BackgroundTraditionally, poverty assessment has relied on a single income criterion, which is inadequate in contexts where absolute poverty has been mitigated. This study developed a weighting scheme for the Multidimensional Health Poverty Index (MHPI) based on sustainable livelihood framework (SLF), aimed at enhancing the accuracy of poverty identification in China.MethodsA two-round Delphi process was used to determine the indicators of the index system. The weight of each indicator was calculated using the analytic hierarchy process (AHP) method. The multidimensional poverty index incorporates SLF’s five dimensions: financial, natural, physical, social, and human capital.ResultsThe initial 46 indicators formed an indicator pool for the Delphi questionnaire. Based on the final consensus of the expert panel, the Delphi consultation resulted in an index system comprising six first-level and 23 second-level indicators. The weight values of the first-level indicators (economic security, health status, education, social capital, healthcare utilisation, and living conditions) were 0.2715, 0.2593, 0.0855, 0.0657, 0.1812, and 0.1363, respectively.ConclusionThis study established a scientific and effective index to evaluate generate weight for estimating multidimensional health poverty in China. Economic security, health status, and healthcare utilisation are the most crucial aspects of multidimensional health poverty. Moreover, the results indicated that vocational training and social capital should be emphasised.