Department of Global Health, Georgetown University, Washington, DC, USA.
Health, Nutrition and Population Global Practice, The World Bank, Dhaka, Bangladesh.
Lancet Planet Health. 2023 Feb;7(2):e137-e146. doi: 10.1016/S2542-5196(22)00315-1.
Climate change has major implications for common mental disorders including depression and anxiety in vulnerable nations such as Bangladesh. However, knowledge gaps exist around national estimations of depression and anxiety, and the associations between the prevalence of these disorders with climate-related and sociodemographic risk factors. To address these gaps, this study analysed data from a nationally representative panel study in Bangladesh that examined climate-related and sociodemographic correlates of depression and anxiety.
Two rounds of nationally representative household panel data were collected from urban and rural areas between August and September, 2019, and January and February, 2020. Households were selected for inclusion across 150 enumeration areas as the primary sampling units with use of a two-stage stratified random sampling design, and survey instruments were administered to the available adult member of the household. Depression and anxiety were measured with the Patient Health Questionnaire-9 and Generalized Anxiety Disorder-7 scales, respectively, and weighted prevalence estimates were calculated on the basis of the 2011 national population census. Data on temperature and humidity were collected from 43 weather stations and constructed as mean values for the 2-month period preceding each round of the survey. Self-reported exposure to flooding was collected for a 12-month recall period. We applied a weighted population average logistic model on the pooled sample of both surveys to analyse the associations between ambient temperature, humidity, exposure to flooding, seasonality, sociodemographic variables, and three outcome conditions (depression, anxiety, and co-occurring depression and anxiety; at the level of p<0·1). The models accounted for temporal and spatial heterogeneity. Standard errors were clustered at the level of each primary sampling unit.
3606 individuals were included with 3·5% dropout in the second survey round (pooled sample n=7086; age range 15-90 years; 2898 [40·9%] men and 4188 [59·1%] women). National weighted prevalence estimates were 16·3% (95% CI 14·7-17·8) for depression, 6·0% (4·7-7·3) for anxiety, and 4·8% (3·7-5·9) for co-occurring depression and anxiety. We observed no significant associations between overall seasonality (summer vs winter) and the odds of depression (adjusted odds ratio 3·14 [95% CI 0·52-19·13], p=0·22), anxiety (0·16 [0·02-1·41], p=0·10), or co-occurring depression and anxiety (0·13 [0·01-1·49], p=0·10). An increase in mean temperature of 1°C within the 2 months preceding the surveys was associated with increased odds of anxiety (1·21 [1·00-1·47], p=0·046) and increased odds of co-occurring depression and anxiety (1·24 [1·00-1·53], p=0·045), whereas increased temperature was not associated with depression (0·90 [0·77-1·04], p=0·15). An increase in mean humidity of 1 g/m was not associated with depression (0·99 [0·96-1·02], p=0·60) or anxiety (1·04 [0·99-1·09], p=0·13), but was associated with co-occurring depression and anxiety (1·06 [1·00-1·12], p=0·064). Exposure to flooding within the 12 months preceding the survey rounds was associated with increased odds of all outcome conditions (depression, 1·31 [1·00-1·70], p=0·047; anxiety, 1·69 [1·21-2·36], p=0·0020; and co-occurring depression and anxiety, 1·87 [1·31-2·68], p=0·0006).
Climate-related shocks and other stressors have an important association with the burden of depression and anxiety in Bangladesh. Community-level interventions for common mental disorders need to be developed and assessed for safety, feasibility, and effectiveness in a Bangladeshi context. Further research on climate-related stressors is needed over different timespans and time intervals.
The World Bank.
气候变化对脆弱国家（如孟加拉国）的常见精神障碍（包括抑郁和焦虑）有重大影响。然而，在国家对抑郁和焦虑的估计方面，以及这些疾病的流行与气候相关和社会人口学风险因素之间的关联方面，仍存在知识空白。为了解决这些差距，本研究分析了来自孟加拉国全国代表性的面板研究的数据，该研究调查了抑郁和焦虑与气候相关和社会人口学的相关性。
2019 年 8 月至 9 月和 2020 年 1 月至 2 月期间，在城市和农村地区进行了两轮全国代表性的家庭面板数据收集。使用两阶段分层随机抽样设计，从 150 个普查区中选择纳入家庭作为主要抽样单位，并向家庭中可获得的成年成员发放调查工具。使用患者健康问卷-9 量表和广义焦虑症-7 量表分别测量抑郁和焦虑，根据 2011 年全国人口普查计算加权患病率估计值。从 43 个气象站收集温度和湿度数据，并构建为调查前两个月的平均值。收集了为期 12 个月的洪水暴露情况的回顾性数据。我们对两轮调查的汇总样本应用加权人口平均逻辑模型，分析环境温度、湿度、洪水暴露、季节性、社会人口统计学变量与三种结果情况（抑郁、焦虑和同时存在的抑郁和焦虑；在 p<0·1 的水平）之间的关联。该模型考虑了时间和空间的异质性。标准误差按每个主要抽样单位进行聚类。
纳入了 3606 人，第二轮调查中有 3·5%的人中途退出（汇总样本 n=7086；年龄范围为 15-90 岁；2898 名男性[40·9%]和 4188 名女性[59·1%]）。全国加权患病率估计值分别为抑郁 16·3%（95%CI 14·7-17·8）、焦虑 6·0%（4·7-7·3）和同时存在的抑郁和焦虑 4·8%（3·7-5·9）。我们没有观察到总体季节性（夏季与冬季）与抑郁（调整后的优势比 3·14[95%CI 0·52-19·13]，p=0·22）、焦虑（0·16[0·02-1·41]，p=0·10）或同时存在的抑郁和焦虑（0·13[0·01-1·49]，p=0·10）之间的显著关联。调查前两个月内平均温度升高 1°C，与焦虑（1·21[1·00-1·47]，p=0·046）和同时存在的抑郁和焦虑（1·24[1·00-1·53]，p=0·045）的发病几率增加有关，而温度升高与抑郁（0·90[0·77-1·04]，p=0·15）无关。平均湿度增加 1 g/m 与抑郁（0·99[0·96-1·02]，p=0·60）或焦虑（1·04[0·99-1·09]，p=0·13）无关，但与同时存在的抑郁和焦虑有关（1·06[1·00-1·12]，p=0·064）。在调查前 12 个月内暴露于洪水与所有结果情况（抑郁，1·31[1·00-1·70]，p=0·047；焦虑，1·69[1·21-2·36]，p=0·0020；和同时存在的抑郁和焦虑，1·87[1·31-2·68]，p=0·0006）的发病几率增加有关。
气候相关的冲击和其他压力源与孟加拉国的抑郁和焦虑负担有重要关联。需要在孟加拉国的背景下制定和评估针对常见精神障碍的社区一级干预措施，以确保其安全性、可行性和有效性。需要进一步研究不同时间跨度和时间间隔的气候相关压力源。
世界银行。
