Maria Teresa Espinosa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Interests
Maria Teresa Espinosa has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
Catharina Wesseling
Universidad Nacional, Heredia, Costa Rica
1 shared publicationHéctor Posso
Instituto Nacional de Cancerología, Santa Fe de Bogotá, Colombia
1 shared publicationJorge Chaves
Instituto Tecnológico de Costa Rica, San José, Costa Rica
1 shared publicationLuís Blanco
Universidad Nacional Autónoma, León, Nicaragua
1 shared publicationMarion Piñeros
Instituto Nacional de Cancerología, Santa Fe de Bogotá, Colombia
1 shared publicationPatricia Monge
Universidad Nacional, Heredia, Costa Rica
1 shared publication
Latest Publications
- Open Access
- 1 September 2005
- Published by FapUNIFESP (SciELO) in Revista Panamericana de Salud Pública
- Vol. 18 (3) , 187-196
Determinación del historial de exposiciones en la epidemiolgía ocupacional
In epidemiology, it is necessary that exposure indicators have good validity in order to obtain valid results when measuring the risks associated with occupational exposure to environmental noxious agents. However, ensuring the validity of past exposure data is no easy task. Because there are no environmental hygiene measures or representative levels of bioindicators signaling past exposure, self-reports have been used as a source of indirect exposure data. Unfortunately, data on specific agents are commonly poor and need to be complemented with data on the determinants of exposure. The validity of self-reports improves when certain techniques, such as control lists and icons, are employed, and the quality of individual exposure data improves when secondary data on exposure and its conditioning or determining factors are incorporated. Exposure can be determined by means of exposure matrices, assessment by experts, and exposure models, and by using a combination of primary and secondary data on exposure and its conditioning factors. Matrices contain pooled data and can thus lead to errors in classifying individual exposure and to biased risk estimates. Assessment by experts is probably the method with the highest validity, but it can become expensive when studies are large. It is also feasible to use a formal model for assessing perceivable exposures, complemented with expert assessments whenever the results of the model appear to deviate from reality.
- 1 September 2005
- Published by FapUNIFESP (SciELO) in Revista Panamericana de Salud Pública
- Vol. 18 (3) , 187-196
Open AccessDeterminación del historial de exposiciones en la epidemiolgía ocupacional
In epidemiology, it is necessary that exposure indicators have good validity in order to obtain valid results when measuring the risks associated with occupational exposure to environmental noxious agents. However, ensuring the validity of past exposure data is no easy task. Because there are no environmental hygiene measures or representative levels of bioindicators signaling past exposure, self-reports have been used as a source of indirect exposure data. Unfortunately, data on specific agents are commonly poor and need to be complemented with data on the determinants of exposure. The validity of self-reports improves when certain techniques, such as control lists and icons, are employed, and the quality of individual exposure data improves when secondary data on exposure and its conditioning or determining factors are incorporated. Exposure can be determined by means of exposure matrices, assessment by experts, and exposure models, and by using a combination of primary and secondary data on exposure and its conditioning factors. Matrices contain pooled data and can thus lead to errors in classifying individual exposure and to biased risk estimates. Assessment by experts is probably the method with the highest validity, but it can become expensive when studies are large. It is also feasible to use a formal model for assessing perceivable exposures, complemented with expert assessments whenever the results of the model appear to deviate from reality.