Frank J. Massey

The UCLA population studies of CORD: X. A cohort study of changes in respiratory function associated with chronic exposure to SOx, NOx, and hydrocarbons

Two never-smoking cohorts in Southern California, one in Lancaster (N = 2340) exposed only to moderate levels of oxidants and the other in Long Beach (N = 1326) exposed to high levels of SOx, NO2, hydrocarbons and particulates completed spirometry and the single-breath nitrogen test five to six years apart. Forty-seven percent and 45 percent of the participants were retested. Mean results at baseline for those tested and not retested were similar. Loss to follow-up was primarily due to moving (39 percent and 47 percent). Every difference of consequence indicated greater deterioration in lung function in Long Beach. The level of significance of the difference was greatest, even in the youngest age groups, for delta N2(750-1250), suggesting that the earliest site of impairment may occur in the small airways. Greater deterioration in spirometric parameters was observed in every age group in Long Beach females above seven years of age at baseline and in Long Beach males above 15 years of age, suggesting that chronic exposure to the pollutant mix occurring in Long Beach ultimately adversely affects the large airways as well as small airways.

Electrocardiographic data recorded with Frank leads in subjects without cardiac disease and those with left ventricular overload.

Abstract Amplitudes and durations of scalar deflections were measured in Frank lead electrocardiograms from 505 men without clinical evidence of cardiac disease. Amplitudes of the major QRS deflections and of T waves were significantly correlated with age; they were also correlated to a lesser degree with chest diameter and relative body weight. The results indicate that normal limits for some Frank lead electrocardiographic variables should be stratified by age. Discrimination between normal subjects and cases with left ventricular overload was not so good with selected Frank lead variables as with standard lead electrocardiographic criteria.

Computer analysis of Frank lead electrocardiographic data recorded in an epidemiologic study

Abstract This report describes results of computer analysis of Frank lead electrocardiographic data recorded during a population study. Computer and visual measurements were compared in records from the same subjects. The prevalence of several Frank lead findings and standard lead electrocardiographic criteria were compared in subjects without clinical evidence of cardiac disease, in subjects with a clinical history of myocardial infarction or other evidence of cardiovascular abnormality, and in an unclassified group. A multivariate statistical method was used to evaluate the capability of several sets of Frank lead variables to identify records from subjects with standard lead electrocardiographic abnormalities and to discriminate between normal subjects and those with cardiovascular abnormalities. Most, but not all, records from subjects with standard lead electrocardiographic abnormalities were identified with the use of Frank lead variables. A multivariate statistic using four points during the ST-T wave may be more effective than some standard lead ST-T findings in discriminating between normal subjects and those with cardiovascular abnormalities. Other Frank lead variables did not give better discrimination between these two groups than standard lead criteria. More subjects in the unclassified group had Frank lead findings outside normal limits than had abnormalities in the standard leads; further study will be necessary to determine whether this indicates that these Frank lead variables have higher sensitivity or lower specificity for cardiac disease.

Quality Evaluation and Control Methods in Computer-Assisted Screening

A computer-assisted, chronic respiratory disease (CRD) screening project undertook a critical evaluation of the performance of the screening equipment and techniques used in administering the tests and interpreting the measurements. Based upon screening tests of 13,494 adults, the analysis led to several conclusions: the use of trained volunteers to administer the questionnaire was feasible and reproducible within the expectation of such a technique; the “nonexertional wheezing” question was particularly associated with spirometry data; the spirometry system was found to be accurate and stable under screening conditions; technician improvement was observed; the spirometry technique was modified based upon an experiment performed during the course of the screening program; and the dual positive film readings were in agreement in only a small proportion of instances, necessitating a major effort aimed to increase concordance.

Screening by Chest Photofluorography in Los Angeles

Two decades ago, when extensive tuberculosis case-finding was underway, a minifilm of the chest was a high yield mass screening test. Today readings of “suspect tuberculosis” comprise a minority of significant minifilm findings, based upon the experience of the Breathmobile Project. Both a positive minifilm and a positive smoking history function as risk factors for questionnaire and spirometry evidence of early chronic respiratory disease. Two decades ago the variability of mini-film readings was the subject of several classical studies of screening techniques. Based upon an analysis of 26,000 Breathmobile minifilms, positive discordance continues to be a prominent phenomenon, the chest physicians and radiologists favoring different reading categories. A validity study of 837 minifilms showed dual readings had an average sensitivity of 72% and a specificity of 96%, compared with single readings whose values were 52% and 98%, respectively.

Chapter 16 Discussion and conclusions

In this chapter we discuss some of the existing techniques for symmetric smoothing, as well as the various generalizations of principal components and factor analysis. We compare these techniques with the methodology developed here. The chapter concludes with a summary of the uses of principal curves and surfaces. 7.1. Alternative techniques. Other non-linear generalizations of principal components exist in the literature. They can be broadly classified according to two dichotomies. l We can estimate either the non-linear manifold or the non-linear constraint that defines the manifold. In linear principal components the approaches are equivalent. l The non-linearity can be achieved by transforming the space or by transforming the model. The principal curve and surface procedures model the non-linear manifold by transforming the model. 7.1.1. Generalized linear principal components. This approach corresponds to modeling either the nonlinear constraint or the manifold by transforming the space. The idea here is to introduce some extra variables, where each new variable is some non-linear transformation of the existing coordinates. One then seeks a subspace of this non linear coordinate system that models the data well. The subspace is found by using the usual linear eigenvector solution in the new enlarged space. This technique was first suggested by Gnanadesikan & Wilk (1966,1968), and a good description can be found in Gnanadesikan (1977). They suggested using polynomial functions of the original p coordinates. The resulting linear combinations are then of the form (for p = 2 and quadratic polynomials) 4 = aljzl + a2jz2 + WjZlZ2 + a4jZ: + asjzi (7.1)