PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Effect Modifier, Epidemiologic”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources
PubMed1990

Questioning the status quo: sickness absence research so far claims more than it should!

This article is in the nature of a cautionary tale for those working alongside any group of medics who are required to collect epidemiological data, or plan based upon data collected by medical experts without having access to that data. The field of work is occupational health and the specialised area is the relationship between socio-economic variables and absence from work due to sickness. The over-riding criticism of the sickness absence literature to date is that the use of analytical statistics is poor or in most cases almost non-existent and that there is very little understanding of the epidemiological concepts of 'confounding' and 'effect modification'. An example is given of one particular statistical procedure, multi-variate hierarchical log-linear modelling, which if used by future researchers could prevent the recurrence of some of the previous problems.

Absenteeism↗
PubMed1989

Analytical implications of epidemiological concepts of interaction.

In contrast to definitions based on statistical or biological concepts, Rothman has adopted an unambiguous epidemiological definition of interaction in which two factors are not 'independent' if they are component causes in the same sufficient cause. This leads to the adoption of additivity of incidence rates as the state of 'no interaction'. However, there are other considerations which generally favour the use of multiplicative models. This implies an apparent dilemma as to how an analysis can be conducted which combines the advantages of ratio measures of effect with the assessment of independence in terms of a departure from additivity. These apparently contradictory goals can be reconciled through the analysis of separate and joint effects. This approach is discussed with reference to studies of asbestos exposure, cigarette smoking and lung cancer.

Causality↗
PubMed1991

Polycystic ovaries and the risk of breast cancer.

Data from a case-control study that was conducted between 1980 and 1982 were analyzed to investigate the possible association between polycystic ovaries and the risk of breast cancer. The multicenter, population-based study included in-home interviews with 4,730 women with breast cancer and 4,688 control women aged 20-54 years. The age-adjusted odds ratio for breast cancer among women with a self-reported history of physician-diagnosed polycystic ovaries was 0.52 (95% confidence interval 0.32-0.87). The inverse association was not an artifact of infertility, age at first birth, or surgical menopause. Because women with this syndrome have abnormal levels of certain endogenous hormones, the observation of a low risk of breast cancer in this group may provide new insights into hormonal influences on breast cancer.

Adult↗
PubMed1992

Selection of controls in case-control studies. I. Principles.

A synthesis of classical and recent thinking on the issues involved in selecting controls for case-control studies is presented in this and two companion papers (S. Wacholder et al. Am J Epidemiol 1992;135:1029-50). In this paper, a theoretical framework for selecting controls in case-control studies is developed. Three principles of comparability are described: 1) study base, that all comparisons be made within the study base; 2) deconfounding, that comparisons of the effects of the levels of exposure on disease risk not be distorted by the effects of other factors; and 3) comparable accuracy, that any errors in measurement of exposure be nondifferential between cases and controls. These principles, if adhered to in a study, can reduce selection, confounding, and information bias, respectively. The principles, however, are constrained by an additional efficiency principle regarding resources and time. Most problems and controversies in control selection reflect trade-offs among these four principles.

Case-Control Studies↗
PubMed1990

Use of the confidence interval function.

Graphics displaying all confidence intervals around a point estimate have been referred to as P-value functions and consonance intervals. We recommend use of the term confidence interval function (CI function) rather than P-value function. The CI function is useful because it simultaneously depicts point estimation, variability, and the relation of these two factors to the null value. The usefulness of the CI function in demonstrating the concepts of effect modification and confounding, in meta-analysis, and in the comparison of various confidence interval procedures is evaluated. Software packages that produce CI functions are described.

Confidence Intervals↗
PubMed1992

Identifiability and exchangeability for direct and indirect effects.

We consider the problem of separating the direct effects of an exposure from effects relayed through an intermediate variable (indirect effects). We show that adjustment for the intermediate variable, which is the most common method of estimating direct effects, can be biased. We also show that even in a randomized crossover trial of exposure, direct and indirect effects cannot be separated without special assumptions; in other words, direct and indirect effects are not separately identifiable when only exposure is randomized. If the exposure and intermediate never interact to cause disease and if intermediate effects can be controlled, that is, blocked by a suitable intervention, then a trial randomizing both exposure and the intervention can separate direct from indirect effects. Nonetheless, the estimation must be carried out using the G-computation algorithm. Conventional adjustment methods remain biased. When exposure and the intermediate interact to cause disease, direct and indirect effects will not be separable even in a trial in which both the exposure and the intervention blocking intermediate effects are randomly assigned. Nonetheless, in such a trial, one can still estimate the fraction of exposure-induced disease that could be prevented by control of the intermediate. Even in the absence of an intervention blocking the intermediate effect, the fraction of exposure-induced disease that could be prevented by control of the intermediate can be estimated with the G-computation algorithm if data are obtained on additional confounding variables.

Algorithms↗
PubMed1991

Prolonged breastfeeding and malnutrition: confounding and effect modification in a Brazilian cohort study.

We examined the association between prolonged breastfeeding and anthropometric status in a population-based cohort study of 5,914 liveborns from the city of Pelotas in Southern Brazil. When children from all socioeconomic groups were studied, there was no important association between current breastfeeding and anthropometric status at age 12 months. Children who were still breastfed at age 20 months--and, to a lesser extent, at 43 months--presented with poorer anthropometric status than their nonbreastfed counterparts. We did not find the same pattern in all socioeconomic groups, however. Children from low-income families who were breastfed tended to present better anthropometric status than those who were not, whereas the reverse was observed for children of middle- and high-income families. After controlling for confounding variables, the nutritional advantage of breastfeeding among low-income families was no longer clear, while the superiority of nonbreastfed infants amongst middle- and high-income children persisted. These findings indicate that some of the controversy regarding the nutritional effects of prolonged breastfeeding may have been caused by confounding and effect modification. Any decisions on whether or not breastfeeding should be encouraged after the first year of life should take into account the characteristics of the population as well as the anti-infective and birth-spacing properties of breastfeeding.

Anthropometry↗
PubMed1992

Definition, sources, magnitude, effect modifiers, and strategies of reduction of the healthy worker effect.

This article summarizes, compares, and contrasts the definition, sources, magnitude, effect modifiers, and strategies of reduction of the healthy worker effect (HWE), based on the opinion expressed in the papers of nine contributors who responded to the request of the Industrial Disease Standards Panel (IDSP), Ontario, Canada. It provides an insight into the complex issues relating to the HWE. In addition, the catalog of 15 strategies to reduce the HWE is deemed to be useful for investigators in occupational epidemiology.

Age Factors↗
PubMed1992

Divergent biases in ecologic and individual-level studies.

Several authors have shown that ecologic estimates can be biased by effect modification and misclassification in a different fashion from individual-level estimates. This paper reviews and discusses ecologic biases induced by model misspecification; confounding; non-additivity of exposure and covariate effects (effect modification); exposure misclassification; and non-comparable standardization. Ecologic estimates can be more sensitive to these sources of bias than individual-level estimates, primarily because ecologic estimates are based on extrapolations to an unobserved conditional (individual-level) distribution. Because of this sensitivity, one should not rely on a single regression model for an ecologic analysis. Valid ecologic estimates are most feasible when one can obtain accurate estimates of exposure and covariate means in regions with internal exposure homogeneity and mutual covariate comparability; thus, investigators should seek out such regions in the design and analysis of ecologic studies.

Bias↗
PubMed1989

A review of the effects of random measurement error on relative risk estimates in epidemiological studies.

Many articles in the recent epidemiological literature have discussed the effects of random error and misclassification on effect estimation, but many of these have been unclear and hard to follow. This paper reviews and interprets many of these and summarizes the use of the correlation coefficient in assessing the likely effect of measurement error on relative risk estimates for variables that are either continuous or ordered. A table of expected values of relative risks (RRs) calculated in different ways for different levels of random error is presented and the typically large expected attenuation in RR values is shown. The recommendation of taking repeated or multiple measurements whenever possible is endorsed.

Confounding Factors, Epidemiologic↗
PubMed1992

The risk of epithelial ovarian cancer in short-term users of oral contraceptives.

Short-term use (less than 1 year) or oral contraceptives has been associated with increased to slightly decreased risks of epithelial ovarian cancer in several studies. To determine what might account for a statistically significant 40% reduction in risk associated with as little as 3 to 6 months of use, a finding previously reported from the Cancer and Steroid Hormone Study, and to consider the implications for mechanisms of pathogenesis, the authors compared numerous characteristics of short-term users of oral contraceptives (41 cases, 412 controls) with those of never users (242 cases, 1,517 controls). The reduced risk among short-term users was consistently restricted to women who stopped using oral contraceptives for medical reasons, which were essentially side effects; there was little evidence of a protective effect among women who stopped for nonmedical reasons. Factors such as age, parity, family history of ovarian cancer, estrogen dose, history of sterilization, and latency (interval from first use) could not account for the finding. These analyses suggest that short-term use of oral contraceptives has little to no effect per se on reducing the risk of epithelial ovarian cancer and that side effects resulting in cessation of oral contraceptive use shortly after it was begun may be indicative of factors that are protective against the disease.

Adult↗
PubMed1990

Loglinear models, sexual behavior and HIV: epidemiological implications of heterosexual transmission.

An analysis is presented of sexual behaviour data from a "random" sample (n = 780) of the adult population of England and Wales. Sex stratified frequency distributions of the reported number of sex partners/time were analysed for heterosexuals, demographic characteristics associated with the number of sex partners/time were identified, and epidemiological parameters (the basic reproductive rate of HIV and the doubling time of the epidemic) were calculated. These analyses suggest that the size of the group at risk for acquiring the virus by heterosexual transmission may be large and that the age of first sexual intercourse (for males and females) is decreasing in younger cohorts. Members of the potential heterosexual at-risk group may be identified by demographic variables such as marital status (males and females) and age (females only), but not by socioeconomic class. The epidemiological implications of our results for the heterosexual transmission of HIV are discussed.

Adolescent↗
PubMed1990

Differential recall as a source of bias in epidemiologic research.

Differential recall between compared groups is discussed most often in the context of case-control studies. Cases, compared to controls, are suspected of providing a more complete report of their true exposure to an hypothesized risk factor, thereby biasing upwards the estimate of its effect. The present paper describes how differential recall can arise with any observational design in epidemiology; with any class of study variable, not only exposures; and may inflate or deflate the true value of the estimate of effect size. We list a variety of study designs and questionnaire tactics that aim to remedy these problems. The scope and magnitude of the bias created by differential recall and the efficacy of proposed remedies require further study.

Bias↗
PubMedDate not supplied

Confounding bias and effect modification in epidemiologic research.

The medical literature contains many studies of the clinical effectiveness of diagnostic tests and therapeutic interventions. Common problems in experimental design occur that influence the usefulness of original research. Confounding bias and effect modification are two important factors that affect whether clinicians ought to apply the findings of clinical research to the care of their patients. Investigators should minimize confounding biases in their work. Effect modification should be described so that readers can decide which of their patients will benefit from a particular study. This article uses a number of clinical examples to help the clinician and investigator understand the influences of bias and effect modification on clinical research.

Age Factors↗
PubMed1989

Diet history: questionnaire and interview techniques used in some retrospective studies of cancer.

A detailed diet history method that has been used by the Epidemiology Unit of the National Cancer Institute of Canada, Toronto, in retrospective case-control studies is discussed. The questionnaire format and the method of administration by trained interviewers, together with instructions used to train the interviewers, are described. The questionnaire items elicited the usual frequency of foods consumed in the Canadian diet for one or two specified periods of time, including current and/or past diet history. Various probes required to elicit proper information on frequency, serving size, and additions to food are discussed. Portions were quantified by using geometric food models made of papier-mâché, wood, and plastic. The frequency and volumetric measurements obtained from the diet history questionnaire were converted to gram weights using the density of each food consumed; the nutrient intakes were then computed.

Canada↗
PubMedDate not supplied

Is weight loss a modifier of the cholesterol-heart disease relationship in older persons? Data from the NHANES I Epidemiologic Follow-up Study.

The relationship between cholesterol and 14-year incidence of coronary heart disease was compared for men and women of two age groups, 25 to 64 years and 65 to 74 years. While cholesterol levels of 6.2 mmol/L or higher were associated with a risk of coronary heart disease in the younger group, this was not true for either men or women aged 65 to 74. Further analyses for older persons showed that weight loss modified the cholesterol-heart disease relationship. Those with stable weight showed a positive relationship between cholesterol and coronary heart disease, similar to the younger age group (relative risk [RR] = 1.8 [95% confidence interval: 1.1, 2.9] for men; RR = 1.6 [.7, 3.4] for women). Among those with a weight loss of 10% or more, the relationship of cholesterol to heart disease was inverse (RR = .8 [.5, 1.2] for men; RR = .6 [.3, 1.0] for women). These data suggest that the relationship of cholesterol to coronary disease in healthier older persons may be similar to that in younger persons, and that health status should be considered in analyses of cholesterol risk in old age.

Adult↗
PubMedDate not supplied

Differences in histology between first and second primary lung cancer.

Data from the Surveillance, Epidemiology, and End Results (SEER) Program were used to compare the histological distribution of second lung cancer following an initial cancer of the lung, head and neck, and breast to primary lung carcinoma occurring as a first cancer. Following initial head and neck cancer or initial squamous cell carcinoma of the lung, the proportion of second primary lung cancer which was of squamous cell histology rose dramatically, while the proportion of pulmonary adenocarcinomas rose following initial adenocarcinoma of the lung. The histological distribution of lung cancer following an initial breast cancer in women was similar to the distribution of de novo lung cancer in women. These results persisted as the time interval between diagnosis of the two primaries was increased from 12 to 48 months. We conclude that the histology of a second primary lung cancer following an initial cancer of the lung or head and neck tends to repeat the histology of the initial cancer (field effect), and this observation is not likely to be due to misdiagnosis of a recurrence of the initial cancer.

Adenocarcinoma↗