on mental health in vulnerable individuals [30]. Additionally, increased nighttime temperature may disturb sleep–wake cycles governed by circadian rhythms [31], which induces excessive activity of the Hypothalamic–Pituitary–Adrenal (HPA) Axis and further mental disorders [32]. Specifically, for those who are prone to acute or chronic mental problems, the long-term inability to get rid of high temperature might cause irritability and intermittent psychological distress [33]. Reduced cutaneous thermal sensitivity and subjective thermal perception during cooling may also make older people more susceptible to cold exposure [34]. However, the biological mechanism of how high or low temperature affects depressive symptoms needs further investigation.
The results of this study showed that the effects of low temperature on depression vary between northern and southern regions in China. This may be interpreted as the result of the China’s Huai River policy, which provides central heating or low-cost indoor heating during the winter to cities north of the Huai River, but not to those in the south. In northern China, people use either central heating provided by local systems or domestic stoves burning subsidized bulk coal for heating [35]. China is facing a dilemma, in which the results in this study suggest that heating policy is associated with decreased temperature-related depressive symptoms, but both heating modes release air pollutants that damage human health [36]. Therefore, it is important to promote more efficient, clean heating strategies, especially in the construction of future heating systems in cities that are affected by low temperatures.
Younger individuals and those with low socioeconomic status were reported to be more vulnerable to heat-related anxiety and mood disorders [37,38], which is consistent to our results shown in Figure 3. However, other research has indicated that older individuals may be the most vulnerable to extreme temperature [9,10,39]. It is necessary to conduct further studies using large datasets that include various mental disorders to determine the modifying effects and identify vulnerable populations in order to implement effective prevention strategies.

4.2. Strengths and Limitations

This study has several strengths. The study provided novel findings of the non-linear association between long-term apparent temperature exposure and depressive symptoms. Our findings suggested that besides extreme hot, extreme cold is also concerning due to climate change.
However, there are some limitations in this study. First, because of the limitation of geographical information, exposure of temperature was assessed at the city level, which may induce some degree of exposure misclassification. However, this might not attenuate the risk estimates [40]. Second, since the research objects are middle-aged and older people over 45 years old, the results cannot represent the impact of temperature on depression in younger people. Third, the long-term temperature exposure could be affected by other potential unknown confounding factors, which may lead to inaccurate estimation.

5. Conclusions

In a longitudinal study of an aging population, we observed that long-term residence in regions with extreme cold or heat was associated with the risk of depressive symptoms. The results also indicated that people living in northern China have attenuated risk of low apparent temperature. Middle-aged people and people with lower socioeconomic status are more vulnerable to extreme heat, and older people are more vulnerable to extreme cold. Under climate change and the aging of the population, these findings provided implications for policy making and adaptive strategies. Further studies are needed to investigate the underlying mechanisms for the reported association.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijerph20043229/s1, Appendix method; Figure S1: Flowchart of the study samples; Figure S2: Cross-validation for the monthly estimates of apparent temperature (2011-2018) in China; Figure S3: Conceptual causal pathway is shown as directed acyclic graph (DAG); Figure S4: Hazards ratios of depressive symptoms each percent increase of annual change of ice days by baseline characteristics; Figure S5: Hazards ratios of depressive symptoms each percent increase of annual change of cool nights by base-line characteristics; Figure S6: Hazards ratios of depressive symptoms each percent increase of annual change of cool days by baseline characteristics; Figure S7: Hazards ratios of depressive symptoms each percent increase of annual change of cool duration by baseline characteristics; Figure S8: The exposure-response curve of long-term ambient temperature exposure and depressive symptoms; Table S1: Definition of extreme temperature indicators; Table S2: Index table of the variables; Table S3: Summary statistics on population characteristics (N = 17,596); Table S4: The associations between temperature and depressive symptoms in sensitive analysis, by using 12 scores as a cut-off of depressive symptoms, or 2-year moving average apparent temperature before survey time.

Author Contributions

Conceptualization, G.L. and J.H.; methodology, J.J. and Z.X.; formal analysis, J.J., R.C., Y.W., Q.Z. and X.P.; writing—original draft preparation, J.J.; supervision, G.L. and J.H. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by 2018YFE0115300 from the China National Key Research and Development Program.

