Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “causal inference”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 127 records · Page 7Linked to original sources
PubMed2000

Tobacco marketing and adolescent smoking: more support for a causal inference.

OBJECTIVES: This prospective study examined the effect of tobacco marketing on progression to established smoking. METHODS: Massachusetts adolescents (n = 529) who at baseline had smoked no more than 1 cigarette were reinterviewed by telephone in 1997. Analyses examined the effect of receptivity to tobacco marketing at baseline on progression to established smoking, controlling for significant covariates. RESULTS: Adolescents who, at baseline, owned a tobacco promotional item and named a brand whose advertisements attracted their attention were more than twice as likely to become established smokers (odds ratio = 2.70) than adolescents who did neither. CONCLUSIONS: Participation in tobacco marketing often precedes, and is likely to facilitate, progression to established smoking. Hence, restrictions on tobacco marketing and promotion could reduce addiction to tobacco.

Adolescent↗
PubMed1981

Alcohol and high-density lipoprotein cholesterol: causal inference from diverse study designs.

The association between reported alcohol intake and plasma high-density lipoprotein (HDL) cholesterol concentration is examined in an effort to establish whether it was a cause-and-effect basis. A cross-sectional descriptive study of several populations reveals a strong and consistent dose-response pattern: Social drinkers have mean HDL cholesterol levels that are higher than those of teetotalers by as much as 33%. Cross-sectional analyses in another epidemiological study reveal the association to be independent of potential confounding factors such as smoking and body weight, and longitudinal analyses suggest that it is also not a result of certain unmeasured sources of confounding. A small experiment reveals a 15% reduction in HDL cholesterol levels among social drinkers who abstain from alcohol from a 2-week period. The evidence supports the conclusion that alcohol habits are probably one of the determinants of plasma HDL cholesterol level. A clarification of the relevance of this phenomenon to clinical medicine awaits future clinical efforts.

Alcohol Drinking↗
PubMed1991

Practical implications of modes of statistical inference for causal effects and the critical role of the assignment mechanism.

Causal inference in an important topic and one that is now attracting serious attention of statisticians. Although there exist recent discussions concerning the general definition of causal effects and a substantial literature on specific techniques for the analysis of data in randomized and nonrandomized studies, there has been relatively little discussion of modes of statistical inference for causal effects. This presentation briefly describes and contrasts four basic modes of statistical inference for causal effects, emphasizes the common underlying causal framework with a posited assignment mechanism, and describes practical implications in the context of an example involving the effects of switching from a name-band to a generic drug. A fundamental conclusion is that in such nonrandomized studies, sensitivity of inference to the assignment mechanism is the dominant issue, and it cannot be avoided by changing modes of inference, for instance, by changing from randomization-based to Bayesian methods.

Bayes Theorem↗
PubMed1997

Does diesel exhaust cause human lung cancer?

Recent reviews of epidemiological evidence on the relation between exposure to diesel exhaust (DE) and lung cancer risk have reached conflicting conclusions, ranging from belief that there is sufficient evidence to conclude that DE is a human lung carcinogen (California EPA, 1994) to conclusions that there is inadequate evidence to support a causal association between DE and human lung cancer (Muscat and Wynder, 1995). Individual studies also conflict, with both increases and decreases in relative risks of lung cancer mortality being cited with 95% statistical confidence. On balance, reports of elevated risk outnumber reports of reduced risk. This paper reexamines the evidence linking DE exposures to lung cancer risk. After briefly reviewing animal data and biological mechanisms, it surveys the relevant epidemiological literature and examines possible explanations for the discrepancies. These explanations emphasize the distinction between statistical associations, which have been found in many studies, and causal associations, which appear not to have been established. Methodological threats to valid causal inference are identified and new approaches for controlling them are proposed using recent techniques from artificial intelligence (AI) and computational statistics. These threats have not been adequately controlled for in previous epidemiological studies. They provide plausible noncausal explanations for the reported increases in relative risks, making it impossible to infer causality between DE exposure and lung cancer risk from these studies. A key contribution is to show how recent techniques developed in the AI-and-statistics literature can help clarify the causal interpretation of complex multivariate data sets used in epidemiological risk assessments. Applied to the key study of Garshick et al. (1988), these methods show that DE concentration has no positive causal association with occupational lung cancer mortality risk.

Animals↗
PubMed2006

Proposal for an integrated evaluation model for the study of whole systems health care in cancer.

For more than 200 years, biomedicine has approached the treatment of disease by studying disease processes (patho-genesis), inferring causal connections and developing specific approaches for therapeutically interfering with those processes. This pathogenic approach has been highly successful in acute and traumatic disease but less successful in chronic disease, primarily because of the complex, multi-factorial nature of most chronic disease, which does not allow for simple causal inference or for simple therapeutic interventions. This article suggests that chronic disease is best approached by enhancing healing processes (salutogenesis) as a whole system. Because of the nature of complex systems in chronic disease, an evaluation model based on integrative medicine is felt to be more appropriate than a disease model. The authors propose and describe an integrated model for the evaluation of healing (IMEH) that collects multilevel "thick case" observational data in assessing complex practices for chronic disease. If successful, this approach could become a blueprint for studying healing capacity in whole medical systems, including complementary medicine, traditional medicine, and conventional primary care. In addition, streamlining data collection and applying rapid informatics management might allow for such data to be used in guiding clinical practice. The IMEH involves collection, integration, and potentially feedback of relevant variables in the following areas: (1) sociocultural, (2) psychological and behavioral, (3) clinical (diagnosis based), and (4) biological. Evaluation and integration of these components would involve specialized research teams that feed their data into a single data management and information analysis center. These data can then be subjected to descriptive and pathway analysis providing "bench and bedside" information.

Chronic Disease↗
PubMed2004

Temporal aggregation bias and inference of causal regulatory networks.

Time course experiments with microarrays have begun to provide a glimpse into the dynamic behavior of gene expression. In a typical experiment, scientists use microarrays to measure the abundance of mRNA at discrete time points after the onset of a stimulus. Recently, there has been much work on using these data to infer causal regulatory networks that model how genes influence each other. However, microarray studies typically have slow sampling rates that can lead to temporal aggregation of the signal. That is, each successive sampling point represents the sum of all signal changes since the previous sample. In this paper, we show that temporal aggregation can bias algorithms for causal inference and lead them to discover spurious relations that would not be found if the signal were sampled at a much faster rate. We discuss the implications of temporal aggregation on inference, the problems it creates, and potential directions for solutions.

Computational Biology↗
PubMed2006

Time as a guide to cause.

How do people learn causal structure? In 2 studies, the authors investigated the interplay between temporal-order, intervention, and covariational cues. In Study 1, temporal order overrode covariation information, leading to spurious causal inferences when the temporal cues were misleading. In Study 2, both temporal order and intervention contributed to accurate causal inference well beyond that achievable through covariational data alone. Together, the studies show that people use both temporal-order and interventional cues to infer causal structure and that these cues dominate the available statistical information. A hypothesis-driven account of learning is endorsed, whereby people use cues such as temporal order to generate initial models and then test these models against the incoming covariational data.

Computers↗
PubMed1995

Children's inferences of causal events during story understanding.

Second-grade children's inference of causal events was studied. Each of the 24 children heard 2 stories containing 6 event episodes. Each story was followed by either a sentence recognition or a cued-recall test. Event episodes explicitly stated an event followed by its outcome. Enablement episodes explicitly stated an action that was causally antecedent to the event, followed by the outcome of the event. Filler (control) episodes explicitly stated an action causally unrelated to the event, followed by the outcome of the event. The results indicate that the children more frequently inferred events from enablements than they inferred enablements from events, and that the children did not make the inferences at the time of reading, but instead made them when queried at the time of the test. This conclusion is suggested because there was false recognition of events but no false recall.

Attention↗
PubMed2001

Homocysteine and atherothrombosis.

Atherosclerosis with or without thrombosis superimposed is the most frequent cause of ischemic heart disease (IHD), peripheral arterial disease, and a main cause of stroke. Conflicting results have been reported in genetic, observational, and experimental studies on the relationship between homocysteine and these atherothrombotic diseases. Although cardiovascular complications are common in homocystinuric patients (severe hyperhomocysteinemia), IHD, the most frequent manifestation of atherothrombosis in the general population, appears to be rare. On the basis of findings in individuals with hyperhomocysteinemia of genetic origin, there is in fact no clear evidence for a causal role of homocysteine in the pathogenesis of atherothrombotic disease, and the positive association between plasma homocysteine and IHD observed in many, but not all epidemiologic studies does not prove causality. To infer causality from observational studies, there should be a temporal, consistent, strong, independent, graded (dose-response effect), and duration-dependent relationship between exposure and outcomes, and a biologically plausible mechanism should exist. The relationship between plasma homocysteine levels and IHD does not fulfill these criteria beyond reasonable doubt. In the general population, plasma homocysteine levels are to a great extent determined by dietary habits, and plasma homocysteine could be a marker, or a consequence, of atherothrombosis and/or risk-associated behavior (e.g., a diet low in fruits and vegetables) rather than a cause of atherothrombosis. Experimentally, hyperhomocysteinemia is not in itself atherogenic in normal animals with relatively low plasma cholesterol levels. The homocysteine theory of atherosclerosis should be tested more thoroughly in hypercholesterolemic animals that develop atherosclerosis spontaneously to determine whether elevated plasma homocysteine levels are harmful under atherogenic conditions. A causal role of homocysteine in atherothrombotic disease remains to be established.

Arteriosclerosis↗