School of Public Health, Ningxia Medical University, Yinchuan, China.
Key Laboratory of Environmental Factors and Chronic Disease Control, Yinchuan, China.
BMC Public Health. 2024 Feb 14;24(1):459. doi: 10.1186/s12889-024-18035-6.
Although China has eliminated absolute poverty, the effects of sickness still pose a threat to the prospect of returning to poverty in western rural areas. However, poverty governance extends beyond solving absolute poverty, and should enhance the family's ability to resist risks, proactively identify the existence of risks, and facilitate preventive measures to reduce the probability of falling into poverty again. This study aimed to assess the health poverty vulnerability of rural households in western China and decompose its determinants.
Based on survey data from 2022, the three-stage feasible generalized least squares method was used to calculate the health poverty vulnerability index. Then, Anderson's health behavior theory model was extended to analyse various influencing factors using binary logistic regression, and the contribution of each influencing factor was decomposed using the Shapley index. Finally, Tobit regression and the censored least absolute deviations estimation (clad) method were used to test the model's robustness.
A total of 5455 families in the rural Ningxia region of western China were included in the study. The health poverty vulnerability index of the sample population in 2022 was 0.3000 ± 0.2223, and families with vulnerability ≥0.5 accounted for 16.9% of the sample population. From the Anderson behavioral model, the three models including propensity, enabling, and demand factors had the best fit, and the AIC and BIC values were the smallest. The Shapley decomposition showed that the dimensions of the propensity factor, number of residents, age and educational level of the household head, and dependency ratio were the most important factors influencing vulnerability to health poverty. Tobit regression and the clad method proved the reliability of the constructed model through a robustness test.
Rural areas still face the risk of becoming poor or falling into poverty owing to residents' health problems. Health poverty alleviation should gradually change from a focus on treatment to prevention, and formulate a set of accurate and efficient intervention policies from a forward-looking perspective to consolidate the results of health poverty alleviation and prevent widescale poverty return.
虽然中国已经消除了绝对贫困，但疾病的影响仍然对西部农村地区返贫的前景构成威胁。然而，贫困治理不仅仅是解决绝对贫困问题，还应增强家庭的抗风险能力，主动识别风险的存在，并采取预防措施，降低再次陷入贫困的概率。本研究旨在评估中国西部农村家庭的健康贫困脆弱性，并分解其决定因素。
基于 2022 年的调查数据，采用三阶段可行广义最小二乘法计算健康贫困脆弱性指数。然后，扩展了安德森健康行为理论模型，使用二元逻辑回归分析各种影响因素，并使用 Shapley 指数分解各影响因素的贡献。最后，使用 Tobit 回归和截尾最小绝对偏差估计（clad）方法检验模型的稳健性。
共纳入中国西部宁夏农村地区的 5455 户家庭。2022 年样本人口的健康贫困脆弱性指数为 0.3000±0.2223，脆弱性≥0.5 的家庭占样本人口的 16.9%。从安德森行为模型来看，包含倾向、赋权和需求因素的三个模型拟合度最好，AIC 和 BIC 值最小。Shapley 分解表明，倾向因素的维度、居民人数、家庭户主年龄和教育程度以及抚养比是影响健康贫困脆弱性的最重要因素。Tobit 回归和 clad 方法通过稳健性检验证明了所构建模型的可靠性。
农村地区仍然面临着因居民健康问题而陷入贫困或返贫的风险。健康扶贫应逐步从注重治疗转向预防，并从前瞻性角度制定一套准确、高效的干预政策，巩固健康扶贫成果，防止大规模返贫。
