Dataset Information

Comparing the estimates of effect obtained from statistical causal inference methods: An example using bovine respiratory disease in feedlot cattle.

ABSTRACT: The causal effect of an exposure on an outcome of interest in an observational study cannot be estimated directly if the confounding variables are not controlled. Many approaches are available for estimating the causal effect of an exposure. In this manuscript, we demonstrate the advantages associated with using inverse probability weighting (IPW) and doubly robust estimation of the odds ratio in terms of reduced bias. IPW approach can be used to adjust for confounding variables and provide unbiased estimates of the exposure's causal effect. For cluster-structured data, as is common in animal populations, inverse conditional probability weighting (ICPW) approach can provide a robust estimation of the causal effect. Doubly robust estimation can provide a robust method even when the specification of the model form is uncertain. In this paper, the usage of IPW, ICPW, and doubly robust approaches are illustrated with a subset of data with complete covariates from the Australian-based National Bovine Respiratory Disease Initiative as well as simulated data. We evaluate the causal effect of prior bovine viral diarrhea exposure on bovine respiratory disease in feedlot cattle. The results show that the IPW, ICPW and doubly robust approaches would provide a more accurate estimation of the exposure effect than the traditional outcome regression model, and doubly robust approaches are the most preferable overall.

SUBMITTER: Ji J

PROVIDER: S-EPMC7316239 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comparing the estimates of effect obtained from statistical causal inference methods: An example using bovine respiratory disease in feedlot cattle.

Ji Ju J Wang Chong C He Zhulin Z Hay Karen E KE Barnes Tamsin S TS O'Connor Annette M AM

PloS one 20200625 6

The causal effect of an exposure on an outcome of interest in an observational study cannot be estimated directly if the confounding variables are not controlled. Many approaches are available for estimating the causal effect of an exposure. In this manuscript, we demonstrate the advantages associated with using inverse probability weighting (IPW) and doubly robust estimation of the odds ratio in terms of reduced bias. IPW approach can be used to adjust for confounding variables and provide unbi ...[more]

PMID: 32584812

Similar Datasets

Project description:PurposeThe purpose of this tutorial is to provide visual scientists with various approaches for comparing two or more groups of data using parametric statistical tests, which require that the distribution of data within each group is normal (Gaussian). Non-parametric tests are used for inference when the sample data are not normally distributed or the sample is too small to assess its true distribution.MethodsMethods are reviewed using retinal thickness, as measured by optical coherence tomography (OCT), as an example for comparing two or more group means. The following parametric statistical approaches are presented for different situations: two-sample t-test, Analysis of Variance (ANOVA), paired t-test, and the analysis of repeated measures data using a linear mixed-effects model approach.ResultsAnalyzing differences between means using various approaches is demonstrated, and follow-up procedures to analyze pairwise differences between means when there are more than two comparison groups are discussed. The assumption of equal variance between groups and methods to test for equal variances are examined. Examples of repeated measures analysis for right and left eyes on subjects, across spatial segments within the same eye (e.g. quadrants of each retina), and over time are given.ConclusionsThis tutorial outlines parametric inference tests for comparing means of two or more groups and discusses how to interpret the output from statistical software packages. Critical assumptions made by the tests and ways of checking these assumptions are discussed. Efficient study designs increase the likelihood of detecting differences between groups if such differences exist. Situations commonly encountered by vision scientists involve repeated measures from the same subject over time, measurements on both right and left eyes from the same subject, and measurements from different locations within the same eye. Repeated measurements are usually correlated, and the statistical analysis needs to account for the correlation. Doing this the right way helps to ensure rigor so that the results can be repeated and validated.

