Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data

Detalhes bibliográficos
Autor(a) principal: Adrega, Tiago
Data de Publicação: 2021
Outros Autores: Rocha, Anabela, Miranda, M. Cristina
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10773/32821
Resumo: With observational data, an important step of the research process is skipped, resulting in some restrictions to make inferences concerning the treatment effects. Some methodologies have been developed in order to reduce the imbalance in the samples of treated and control units. Propensity Score Matching (PSM) is still one of the most common approaches applied but coarsened Exact Matching (CEM) appears to produce better results, most of the times in which it is used. This work illustrates the application of each of the two techniques to a set of data from the Nepal population. Our aim is to compare the two methodologies and evaluate in what way their use adds information about the prevalence of Cardio Vascular Disease (CVD) risk. Data refers to a remote village population that was separated into two groups after the incidents of the May 2015 earthquake. The study was carried out during a humanitarian mission in Nepal, aimed to provide medical care to the people of Sindhupalchok, a northern Nepalese region, with approximately 1200inhabitants. With the seismic event, this population got separated into groups of dislodged individuals: victims that stayed nearby village areas and those who went towards Kathmandu looking for support in temporary settlements. Both these populations were supported by the medical mission. Cross-sectional data were collected approximately 18 months after the earthquake and included demographic data, anthropometric data, previous medical history, CVD risk factors, and health behaviors. The assessment of CVD risk factors and health behaviours was based on a question-by-question guide provided by the WHO. In order to compare both approaches, we computed two imbalance measures, L1 and Percent Bias Reduction (PBR). The results show that CEM dominates PSM. From the application of the two approaches, we find that the results are generally in agreement but CEM methodology allowed to highlight some data features not seen before with PSM.
id RCAP_d943234279fe33e4a337667e62d07955
oai_identifier_str oai:ria.ua.pt:10773/32821
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational dataCardiovascular disease risk factorsPropensity score matchingHealth behavioursCoarsened exact matchingObservational studiesWith observational data, an important step of the research process is skipped, resulting in some restrictions to make inferences concerning the treatment effects. Some methodologies have been developed in order to reduce the imbalance in the samples of treated and control units. Propensity Score Matching (PSM) is still one of the most common approaches applied but coarsened Exact Matching (CEM) appears to produce better results, most of the times in which it is used. This work illustrates the application of each of the two techniques to a set of data from the Nepal population. Our aim is to compare the two methodologies and evaluate in what way their use adds information about the prevalence of Cardio Vascular Disease (CVD) risk. Data refers to a remote village population that was separated into two groups after the incidents of the May 2015 earthquake. The study was carried out during a humanitarian mission in Nepal, aimed to provide medical care to the people of Sindhupalchok, a northern Nepalese region, with approximately 1200inhabitants. With the seismic event, this population got separated into groups of dislodged individuals: victims that stayed nearby village areas and those who went towards Kathmandu looking for support in temporary settlements. Both these populations were supported by the medical mission. Cross-sectional data were collected approximately 18 months after the earthquake and included demographic data, anthropometric data, previous medical history, CVD risk factors, and health behaviors. The assessment of CVD risk factors and health behaviours was based on a question-by-question guide provided by the WHO. In order to compare both approaches, we computed two imbalance measures, L1 and Percent Bias Reduction (PBR). The results show that CEM dominates PSM. From the application of the two approaches, we find that the results are generally in agreement but CEM methodology allowed to highlight some data features not seen before with PSM.Universidade de Aveiro; Centro Hospitalar do Baixo Vouga2022-01-07T11:11:44Z2021-12-14T00:00:00Z2021-12-14info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10773/32821eng10.34624/jshd.v3i2.24256Adrega, TiagoRocha, AnabelaMiranda, M. Cristinainfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-02-22T12:03:15Zoai:ria.ua.pt:10773/32821Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T03:04:23.500212Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
title Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
spellingShingle Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
Adrega, Tiago
Cardiovascular disease risk factors
Propensity score matching
Health behaviours
Coarsened exact matching
Observational studies
title_short Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
title_full Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
title_fullStr Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
title_full_unstemmed Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
title_sort Evaluation of cardiovascular disease risk factors with Propensity Score Matching and Coarsened Exact Matching: Nepalese post-seismic observational data
author Adrega, Tiago
author_facet Adrega, Tiago
Rocha, Anabela
Miranda, M. Cristina
author_role author
author2 Rocha, Anabela
Miranda, M. Cristina
author2_role author
author
dc.contributor.author.fl_str_mv Adrega, Tiago
Rocha, Anabela
Miranda, M. Cristina
dc.subject.por.fl_str_mv Cardiovascular disease risk factors
Propensity score matching
Health behaviours
Coarsened exact matching
Observational studies
topic Cardiovascular disease risk factors
Propensity score matching
Health behaviours
Coarsened exact matching
Observational studies
description With observational data, an important step of the research process is skipped, resulting in some restrictions to make inferences concerning the treatment effects. Some methodologies have been developed in order to reduce the imbalance in the samples of treated and control units. Propensity Score Matching (PSM) is still one of the most common approaches applied but coarsened Exact Matching (CEM) appears to produce better results, most of the times in which it is used. This work illustrates the application of each of the two techniques to a set of data from the Nepal population. Our aim is to compare the two methodologies and evaluate in what way their use adds information about the prevalence of Cardio Vascular Disease (CVD) risk. Data refers to a remote village population that was separated into two groups after the incidents of the May 2015 earthquake. The study was carried out during a humanitarian mission in Nepal, aimed to provide medical care to the people of Sindhupalchok, a northern Nepalese region, with approximately 1200inhabitants. With the seismic event, this population got separated into groups of dislodged individuals: victims that stayed nearby village areas and those who went towards Kathmandu looking for support in temporary settlements. Both these populations were supported by the medical mission. Cross-sectional data were collected approximately 18 months after the earthquake and included demographic data, anthropometric data, previous medical history, CVD risk factors, and health behaviors. The assessment of CVD risk factors and health behaviours was based on a question-by-question guide provided by the WHO. In order to compare both approaches, we computed two imbalance measures, L1 and Percent Bias Reduction (PBR). The results show that CEM dominates PSM. From the application of the two approaches, we find that the results are generally in agreement but CEM methodology allowed to highlight some data features not seen before with PSM.
publishDate 2021
dc.date.none.fl_str_mv 2021-12-14T00:00:00Z
2021-12-14
2022-01-07T11:11:44Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10773/32821
url http://hdl.handle.net/10773/32821
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv 10.34624/jshd.v3i2.24256
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade de Aveiro; Centro Hospitalar do Baixo Vouga
publisher.none.fl_str_mv Universidade de Aveiro; Centro Hospitalar do Baixo Vouga
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799137699526868992