State Key Laboratory of Quality Research in Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Macau, China.
Department of Public Health and Medicinal Administration, Faculty of Health Sciences, University of Macau, Macau, China.
Front Public Health. 2022 Dec 14;10:953695. doi: 10.3389/fpubh.2022.953695. eCollection 2022.
The maldistributions of the health workforce showed great inconsistency when singly measured by population quantity or geographic area in China. Meanwhile, earlier studies mainly employed traditional econometric approaches to investigate determinants for the health workforce, which ignored spillover effects of influential factors on neighboring regions. Therefore, we aimed to analyze health workforce allocation in China from demographic and geographic perspectives simultaneously and then explore the spatial pattern and determinants for health workforce allocation taking account of the spillover effect.
The health resource density index (HRDI) equals the geometric mean of health resources per 1,000 persons and per square kilometer. First, the HRDI of licensed physicians (HRDI_P) and registered nurses (HRDI_N) was calculated for descriptive analysis. Then, global and local Moran's I indices were employed to explore the spatial features and aggregation clusters of the health workforce. Finally, four types of independent variables were selected: supportive resources (bed density and government health expenditure), healthcare need (proportion of the elderly population), socioeconomic factors (urbanization rate and GDP per capita), and sociocultural factors (education expenditure per pupil and park green area per capita), and then the spatial panel econometric model was used to assess direct associations and intra-region spillover effects between independent variables and HRDI_P and HRDI_N.
Global Moran's I index of HRDI_P and HRDI_N increased from 0.2136 ( = 0.0070) to 0.2316 ( = 0.0050), and from 0.1645 ( = 0.0120) to 0.2022 ( = 0.0080), respectively. Local Moran's I suggested spatial aggregation clusters of HRDI_P and HRDI_N. For HRDI_P, bed density, government health expenditure, and GDP had significantly positive associations with local HRDI_P, while the proportion of the elderly population and education expenditure showed opposite spillover effects. More precisely, a 1% increase in the proportion of the elderly population would lead to a 0.4098% increase in HRDI_P of neighboring provinces, while a 1% increase in education expenditure leads to a 0.2688% decline in neighboring HRDI_P. For HRDI_N, the urbanization rate, bed density, and government health expenditure exerted significantly positive impacted local HRDI_N. In addition, the spillover effect was more evident in the urbanization rate, with a 1% increase in the urbanization rate relating to 0.9080% growth of HRDI_N of surrounding provinces. Negative spillover effects of education expenditure, government health expenditure, and elderly proportion were observed in neighboring HRDI_N.
There were substantial spatial disparities in health workforce distribution in China; moreover, the health workforce showed positive spatial agglomeration with a strengthening tendency in the last decade. In addition, supportive resources, healthcare needs, and socioeconomic and sociocultural factors would affect the health labor configuration not only in a given province but also in its nearby provinces.
在中国，仅通过人口数量或地理区域来衡量卫生人力的分布情况存在很大的不一致性。同时，早期的研究主要采用传统的计量经济学方法来研究卫生人力的决定因素，而忽略了影响因素对邻近地区的溢出效应。因此，我们旨在同时从人口统计学和地理角度分析中国的卫生人力配置，并探讨考虑到溢出效应的卫生人力配置的空间格局和决定因素。
卫生资源密度指数（HRDI）等于每 1000 人及每平方公里的卫生资源几何平均值。首先，计算执业医师（HRDI_P）和注册护士（HRDI_N）的 HRDI 进行描述性分析。然后，采用全局和局部 Moran's I 指数来探索卫生人力的空间特征和集聚群。最后，选择了四类独立变量：支持性资源（床位数和政府卫生支出）、医疗保健需求（老年人口比例）、社会经济因素（城市化率和人均国内生产总值）和社会文化因素（生均教育支出和人均公园绿地面积），然后采用空间面板计量经济学模型评估独立变量与 HRDI_P 和 HRDI_N 之间的直接关联和区域内溢出效应。
HRDI_P 和 HRDI_N 的全局 Moran's I 指数从 0.2136（=0.0070）增加到 0.2316（=0.0050）和从 0.1645（=0.0120）增加到 0.2022（=0.0080）。局部 Moran's I 表明 HRDI_P 和 HRDI_N 存在空间集聚群。对于 HRDI_P，床位数、政府卫生支出和人均国内生产总值与当地 HRDI_P 呈显著正相关，而老年人口比例和教育支出则表现出相反的溢出效应。更确切地说，老年人口比例每增加 1%，邻近省份的 HRDI_P 将增加 0.4098%，而教育支出每增加 1%，邻近的 HRDI_P 将减少 0.2688%。对于 HRDI_N，城市化率、床位数和政府卫生支出对当地 HRDI_N 产生显著的正向影响。此外，城市化率的溢出效应更为明显，城市化率每增加 1%，周边省份的 HRDI_N 将增长 0.9080%。教育支出、政府卫生支出和老年人口比例的负溢出效应在邻近的 HRDI_N 中观察到。
中国卫生人力分布存在较大的空间差异；此外，在过去十年中，卫生人力呈正空间集聚，且集聚趋势增强。此外，支持性资源、医疗保健需求以及社会经济和社会文化因素不仅会影响到特定省份的卫生劳动力配置，还会影响到其邻近省份的卫生劳动力配置。