Project description:Bovine respiratory disease (BRD) is the most important cause of clinical disease and death in feedlot cattle. Respiratory viral infections are key components in predisposing cattle to the development of this disease. To quantify the contribution of four viruses commonly associated with BRD, a case-control study was conducted nested within the National Bovine Respiratory Disease Initiative project population in Australian feedlot cattle. Effects of exposure to Bovine viral diarrhoea virus 1 (BVDV-1), Bovine herpesvirus 1 (BoHV-1), Bovine respiratory syncytial virus (BRSV) and Bovine parainfluenza virus 3 (BPIV-3), and to combinations of these viruses, were investigated. Based on weighted seroprevalences at induction (when animals were enrolled and initial samples collected), the percentages of the project population estimated to be seropositive were 24% for BoHV-1, 69% for BVDV-1, 89% for BRSV and 91% for BPIV-3. For each of the four viruses, seropositivity at induction was associated with reduced risk of BRD (OR: 0.6-0.9), and seroincrease from induction to second blood sampling (35-60 days after induction) was associated with increased risk of BRD (OR: 1.3-1.5). Compared to animals that were seropositive for all four viruses at induction, animals were at progressively increased risk with increasing number of viruses for which they were seronegative; those seronegative for all four viruses were at greatest risk (OR: 2.4). Animals that seroincreased for one or more viruses from induction to second blood sampling were at increased risk (OR: 1.4-2.1) of BRD compared to animals that did not seroincrease for any viruses. Collectively these results confirm that prior exposure to these viruses is protective while exposure at or after feedlot entry increases the risk of development of BRD in feedlots. However, the modest increases in risk associated with seroincrease for each virus separately, and the progressive increases in risk with multiple viral exposures highlights the importance of concurrent infections in the aetiology of the BRD complex. These findings indicate that, while efficacious vaccines could aid in the control of BRD, vaccination against one of these viruses would not have large effects on population BRD incidence but vaccination against multiple viruses would be expected to result in greater reductions in incidence. The findings also confirm the multifactorial nature of BRD development, and indicate that multifaceted approaches in addition to efficacious vaccines against viruses will be required for substantial reductions in BRD incidence.

Project description:Bovine digital dermatitis (DD) is a skin disorder that is a significant cause of infectious lameness in cattle around the world. However, very little is known about the etiopathogenesis of the disease and the microbiota associated with DD in beef cattle. In this study, we provide a comprehensive characterization of DD and healthy skin microbiota of feedlot beef cattle. We also developed and validated a novel multiplex quantitative PCR (qPCR) assay to quantify the distribution of DD-associated bacterial species across DD lesion stages. We determined the DD-associated microbiota with deep amplicon sequencing of the V3-V4 hypervariable region of the 16S rRNA gene, followed by the application of novel and existing qPCR assays to quantify species distributions of Treponema, Porphyromonas, Fusobacterium, and Bacteroides across lesion stages. Deep amplicon sequencing revealed that Treponema, Mycoplasma, Porphyromonas, and Fusobacterium were associated with DD lesions. Culturing of DD biopsy specimens identified Porphyromonas levii, Bacteroides pyogenes, and two Fusobacterium spp. within DD lesions. Using species-specific qPCR on DD lesion DNA, we identified P. levii in 100% of active lesion stages. Early-stage lesions were particularly associated with Treponema medium, T. phagedenis, and P. levii. This study suggests a core DD microbial group consisting of species of Treponema, Fusobacterium, Porphyromonas, and Bacteroides, which may be closely tied with the etiopathogenesis of DD. Further characterizations of these species and Mycoplasma spp. are necessary to understand the microbial factors involved in DD pathogenesis, which will help elucidate DD etiology and facilitate more targeted and effective mitigation and treatment strategies. IMPORTANCE Previous work, primarily in dairy cattle, has identified various taxa associated with digital dermatitis (DD) lesions. However, there is a significant gap in our knowledge of DD microbiology in beef cattle. In addition, characterization of bacteria at the species level in DD lesions is limited. In this study, we provide a framework for the accurate and reproducible quantification of major DD-associated bacterial species from DNA samples. Our findings support DD as a polymicrobial infection, and we identified a variety of bacterial species spanning multiple genera that are consistently associated with DD lesions. The DD-associated microbiota identified in this study may be capable of inducing the formation and progression of DD lesions and thus should be primary targets in future DD pathogenesis studies.

Dataset Information

Comparing the estimates of effect obtained from statistical causal inference methods: An example using bovine respiratory disease in feedlot cattle.

Publications

Comparing the estimates of effect obtained from statistical causal inference methods: An example using bovine respiratory disease in feedlot cattle.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets