PubMed Health⌕ Search

Biomedical subjects

B G Armstrong

Publications and source records attributed to B G Armstrong.

At least 19 recordsLinked to original sources
PubMed1995

Comparing standardized mortality ratios.

It is well known that the ratio of two standardized mortality ratios (SMRs) is not in general an unbiased estimate of the underlying within-stratum ratio of rates of one subcohort relative to another. It is also established, although less well known, that a sufficient condition for unbiasedness is that the underlying stratum-specific rates in each of the two subcohorts be proportional to the reference population. Further, the ratio of SMRs is more precise than the wholly internal (Poisson regression) estimate of rate ratio. In data that are compatible with the proportionality assumption, use of the ratio of SMRs thus buys precision at the cost of increased vulnerability to bias. To further elucidate choice between methods, we derive expressions for the asymptotic precision of each. These show that improved precision of ratio of SMRs depends on the extent to which the distribution of expected deaths over strata is different in the two cohorts, or equivalently, on the variance over strata of the proportion of expected deaths falling in the first cohort. The results are illustrated by hypothetic examples.

Bias↗
PubMed1992

Cigarette, alcohol, and coffee consumption and spontaneous abortion.

We analyzed data from a survey of occupational factors and pregnancy outcome to examine the effects of cigarette, alcohol, and coffee consumption on pregnancy outcome. Clear and statistically significant associations were found between cigarette and alcohol consumption and spontaneous abortion. There was a weaker but statistically significant association with coffee consumption: If the associations were casual, 11% of the spontaneous abortions could be attributed to smoking, 5% to alcohol, and 2% to coffee.

Abortion, Spontaneous↗
PubMed1992

Cigarette, alcohol, and coffee consumption and prematurity.

We analyzed data from a survey of occupational and other factors in pregnancy to assess the effects of cigarette, alcohol, and coffee consumption on pregnancy outcome. The risk of low birth weight for gestational age was found to increase substantially with smoking. Occasional consumers of alcohol had a slightly reduced risk relative to total abstainers. In more frequent drinkers, there was a small increase in risk. Risk increased slightly with coffee consumption.

Adult↗
PubMed1992

Cigarette, alcohol, and coffee consumption and congenital defects.

We analyzed data from a survey of occupational and other factors and pregnancy outcome to assess the effects of cigarette, alcohol, and coffee consumption. There was no evidence of an association between any congenital defect and smoking. Results for alcohol and coffee consumption were largely negative, but there was a weak association with musculoskeletal defects in babies born to women who drank one or more alcoholic drinks a day.

Alcohol Drinking↗
PubMed1991

Comparison of French and English versions of the American Thoracic Society respiratory questionnaire in a bilingual working population.

Standardized French and English versions of the American Thoracic Society (ATS) respiratory disease questionnaire were administered to 204 English-speaking and 406 French-speaking male blue-collar aviation workers unexposed to occupational respiratory hazards. After adjusting for smoking status, age, years of education, foreign birth and maternal language other than French or English, no significant differences between the two questionnaires were found for response rates to usual cough, usual phlegm, mild or moderate dyspnoea, and chronic bronchitis. French-speaking workers reported significantly less wheeze with colds (OR = 0.60, p less than 0.02) and wheeze apart from colds (OR = 0.55, p less than 0.05) than the English-speaking group, but, the occurrence of wheeze on most days or nights was similar for both groups (OR = 1.02, NS). For 66 bilingual workers who completed both French and English questionnaires at a time interval of approximately two months, highly consistent results were found for sociodemographic data, smoking habits, cough, phlegm, breathlessness and chronic bronchitis, but not for wheeze with or apart from colds (agreement less than 90%; Kappa less than 0.50). These results reflect the difficulties in translating the concept of 'wheeze' from English to French. We conclude that most symptoms elicited by the French questionnaire may be generalized to English-speaking populations, but that questions pertaining to wheeze on most days or nights may be preferable to other questions concerning wheeze.

Data Collection↗
PubMed1990

Comparison of indices of ambient exposure to 60-hertz electric and magnetic fields.

