Read e-book Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s)

Free download. Book file PDF easily for everyone and every device. You can download and read online Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s) file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s) book. Happy reading Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s) Bookeveryone. Download file Free Book PDF Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s) at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF Additive and multiplicative models for the joint effect of two risk factors (2005)(en)(9s) Pocket Guide.

If you previously purchased this article, Log in to Readcube.

Additive and multiplicative models for the joint effect of two risk factors.

Log out of Readcube. Click on an option below to access. Log out of ReadCube. Current methods for detecting genetic associations lack full consideration of the background effects of environmental exposures. Recently proposed methods to account for environmental exposures have focused on logistic regressions with gene—environment interactions. In this report, we developed a test for genetic association, encompassing a broad range of risk models, including linear, logistic and probit, for specifying joint effects of genetic and environmental exposures.

We obtained the test statistics by maximizing over a class of score tests, each of which involves modified standard tests of genetic association through a weight function. This weight function reflects the potential heterogeneity of the genetic effects by levels of environmental exposures under a particular model.

Simulation studies demonstrate the robust power of these methods for detecting genetic associations under a wide range of scenarios.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries other than missing content should be directed to the corresponding author for the article. Volume 71 , Issue 3. The full text of this article hosted at iucr.


  • ?
  • Finland in the European Union.
  • Ballistic-Missile Defence and Strategic Stability (Adelphi series).
  • Electrohydrodynamic Patterning of Functional Materials;
  • Discrete Mathematics, Third Edition!
  • Reclaiming Unlived Life: Experiences in Psychoanalysis.

If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. If the address matches an existing account you will receive an email with instructions to retrieve your username. Summer S. Search for more papers by this author.

An exposure‐weighted score test for genetic associations integrating environmental risk factors

Philip S. Neil E. Tools Request permission Export citation Add to favorites Track citation. Share Give access Share full text access. Share full text access.


  • ;
  • Maya Angelou (Black Americans of Achievement).
  • .
  • Measuring additive interaction using odds ratios;
  • Skellig.

Please review our Terms and Conditions of Use and check box below to share full-text version of article. Get access to the full version of this article. View access options below. You previously purchased this article through ReadCube. Table 2 describes the distribution of cases and controls according to categories of exposure to smoking and drinking. The median cumulative tobacco consumption among control patients was 28 pack-years, while the median cumulative alcohol consumption was gram-years. These values were the thresholds used to split smoking and drinking exposures into level-1 moderate consumption and level-2 heavy consumption.

Table 2 also describes the unadjusted assessment of associations between the outcome and these covariates. The highest risk estimates were reported for level-2 smokers OR, 7. In the multivariable assessment Table 3 , Model-1 registered that the two lifestyle exposures, considered both independently and jointly, were highly and significantly associated with oral cancer. The ORs estimated by Model-2 were generally lower than those estimated by Model-1 and the independent effect of alcohol drinking resulted no longer associated with the disease OR, 0.

An analogous result was obtained when both exposures were classified using three categories Table 3. Model-1 showed that all the smoking-drinking categories different from the reference group had a direct and statistically significant association with the disease.

Additive and multiplicative models for the joint effect of two risk factors.

However, according to Model-2, no significant association was observed between the disease and the various smoking-drinking categories for individuals who smoked less than 28 pack-years level-1 smokers and drank less than gram-years level-1 drinkers. The ORs for all the categories of concurrent exposure to smoking and drinking estimated by Model-2 were lower than the ORs estimated by Model According to the validation analysis, the coefficient estimates and goodness of fit of regression models did not change substantially, thus corroborating the robustness of the various risk estimates data not in Table.

The most important findings of this study were that 1 the independent effect of drinking substantially decreased and was no longer associated with oral and oropharyngeal cancer accounting for the smoking-drinking interaction term; 2 the independent effect of smoking also considerably decreased, although it remained significantly associated with the disease; 3 the smoking-drinking joint effect remained significantly associated with oral cancer; 4 regression models accounting for the smoking-drinking interaction term had a significantly better fit than those exclusively assessing the individual effects of behavioural exposures.

In addition, cases and controls belonged to the same homogeneous study population, a situation which helps control for other risk factors for oral cancer, because in these circumstances the unidentified and non-investigated factors are part of the background environment and can be disregarded. This is an important advantage for the investigation of diseases with complex multifactorial aetiology like oral cancer [13] , [14].

Discriminative accuracy of genomic profiling comparing multiplicative and additive risk models

Therefore, large size and homogeneity of sample increased the internal validity of this study, that is, the reliability of the reported risk estimates [28]. Indeed, factors which bring people to public hospitals, such as financial standings, area of residence, ethnicity, religious affiliation, are not distributed uniformly within the underlying study population, thus making it difficult to define the population from which our case patients arose.


  • The Offshore Renminbi: The Rise of the Chinese Currency and Its Global Future.
  • .
  • Statistical Analysis of Designed Experiments: Theory and Applications.
  • Bulk and Boundary Invariants for Complex Topological Insulators: From K-Theory to Physics.
  • .
  • Direct and Inverse Problems of Mathematical Physics.

In order to minimize the consequences of selection bias on the consistency of results, we decided to select hospital-based controls who followed the same referral route as cases. In addition, the selection of case and control patients from the same underlying study population minimized the degree of information bias, since hospital-based controls tend to show the same levels of cooperativeness and accuracy in providing information as the hospital-based cases, thus reducing the potential differences between these two groups in the quality of recall of past exposures [13].

Information bias regarding lifestyle variables, which are generally underreported by heavy users is another important form of bias [29] — [31]. Since reliability of lifestyle variables is particularly low when they are treated quantitatively, we classified lifestyle variables into categories, considering that treating variables semi-quantitatively would reduce the negative effect of information bias on the reliability of risk estimates [32] , [33]. This approach is also preferred by experts in alcohol drinking epidemiology, who make international comparisons using qualitative data [32].

Recall bias and interviewer bias, two specific forms of information bias, are also relevant limitations of case-control studies [13] , [23] , [34]. Patients diagnosed with oral and oropharyngeal cancer may have spent some time pondering on deleterious habits that may have contributed to the disease. Therefore, cases would be more likely to recall alcohol drinking and tobacco smoking than controls.

This limitation is difficult to overcome in the scope of a case-control study. Interviewer bias also could not be excluded, as interviewers were trained, but not blinded. Recall and interviewer biases may have resulted in overestimates of smoking and drinking exposure among cases. Consequently, the smoking and drinking risk estimates might have been artificially higher than the true risks, which would be even lower if these forms of bias were completely controlled.

Supporting Information

Finally, it is possible that other variables, strongly associated with both drinking and smoking, also were associated with oral and oropharyngeal cancer risk. Indeed, smoking and drinking are not only associated with each other, they are also associated with other behavioural risk factors for cancer and other degenerative diseases, such as unsafe sex, use of other addictive substances, unhealthy diet, low physical exercise, etc. Therefore, the joint exposure to smoking and drinking may also imply a potentially etiological role of hidden variables.

The data of the present study were partly corroborated. As noted, multi-centre studies are not homogeneous and, therefore, tend to overestimate the effects of the investigated variables because they do not account for unidentified factors [13]. This study may have implications in the design of effective oral cancer control policy. Therefore, our study demonstrated that the exposure to a single lifestyle is not only uncommon, as previously demonstrated, but it also does not pose an important risk for oral cancer.

Conversely, multiple exposures are very frequent, due to the widespread addictive behaviour, and are a serious risk for oral cancer. Instead, the joint effect of drinking and smoking was significantly associated with this condition.

Using the Time Series Additive Model in Excel

We strongly recommend that observational studies from other countries are designed accounting for such an important interaction, thus allowing investigating whether this result may be generalized to populations with different lifestyles, and with various drinking and smoking modalities. According to these findings, oral cancer control policies should focus primarily on the addictive behaviour which induces people to adopt several unsafe lifestyles.

Browse Subject Areas? Click through the PLOS taxonomy to find articles in your field. Abstract Alcohol drinking and tobacco smoking are assumed to have significant independent and joint effects on oral cancer OC development. Funding: No current external funding sources for this study. Introduction Tobacco smoking and alcohol drinking are lifestyle risk factors which play an etiological role in oral cancer development with sufficient evidence.