Germán Federico Rosati1,2, Tomás Alberto Olego3,4, H Antonio Vazquez Brust3. 1. National Council of Scientific and Technical Research, CONICET, Godoy Cruz 2290, Buenos Aires, C1425FQB, Argentina. grosati@unsam.edu.ar. 2. Institute for High Social Studies, University of San Martín, Av. 25 de Mayo 1021, San Martín, B1650HMI, Argentina. grosati@unsam.edu.ar. 3. Bunge & Born Foundation, 25 de Mayo 501 6∘ Piso, Buenos Aires, C1002ABK, Argentina. 4. Faculty of Social Science, University of Buenos Aires, Santiago del Estero y Carlos Calvo, Buenos Aires, C1075AAU, Argentina.
Abstract
BACKGROUND: Designing public health policies to target the needs of specific places requires highly granular data. When geographic health statistics from official sources are absent or lacking in spatial detail, Sanitary Vulnerability metrics derived from Census and other georeferenced public data can be used to identify areas in particular need of attention. With that aim, a Vulnerability Map was developed, identifying areas with a substantial deficit in its population health coverage. As a result a novel methodology for measuring Sanitary Vulnerability is presented, that can potentially be applied to different time periods or geographies. METHODS: Census, official listings of public health facilities and crowdsourced georeferenced data are used. The Vulnerability Index is built using dimensionality reduction techniques such as Autoencoders and Non-parametric PCA. MAIN RESULTS: The high resolution map shows the geographical distribution of a Sanitary Vulnerability Index, produced using official and crowdsourced open data sources, overcoming the lack of official sources on health indicators at the local level. CONCLUSIONS: The Sanitary Vulnerability Map's value as a tool for place specific policymaking was validated by using it to predict local health related metrics such as health coverage. Further lines of work contemplate using the Map to study the interaction between Sanitary Vulnerability and the prevalence of different diseases, and also applying its methodology in the context of other public services such as education, security, housing, etc.
BACKGROUND: Designing public health policies to target the needs of specific places requires highly granular data. When geographic health statistics from official sources are absent or lacking in spatial detail, Sanitary Vulnerability metrics derived from Census and other georeferenced public data can be used to identify areas in particular need of attention. With that aim, a Vulnerability Map was developed, identifying areas with a substantial deficit in its population health coverage. As a result a novel methodology for measuring Sanitary Vulnerability is presented, that can potentially be applied to different time periods or geographies. METHODS: Census, official listings of public health facilities and crowdsourced georeferenced data are used. The Vulnerability Index is built using dimensionality reduction techniques such as Autoencoders and Non-parametric PCA. MAIN RESULTS: The high resolution map shows the geographical distribution of a Sanitary Vulnerability Index, produced using official and crowdsourced open data sources, overcoming the lack of official sources on health indicators at the local level. CONCLUSIONS: The Sanitary Vulnerability Map's value as a tool for place specific policymaking was validated by using it to predict local health related metrics such as health coverage. Further lines of work contemplate using the Map to study the interaction between Sanitary Vulnerability and the prevalence of different diseases, and also applying its methodology in the context of other public services such as education, security, housing, etc.
Entities:
Keywords:
Article; Census; Health; Open data; Public policy; Sample; Vulnerability
Authors: D J Weiss; A Nelson; H S Gibson; W Temperley; S Peedell; A Lieber; M Hancher; E Poyart; S Belchior; N Fullman; B Mappin; U Dalrymple; J Rozier; T C D Lucas; R E Howes; L S Tusting; S Y Kang; E Cameron; D Bisanzio; K E Battle; S Bhatt; P W Gething Journal: Nature Date: 2018-01-10 Impact factor: 49.962
Authors: Sabrina Juran; P Niclas Broer; Stefanie J Klug; Rachel C Snow; Emelda A Okiro; Paul O Ouma; Robert W Snow; Andrew J Tatem; John G Meara; Victor A Alegana Journal: BMJ Glob Health Date: 2018-08-16