Institutional Review Board Statement

The CHARLS was approved by Peking University’s Ethical Review Committee (IRB 0000105211015).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The population data (CHARLS) that support the findings of this study are available from (http://charls.pku.edu.cn/index/en.html, accessed 15 March 2021). The climate data that support the findings of this study are available from (http://data.cma.cn/data/cdcdetail/dataCode/SURF_CLI_CHN_MUL_DAY.html, accessed 15 March 2021).

Acknowledgments

Thanks to the China Center for Economic Research, National School of Development, Peking University, for providing the CHARLS data.

Conflicts of Interest

Other than the grants listed in the acknowledgement section, the authors declare that they have no other competing interest.

References

  1. Edition, F. Diagnostic and statistical manual of mental disorders. Am. Psychiatr. Assoc. 2013, 21, 143–144. [Google Scholar]
  2. Ménard, C.; Hodes, G.E.; Russo, S.J. Pathogenesis of depression: Insights from human and rodent studies. Neuroscience 2016, 321, 138–162. [Google Scholar] [CrossRef] [PubMed]
  3. Almendra, R.; Loureiro, A.; Silva, G.; Vasconcelos, J.; Santana, P. Short-term impacts of air temperature on hospitalizations for mental disorders in Lisbon. Sci. Total Environ. 2019, 647, 127–133. [Google Scholar] [CrossRef] [PubMed]
  4. Bullock, B.; Murray, G.; Meyer, D. Highs and lows, ups and downs: Meteorology and mood in bipolar disorder. PLoS ONE 2017, 12, e0173431. [Google Scholar] [CrossRef]
  5. Chan, E.Y.Y.; Lam, H.C.Y.; So, S.H.W.; Goggins, W.B.; Ho, J.Y.; Liu, S.; Chung, P.P.W. Association between Ambient Temperatures and Mental Disorder Hospitalizations in a Subtropical City: A Time-Series Study of Hong Kong Special Administrative Region. Int. J. Environ. Res. Public Health 2018, 15, 754. [Google Scholar] [CrossRef] [PubMed]
  6. Sung, T.-I.; Chen, M.-J.; Su, H.-J. A positive relationship between ambient temperature and bipolar disorder identified using a national cohort of psychiatric inpatients. Soc. Psychiatry Psychiatr. Epidemiol. 2013, 48, 295–302. [Google Scholar] [CrossRef]
  7. Zhang, S.; Yang, Y.; Xie, X.; Li, H.; Han, R.; Hou, J.; Sun, J.; Qian, Z.; Wu, S.; Huang, C.; et al. The effect of temperature on cause-specific mental disorders in three subtropical cities: A case-crossover study in China. Environ. Int. 2020, 143, 105938. [Google Scholar] [CrossRef]
  8. O’Hare, C.; O’Sullivan, V.; Flood, S.; Kenny, R.A. Seasonal and meteorological associations with depressive symptoms in older adults: A geo-epidemiological study. J. Affect. Disord. 2016, 191, 172–179. [Google Scholar] [CrossRef] [PubMed]
  9. Chen, N.-T.; Lin, P.-H.; Guo, Y.-L.L. Long-term exposure to high temperature associated with the incidence of major depressive disorder. Sci. Total Environ. 2019, 659, 1016–1020. [Google Scholar] [CrossRef]
  10. Henríquez-Sánchez, P.; Doreste-Alonso, J.; Martínez-González, M.A.; Bes-Rastrollo, M.; Gea, A.; Sánchez-Villegas, A. Geographical and climatic factors and depression risk in the SUN project. Eur. J. Public Health 2014, 24, 626–631. [Google Scholar] [CrossRef]
  11. Berry, H.L.; Bowen, K.; Kjellstrom, T. Climate change and mental health: A causal pathways framework. Int. J. Public Health 2010, 55, 123–132. [Google Scholar] [CrossRef] [PubMed]
  12. Oppermann, E.; Kjellstrom, T.; Lemke, B.; Otto, M.; Lee, J.K.W. Establishing intensifying chronic exposure to extreme heat as a slow onset event with implications for health, wellbeing, productivity, society and economy. Curr. Opin. Environ. Sustain. 2021, 50, 225–235. [Google Scholar] [CrossRef]
  13. Yang, J.; Yang, Y.; Sun, D.; Jin, C.; Xiao, X. Influence of urban morphological characteristics on thermal environment. Sustain. Cities Soc. 2021, 72, 103045. [Google Scholar] [CrossRef]
  14. Zhang, D.; Zhou, C.; Zhou, Y.; Zikirya, B. Spatiotemporal relationship characteristic of climate comfort of urban human settlement environment and population density in China. Front. Ecol. Evol. 2022, 10, 695. [Google Scholar] [CrossRef]
  15. Basu, R.; Gavin, L.; Pearson, D.; Ebisu, K.; Malig, B. Examining the association between apparent temperature and mental health-related emergency room visits in California. Am. J. Epidemiol. 2018, 187, 726–735. [Google Scholar] [CrossRef] [PubMed]
  16. Santurtún, A.; Almendra, R.; Silva, G.L.; Fdez-Arroyabe, P.; Santurtún, M.; Santana, P. Suicide and apparent temperature in the two capitals cities in the iberian peninsula. Soc. Sci. Med. 2020, 265, 113411. [Google Scholar] [CrossRef]
  17. Yi, W.; Zhang, X.; Gao, J.; Wei, Q.; Pan, R.; Duan, J.; Xu, Z.; Zhang, Y.; Bai, L.; Cheng, Q.; et al. Examining the association between apparent temperature and admissions for schizophrenia in Hefei, China, 2005–2014: A time-series analysis. Sci. Total Environ. 2019, 672, 1–6. [Google Scholar] [CrossRef]
  18. Zhao, Y.; Hu, Y.; Smith, J.P.; Strauss, J.; Yang, G. Cohort profile: The China Health and Retirement Longitudinal Study (CHARLS). Int. J. Epidemiol. 2014, 43, 61–68. [Google Scholar] [CrossRef]
  19. Wen, Y.; Liu, C.; Liao, J.; Yin, Y.; Wu, D. Incidence and risk factors of depressive symptoms in 4 years of follow-up among mid-aged and elderly community-dwelling Chinese adults: Findings from the China Health and Retirement Longitudinal Study. BMJ Open 2019, 9, e029529. [Google Scholar] [CrossRef]
  20. Zhao, Y.; Strauss, J.; Yang, G.; Giles, J.; Hu, P.; Hu, Y.; Lei, X.; Park, A.; Smith, J.P.; Wang, Y. China health and retirement longitudinal study–2011–2012 national baseline users’ guide. Beijing Natl. Sch. Dev. Peking Univ. 2013, 2, 1–56. [Google Scholar]
  21. Sun, X.; Li, Y.; Yu, C.; Li, L. Reliability and validity of depression scales of Chinese version: A systematic review. Zhonghua Liu Xing Bing Xue Za Zhi= Zhonghua Liuxingbingxue Zazhi 2017, 38, 110–116. [Google Scholar]
  22. Lei, X.; Sun, X.; Strauss, J.; Zhang, P.; Zhao, Y. Depressive symptoms and SES among the mid-aged and elderly in China: Evidence from the China Health and Retirement Longitudinal Study national baseline. Soc. Sci. Med. 2014, 120, 224–232. [Google Scholar] [CrossRef]
  23. Kovats, R.S.; Hajat, S. Heat stress and public health: A critical review. Annu. Rev. Public Health 2008, 29, 41–55. [Google Scholar] [CrossRef] [PubMed]
  24. Yang, Z.; Wang, Q.; Liu, P. Extreme temperature and mortality: Evidence from China. Int. J. Biometeorol. 2019, 63, 29–50. [Google Scholar] [CrossRef] [PubMed]
  25. Ren, G.; Zhou, Y. Urbanization Effect on Trends of Extreme Temperature Indices of National Stations over Mainland China, 1961–2008. J. Clim. 2014, 27, 2340–2360. [Google Scholar] [CrossRef]
  26. Greaney, J.L.; Kenney, W.L.; Alexander, L.M. Sympathetic regulation during thermal stress in human aging and disease. Auton Neurosci 2016, 196, 81–90. [Google Scholar] [CrossRef] [PubMed]
  27. Harinath, K.; Malhotra, A.S.; Pal, K.; Prasad, R.; Kumar, R.; Sawhney, R.C. Autonomic nervous system and adrenal response to cold in man at Antarctica. Wilderness Env. Med. 2005, 16, 81–91. [Google Scholar] [CrossRef]
  28. Mäkinen, T.M.; Mäntysaari, M.; Pääkkönen, T.; Jokelainen, J.; Palinkas, L.A.; Hassi, J.; Leppäluoto, J.; Tahvanainen, K.; Rintamäki, H. Autonomic nervous function during whole-body cold exposure before and after cold acclimation. Aviat. Space Environ. Med. 2008, 79, 875–882. [Google Scholar] [CrossRef]
  29. Halaris, A. Inflammation and depression but where does the inflammation come from? Curr. Opin. Psychiatry 2019, 32, 422–428. [Google Scholar] [CrossRef]
  30. Ding, N.; Berry, H.L.; Bennett, C.M. The importance of humidity in the relationship between heat and population mental health: Evidence from Australia. PloS ONE 2016, 11, e0164190. [Google Scholar] [CrossRef]
  31. Obradovich, N.; Migliorini, R.; Mednick, S.C.; Fowler, J.H. Nighttime temperature and human sleep loss in a changing climate. Sci. Adv. 2017, 3, e1601555. [Google Scholar] [CrossRef]
  32. Nicolaides, N.C.; Vgontzas, A.N.; Kritikou, I.; Chrousos, G. HPA Axis and Sleep. In Endotext; Feingold, K.R., Anawalt, B., Blackman, M.R., Boyce, A., Chrousos, G., Corpas, E., de Herder, W.W., Dhatariya, K., Hofland, J., Dungan, K., et al., Eds.; MDText.com, Inc.; Copyright © 2000–2023; MDText.com, Inc.: South Dartmouth, MA, USA, 2000. [Google Scholar]
  33. Page, L.A.; Hajat, S.; Kovats, R.S. Relationship between daily suicide counts and temperature in England and Wales. Br. J. Psychiatry 2007, 191, 106–112. [Google Scholar] [CrossRef] [PubMed]
  34. Smolander, J. Effect of cold exposure on older humans. Int. J. Sports Med. 2002, 23, 86–92. [Google Scholar] [CrossRef]
  35. Feng, T.; Du, H.; Coffman, D.M.; Qu, A.; Dong, Z. Clean heating and heating poverty: A perspective based on cost-benefit analysis. Energy Policy 2021, 152, 112205. [Google Scholar] [CrossRef]
  36. Almond, D.; Chen, Y.; Greenstone, M.; Li, H. Winter heating or clean air? Unintended impacts of China’s Huai river policy. Am. Econ. Rev. 2009, 99, 184–190. [Google Scholar] [CrossRef]
  37. Cianconi, P.; Betrò, S.; Janiri, L. The Impact of Climate Change on Mental Health: A Systematic Descriptive Review. Front. Psychiatry 2020, 11, 74. [Google Scholar] [CrossRef] [PubMed]
  38. Thompson, R.; Hornigold, R.; Page, L.; Waite, T. Associations between high ambient temperatures and heat waves with mental health outcomes: A systematic review. Public Health 2018, 161, 171–191. [Google Scholar] [CrossRef]
  39. Lee, S.; Lee, H.; Myung, W.; Kim, E.J.; Kim, H. Mental disease-related emergency admissions attributable to hot temperatures. Sci. Total Environ. 2018, 616–617, 688–694. [Google Scholar] [CrossRef]
  40. Tian, Z.; Li, S.; Zhang, J.; Jaakkola, J.J.K.; Guo, Y. Ambient temperature and coronary heart disease mortality in Beijing, China: A time series study. Environ. Health 2012, 11, 56. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Ijerph 20 03229 g001 550
Figure 1. Map of average annual apparent temperature and incidence of depressive symptoms in each city (2011−2018). Dashed line refers to Kunlun−Qinling−Huaihe River line, dividing cities into the south and the north. Small islands or areas with missing data are not displayed in that map.
Figure 1. Map of average annual apparent temperature and incidence of depressive symptoms in each city (2011−2018). Dashed line refers to Kunlun−Qinling−Huaihe River line, dividing cities into the south and the north. Small islands or areas with missing data are not displayed in that map.
Ijerph 20 03229 g001
Ijerph 20 03229 g002 550
Figure 2. The exposure−response curve of long-term temperature exposure and depressive symptoms. Age, sex, marriage status, household income, education attainment, urbanicity of residence, gross domestic product (GDP), and annual average sunlight hours were adjusted. The solid line represents log hazard ratio, and gray zone indicates 95% confidence interval.
Figure 2. The exposure−response curve of long-term temperature exposure and depressive symptoms. Age, sex, marriage status, household income, education attainment, urbanicity of residence, gross domestic product (GDP), and annual average sunlight hours were adjusted. The solid line represents log hazard ratio, and gray zone indicates 95% confidence interval.
Ijerph 20 03229 g002
Ijerph 20 03229 g003 550
Figure 3. Hazards ratios of depressive symptoms, each percent increase in annual change in tropical nights by baseline characteristics.
Figure 3. Hazards ratios of depressive symptoms, each percent increase in annual change in tropical nights by baseline characteristics.
Ijerph 20 03229 g003
Table
Table 1. Summary statistics of the participants’ baseline characteristics.
Table 1. Summary statistics of the participants’ baseline characteristics.
CharacteristicsParticipants Followed
with Depressive Symptoms		Healthy
Participants		Total Participants
at Baseline
Subjects245731435600
  Individual-level variables
    Age (mean (SD)), yr57.55 (8.46)56.90 (8.04)57.18 (8.23)
    Sex = Male 1080 (44.0%)1849 (58.8%)2936 (52.4%)
    Marriage status = married2248 (91.5%)2943 (93.6%)5191 (92.7%)
    Education attainment = higher than primary school1377 (56.0%)2243 (71.4%)2279 (40.7%)
    Urbanicity = urban residence829 (33.7%)1394 (44.4%)2223 (39.7%)
    Household income (mean (SD)), CNY13,111.11 (29006.50)16,248.56 (27902.13)14,872.01 (28432.06)
Area-level variables
    Geographic location = living in the southern China1429 (58.2%)1501 (47.8%)2930 (52.3%)
    City level GDP (mean (SD)), CNY15.86 (18.53)20.43 (23.28)18.43 (21.44)
Meteorological variables
    Apparent temperature (mean (SD)), °C13.06 (6.42)12.55 (5.98)12.75 (6.03)
    Ambient temperature (mean (SD)), °C13.91 (5.02)13.49 (4.73)137.29 (47.46)
    Relative humidity (mean (SD)), %67.95 (7.86)66.90 (7.36)66.49 (7.16)
    Sunlight (mean (SD)), hours/day49.07 (12.16)50.63 (11.06)50.92 (10.76)
Extreme hot indicators
    Tropical nights (mean (SD)), days76.87 (53.56)75.13 (47.87)76.30 (49.65)
    Summer days (mean (SD)), days135.92 (45.84)133.27 (40.99)131.48 (41.57)
    Warm nights (mean (SD)), days32.45 (4.27)32.07 (4.10)31.13 (7.53)
    Warm days (mean (SD)), days33.23 (5.67)31.75 (5.26)34.89 (9.40)
    Warm spell duration (mean (SD)), days28.60 (6.47)27.13 (6.20)30.08 (10.33)
Extreme cold indicators
    Frost nights (mean (SD)), days69.21 (62.03)77.75 (59.13)73.92 (58.36)
    Ice days (mean (SD)), days20.78 (35.61)23.58 (36.60)20.95 (34.13)
    Cool nights (mean (SD)), days46.83 (4.42)47.21 (4.10)48.81 (6.00)
    Cool days (mean (SD)), days54.34 (5.70)53.98 (5.29)51.24 (5.56)
    Cold spell duration (mean (SD)), days42.98 (5.13)43.40 (4.87)45.35 (6.22)
Note: yr = years; SD = standard difference; GDP = gross domestic product.
Table
Table 2. Cox regression models of time to having depressive symptoms, 2011–2018.
Table 2. Cox regression models of time to having depressive symptoms, 2011–2018.
VariableHazard Ratio (95% CI)
Apparent temperature
High temperature1.027(1.013–1.041) ***
Low temperature1.023(1.011–1.035) ***
Extreme hot
Tropical nights1.344(1.127–1.602) ***
Summer days0.581(0.265–1.273)
Warm nights1.048(0.856–1.284)
Warm days0.801(0.63–1.017) *
Warm spell durations0.836(0.698–1.002) *
Extreme cold
Frost nights1.166(0.981–1.386) *
Ice days1.289(1.114–1.491) ***
Cool nights2.064(1.507–2.825) ***
Cool days1.315(1.061–1.631) **
Cold spell durations1.645(1.306–2.072) ***
Note: *, **, and *** indicate significance at 10%, 5%, and 1% levels, respectively.
Table
Table 3. Hazards ratios of depressive symptoms, each unit increase in low temperature and annual change in cool nights by baseline characteristics.
Table 3. Hazards ratios of depressive symptoms, each unit increase in low temperature and annual change in cool nights by baseline characteristics.
CharacteristicsSubgroupHigh TemperatureLow Temperature
Hazard Ratio
(95% CI)		p-InteractionHazard Ratio
(95% CI)		p-Interaction
SexFemale1.018(1.001,1.037)Ref.1.028(1.013,1.044)Ref.
Male1.036(1.017,1.055)0.1461.016(0.999,1.032)0.218
Age<65 yr1.024(1.008,1.039)Ref.1.021(1.008,1.034)Ref.
≥65 yr1.037(1.01,1.064)0.381.031(1.005,1.057)0.491
Urbanicity of
residence		Rural1.022(1.005,1.038)Ref.1.023(1.009,1.037)Ref.
Urban1.037(1.015,1.059)0.241.022(1.004,1.041)0.911
Household
income		Above average1.021(0.999,1.044)Ref.1.012(0.992,1.033)Ref.
Below average1.029(1.013,1.046)0.5571.027(1.013,1.041)0.197
Education
attainment		Primary school
or below		1.016(0.996,1.036)Ref.1.029(1.011,1.046)Ref.
Junior school
or above		1.035(1.018,1.053)0.1181.019(1.005,1.033)0.351
Geographic
location		Living in
north		0.862(0.703,1.058)Ref.1.031(1.018,1.044)Ref.
Living in
south		1.011(0.994,1.028)0.1181.115(1.053,1.181)0.014
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jin, J.; Xu, Z.; Cao, R.; Wang, Y.; Zeng, Q.; Pan, X.; Huang, J.; Li, G. Long-Term Apparent Temperature, Extreme Temperature Exposure, and Depressive Symptoms: A Longitudinal Study in China. Int. J. Environ. Res. Public Health 2023, 20, 3229. https://doi.org/10.3390/ijerph20043229

AMA Style

Jin J, Xu Z, Cao R, Wang Y, Zeng Q, Pan X, Huang J, Li G. Long-Term Apparent Temperature, Extreme Temperature Exposure, and Depressive Symptoms: A Longitudinal Study in China. International Journal of Environmental Research and Public Health. 2023; 20(4):3229. https://doi.org/10.3390/ijerph20043229

Chicago/Turabian Style

Jin, Jianbo, Zhihu Xu, Ru Cao, Yuxin Wang, Qiang Zeng, Xiaochuan Pan, Jing Huang, and Guoxing Li. 2023. "Long-Term Apparent Temperature, Extreme Temperature Exposure, and Depressive Symptoms: A Longitudinal Study in China" International Journal of Environmental Research and Public Health 20, no. 4: 3229. https://doi.org/10.3390/ijerph20043229

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop