Sökning: id:"swepub:oai:DiVA.org:uu-420342" >
Analytical approach...
-
Dzhambov, Angel M.Med Univ Plovdiv, Fac Publ Hlth, Dept Hyg & Ecomed, Plovdiv, Bulgaria.
(författare)
Analytical approaches to testing pathways linking greenspace to health : A scoping review of the empirical literature
- Artikel/kapitelEngelska2020
Förlag, utgivningsår, omfång ...
-
Elsevier BV,2020
-
printrdacarrier
Nummerbeteckningar
-
LIBRIS-ID:oai:DiVA.org:uu-420342
-
https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-420342URI
-
https://doi.org/10.1016/j.envres.2020.109613DOI
Kompletterande språkuppgifter
-
Språk:engelska
-
Sammanfattning på:engelska
Ingår i deldatabas
Klassifikation
-
Ämneskategori:ref swepub-contenttype
-
Ämneskategori:for swepub-publicationtype
Anmärkningar
-
Background: Inadequate translation from theoretical to statistical models of the greenspace - health relationship may lead to incorrect conclusions about the importance of some pathways, which in turn may reduce the effectiveness of public health interventions involving urban greening. In this scoping review we aimed to: (1) summarize the general characteristics of approaches to intervening variable inference (mediation analysis) employed in epidemiological research in the field; (2) identify potential threats to the validity of findings; and (3) propose recommendations for planning, conducting, and reporting mediation analyses.Methods: We conducted a scoping review, searching PubMed, Scopus, and Web of Science for peer-reviewed epidemiological studies published by December 31, 2019. The list of potential studies was continuously updated through other sources until March 2020. Narrative presentation of the results was coupled with descriptive summary of study characteristics.Results: We found 106 studies, most of which were cross-sectional in design. Most studies only had a spatial measure of greenspace. Mental health/well-being was the most commonly studied outcome, and physical activity and air pollution were the most commonly tested intervening variables. Most studies only conducted single mediation analysis, even when multiple potentially intertwined mediators were measured. The analytical approaches used were causal steps, difference-of-coefficients, product-of-coefficients, counterfactual framework, and structural equation modelling (SEM). Bootstrapping was the most commonly used method to construct the 95% CI of the indirect effect. The product-of-coefficients method and SEM as used to investigate serial mediation components were more likely to yield findings of indirect effect. In some cases, the causal steps approach thwarted tests of indirect effect, even though both links in an indirect effect were supported. In most studies, sensitivity analyses and proper methodological discussion of the modelling approach were missing.Conclusions: We found a persistent pattern of suboptimal conduct and reporting of mediation analysis in epidemiological studies investigating pathways linking greenspace to health; however, recent years have seen improvements in these respects. Better planning, conduct, and reporting of mediation analyses are warranted.
Ämnesord och genrebeteckningar
Biuppslag (personer, institutioner, konferenser, titlar ...)
-
Browning, Matthew H. E. M.Clemson Univ, Dept Pk Recreat & Tourism Management, Clemson, SC USA.
(författare)
-
Markevych, IanaJagiellonian Univ, Inst Psychol, Krakow, Poland.
(författare)
-
Hartig, Terry,1959-Uppsala universitet,Institutet för bostads- och urbanforskning (IBF)(Swepub:uu)terrhart
(författare)
-
Lercher, PeterGraz Univ Technol, Inst Highway Engn & Transport Planning, Graz, Austria.
(författare)
-
Med Univ Plovdiv, Fac Publ Hlth, Dept Hyg & Ecomed, Plovdiv, Bulgaria.Clemson Univ, Dept Pk Recreat & Tourism Management, Clemson, SC USA.
(creator_code:org_t)
Sammanhörande titlar
-
Ingår i:Environmental Research: Elsevier BV1860013-93511096-0953
Internetlänk
Hitta via bibliotek
Till lärosätets databas