Título: Normalization of lung function tests for epidemiological studies
Autores: Ghezzo, Ruben Heberto.
Fecha: 1982
Publicador: McGill University - MCGILL
Fuente:
Tipo: Electronic Thesis or Dissertation
Tema: Health Sciences, Public Health.
Descripción: Environmental pollutants are becoming increasingly conspicuous in modern society. Lung function testing is one of the ways of assessing their impact on the health of the public. However lung function tests present large between individual variation, some of the determinants of which are race, sex, age and morphometric characteristics. These are generally taken into account using prediction formulae derived from the best fit equation to a set of values observed in a particular survey.
In the present study new formulae were developed which used current physiological knowledge of the determinants of lung function. These formulae together with some of the models in current use were fitted to a single training data set and the homogeneity of the fit across all ages was then compared. The models so developed, together with some of the formulae proposed in the literature were then used to predict the values of certain lung function tests in two independent populations to assess the bias of the predictions.
In the working age range (25 to 65) in males all formulae yielded comparable predictions; however in females and teenagers of both sexes the proposed model or the model proposed by Schoenberg predicted the data better.
The model proposed included a smoking module. However in the population studied ex smokers were found to present better lung function than non smokers, so that the model proposed to evaluate the effect of smoking is only applicable for current smokers. The effect of smoking was acceptably predicted for the tests derived from the forced expiratory maneuver.
One of the possible reasons why the proposed model did not fit the training data better than the empirical models may lie in the incomplete knowledge of the determinants of lung function. Further research is needed to identify such determinants and to find an objective and simple methodology for their evaluation.
Idioma: en