Occupational, environmental, or domestic exposure of human beings to extremely low-frequency (50- or 60-Hz) electric and magnetic fields varies continuously over time. In epidemiological studies of possible health effects, exposures over long durations must be aggregated in terms of simple summary indices. However, there are many different, biologically plausible, ways of aggregating the data. While awake, each of 20 electric utility personnel and 16 office workers had provided minute-by-minute measures of incident electric (V/m) and magnetic (muT) fields over a 7-day period via personal dosimeters. Once the measures were aggregated as means, medians, peaks, and other indices, intercorrelations between all index pairs were calculated; correlation matrices are presented for the utility and office workers both by group and when pooled. Product-moment coefficients (r) greater that .80 were found between the time-weighted arithmetic mean (TWA) and indices that explicitly emphasize short but highly intense exposures, such as peak values and time above thresholds. Medians and geometric means were less highly correlated with the TWA. Use of only a few indices, perhaps the TWA alone, may sacrifice but little statistical power in most epidemiological studies of utility workers exposed to ELF fields. However, correlations between electric-field strength and magnetic-field density were generally quite weak, as were correlations of either with high-frequency transients; these findings underscore the need to measure each of these variables in epidemiological studies. Indices of exposure incurred outside the workplace were less strongly correlated, which may indicate the need to use several indices in general-population studies.

Electromagnetic Fields↗
PubMed1990

The effects of measurement errors on relative risk regressions.

This paper concerns the effects of random error in numerical measurements of risk factors (covariates) in relative risk regressions. When not dependent on outcome (nondifferential), such error usually attenuates relative risk estimates (shifts them toward one) and leads to spuriously narrow confidence intervals. The presence of measurement error also reduces precision of estimates and power of significance tests. However, significance levels obtained by using the approximate measurements are usually valid and as powerful as possible given the measurement error. The attenuation in risk estimate depends not only on the size (variance) of the measurement error, but also on its distributional form, on whether it is dependent on the true level of the risk factor (whether it is of "Berkson" type), on the variance and distributional form of true levels of the risk factor, on the functional form of the regression (exponential or linear), and on the confounding variables included in the model. Error in measuring confounding variables leads to loss of control of confounding, leaving residual bias. Uncomplicated techniques of correcting the effects of measurement error in simple models in which distributions are assumed normal are available in the statistical literature. For these corrections, information on measurement error variance is required. Some approaches appropriate for more general models have been proposed, but these appear to be insufficiently developed for routine application.

Bias↗
PubMed1990

Bladder cancer screening among primary aluminum production workers in Quebec.

We evaluated a cytology screening program, offered by a large aluminum producer after the discovery of an excess of bladder cancer due to occupational exposure to coal-tar-pitch volatiles, in terms of early detection and survival, based on information in the public domain. From January 1970 through June 1986, 79 cases of bladder cancer were identified in this cohort of aluminum workers aged 65 years or younger. By the end of 1986, 36 had died, with bladder cancer as the primary cause of death for 53%. Cases diagnosed after the screening program was introduced in 1980 were compared with those diagnosed earlier. In cases diagnosed after 1980, the proportion identified at early stages was higher (77% v 67%) and survival seemed improved but the differences were not statistically significant. Although these results do not encourage an optimistic view of screening effectiveness in this population, the limits inherent in the present study make it impossible to draw any firm conclusion. Studies restricted to public domain information do not appear to have sufficient data to evaluate workplace screening programs.

Adult↗
PubMed1989

Analysis of case-control data with covariate measurement error: application to diet and colon cancer.

We propose a method for estimating odds ratios from case-control data in which covariates are subject to measurement error. The measurement error may contain both a random component and a systematic difference between cases and controls (recall bias). A multivariate normal discriminant analysis model is assumed. If the distribution of measurement error is known, then a simple correction to naive (biased) estimates of odds ratios from logistic regression of disease on fallible measurements of covariates removes bias. The same correction yields confidence intervals and significance tests. We apply the proposed methods to data from a case-control study of colon cancer and diet.

Analysis of Variance↗
PubMed1989

Ordinal regression models for epidemiologic data.

Health status is often measured in epidemiologic studies on an ordinal scale, but data of this type are generally reduced for analysis to a single dichotomy. Several statistical models have been developed to make full use of information in ordinal response data, but have not been much used in analyzing epidemiologic studies. The authors discuss two of these statistical models--the cumulative odds model and the continuation ratio model. They may be interpreted in terms of odds ratios, can account for confounding variables, have clear and testable assumptions, and have parameters that may be estimated and hypotheses that may be tested using available statistical packages. However, calculations of asymptotic relative efficiency and results of simulations showed that simple logistic regression applied to dichotomized responses can in some realistic situations have more than 75% of the efficiency of ordinal regression models, but only if the ordinal scale is collapsed into a dichotomy close to the optimal point. The application of the proposed models to data from a study of chest x-rays of workers exposed to mineral fibers confirmed that they are easy to use and interpret, but gave results quite similar to those obtained using simple logistic regression after dichotomizing outcome in the conventional way.

Asbestos, Amphibole↗
PubMed1989

Work in pregnancy and birth weight for gestational age.

In a recent report on prematurity and work in pregnancy based on the Montreal survey noteworthy increases in both preterm births (less than 37 weeks) and infants of low birth weight (less than or equal to 2500 g) were found in women in certain specific occupations or whose work entailed heavy lifting, shift work, long hours, or great fatigue. Because of the large overlap between preterm births and low birth weight, the latter was further analysed with allowance for gestational age in order better to separate factors retarding fetal growth from those shortening gestation. The association of low birth weight with specific occupations, long working hours, and fatigue largely disappeared, suggesting that the effect of these factors was to shorten gestation. By contrast, the association with lifting heavy weights and with shift work persisted, suggesting that these factors retarded fetal growth as well as increasing the risk of preterm birth.

Adult↗
PubMed1988

Risk of ischemic heart disease among primary aluminum production workers.

The risk of ischemic heart disease (IHD) has been studied in relation to working conditions encountered in a primary aluminum smelter employing over 6,000 men. During the period 1975-1983, 306 new cases of IHD were identified which were matched with 575 referents. A logistic regression analysis was performed to adjust for differences in smoking habits, high blood pressure, hyperglycemia, hypercholesterolemia, and obesity. Results from this showed that white collar workers had a significantly lower risk of IHD (odds ratio 0.47, 95% confidence interval 0.31-0.70). Among blue collar workers, a significantly higher risk was observed for workers in the reduction division of the plant (OR 1.72, CI 1.09-2.97) including, in particular, Soderberg (OR 1.71, CI 1.07-2.72) and prebake (OR 2.26, CI. 1.27-4.02) potroom workers. The risk of IHD did not increase with the length of time worked in these occupations. The search for associations (among blue collar workers) of risk with nine specific contaminants (benzene soluble material, fluoride, total dust, sulfur dioxide, carbon monoxide, thermal stress, noise, physical load, and mental load) proved inconclusive, with no association reaching statistical significance.

Air Pollutants, Occupational↗
PubMed1988

Occupational and residential 60-Hz electromagnetic fields and high-frequency electric transients: exposure assessment using a new dosimeter.

One problem that has limited past epidemiologic studies of cancer and exposure to extremely low-frequency (0-100 Hz) electric and magnetic fields has been the lack of adequate methods for assessing personal exposure to these fields. A new 60-Hz electromagnetic field dosimeter was tested to assess occupational and residential exposures of a group of electrical utility workers and a comparison background group over a 7-day period. Comparing work periods only, utility workers' exposures were significantly higher than background levels by a factor of about 10 for electric (E) and magnetic (B) fields and by a factor of 171 for high-frequency transient electric (HFTE) fields. When overall weekly time-weighted averages combining work and nonwork exposures were compared, ratios of the exposed to background groups were lower. B and HFTE exposure ratios remained statistically significant, with values of 3.5 and 58, respectively, whereas the electric field exposure ratio was no longer significant, with a value of 1.7. E-field exposures of the background group were the highest during the nonwork period, probably reflecting the use of electrical appliances at home. Residential E- and B-field exposures were in the same range as published results from other surveys, whereas occupational E-field exposures tended to be lower than exposures reported in other studies. The high variability associated with occupational exposures probably accounts for the latter discrepancy. Worker acceptance of wearing the dosimeter was good: 95% of participants found it to be of little or no inconvenience while at work. At home, 37% found the device to be inconvenient in its present form but would not object to wearing a slightly smaller and lighter dosimeter.

Electricity↗
PubMed1986

Estimating the relationship between exposure to tar volatiles and the incidence of bladder cancer in aluminum smelter workers.

A previously reported case-referent study of 85 incident cases of bladder cancer among aluminum smelter workers and 255 matched referents revealed an excess risk among workers exposed to coal-tar pitch volatiles. For the study reported in the present investigation these data have been augmented by estimates of past workplace exposure to total tar (benzene-soluble matter) and to benzo-a-pyrene (BaP). From these new data, exposure-response relationships have been estimated by maximum likelihood. A linear relationship between cumulative exposure and relative risk and a minimum latency period of ten years were assumed on a priori grounds and found compatible with the data. Under these assumptions, relative risk increased for each year of exposure to benzene-soluble matter at a concentration of 1 mg/m3 by 13%, the 95% confidence interval being 5-31. The corresponding figure for BaP (as micrograms/m3 X year) was 2.3%. On the basis of these estimates, 40 years of exposure to benzene-soluble matter at the current exposure limit of 0.2 mg/m3 would lead to a relative risk of 2.4. There was suggestive but not conclusive evidence that relative risks due to exposure to tar volatiles and to cigarette smoke combined multiplicatively.

Air Pollutants, Occupational↗
PubMed1985

The problem of multiple inference in studies designed to generate hypotheses.

Epidemiologic research often involves the simultaneous assessment of associations between many risk factors and several disease outcomes. In such situations, often designed to generate hypotheses, multiple univariate hypothesis-testing is not an appropriate basis for inference. The number of true positive associations in a collection of many associations can be estimated by comparing the observed distribution of p values for the positive associations to a theoretical uniform distribution, or to the observed distribution of negative associations, or to an empiric randomization distribution. None of these approaches, however, will distinguish the true from the false positive associations. Various criteria for selecting a subset of associations to report are considered by the authors, including Bonferoni adjustment of p values, splitting the sample for searching and testing, Bayesian inference, and decision theory. The authors prefer an approach in which all associations in the data are reported, whether significant or not, followed by a ranking in order of priority for investigation using empirical Bayes techniques. Methods are illustrated by application to preliminary data from a study aimed at identifying hitherto unsuspected occupational carcinogens.

Bayes Theorem↗
PubMed1985

Prostatic cancer and chronic respiratory and renal disease in British cadmium workers: a case control study.

Deaths due to prostatic cancer, renal cancer, bronchitis or emphysema, and nephritis or nephrosis in three cohorts of cadmium workers have been investigated in a case-control study. Evidence of an association of risk for these diseases with intensity and duration of exposure to cadmium was sought. The only clearly statistically significant finding was of an association of deaths coded as bronchitis or emphysema with "high" levels of exposure to cadmium fume, which was related also to duration of exposure. There was suggestive evidence also (p congruent to 0.10) of an increased risk for nephritis or nephrosis after high exposure. Marginally increased risks were observed for prostatic cancer after high or "medium" exposure, but these were not statistically significant.

Aged↗
PubMed1984

Renal function in relation to low levels of cadmium exposure in a group of smelter workers.

Blood and urine samples were obtained from 274 smelter workers and urine samples from 48 controls. Cadmium, beta 2-microglobulin, and creatinine were estimated in blood and urine, and total protein in urine. Concentrations of cadmium in urine (mean 2.0 nmole/mmole creatinine) and blood (mean 21.8 nmole/L) observed in the smelter workers confirmed that this group had absorbed more cadmium than the general population, but less than most other occupationally exposed groups studied. Mean beta 2-microglobulin in urine was not significantly different in the smelter workers and the controls. The mean total protein in urine was 20% higher in the smelter workers, a difference which was significant (p congruent to 0.01). There was no consistent picture within the smelter workers of a relationship between history of cadmium exposure and the effect measures of beta 2-microglobulin in urine and blood, total protein in urine, creatinine clearance and relative beta 2-microglobulin clearance. Small but significant positive correlation coefficients were observed between cadmium in urine and beta 2-microglobulin in urine (r = 0.13), total protein in urine (r = 0.23) and beta 2-microglobulin clearance (r = 0.15), although these may be artifactual.

Adult↗