Robust Bayesian approach to logistic regression modeling in small sample size utilizing a weakly informative student’s t prior distribution

Author(s):  
Kenneth Chukwuemeka Asanya ◽  
Mohamed Kharrat ◽  
Akaninyene Udo Udom ◽  
Emmanuel Torsen
Plant Disease ◽  
2006 ◽  
Vol 90 (8) ◽  
pp. 1039-1044 ◽  
Author(s):  
L. Cadle-Davidson ◽  
M. E. Sorrells ◽  
S. M. Gray ◽  
G. C. Bergstrom

Soilborne wheat mosaic virus (SBWMV) was detected in New York in 1998 for the first time and has been associated with yield loss where identified. We assessed 115 regionally adapted small grains genotypes for resistance to SBWMV over four growing seasons. Resistance to SBWMV reduces the percentage of plants that develop detectable viral titer and symptoms. Logistic regression was used to analyze disease incidence data and was compared with a general linear model for categorizing relative resistance to SBWMV. Logistic regression facilitated assessment of the effects of small sample size, low disease incidence, and nonuniform disease distribution. By increasing sample size from 20 to 30 stems per replicate, the number of resistance categories was increased through improved resolution of intermediate resistance classes. In environments with low disease incidence, the number of genotypes categorized as susceptible decreased while intermediate genotypes appeared to be resistant in the analysis. Inclusion of disease distribution data as covariates in a spatially balanced experiment did not increase the power of the logistic analysis. No genotype assessed in multiple years was immune to infection. However, 41 of the regionally adapted genotypes tested repeatedly expressed strong resistance to SBWMV, providing growers a choice of cultivars resistant to SBWMV.


2020 ◽  
Author(s):  
michael e johansen

Introduction: I attempted to determine rates of use and associations with use of acid-suppressing medications in infants under 2 years old given these are not well studied. Methods: The 2002-2018 Medical Expenditure Panel Survey (MEPS) was used for the analysis. The survey is sponsored by the Agency of Healthcare Research and Quality and is conducted on households to be nationally representative of the non-institutionalized population of the United States. The survey is comprised of two overlapping panels that are each included in the survey for two years. Newborns and infants under 2 years old on December 31 of a survey year were included in the study. Additionally, adult men and women living in the same household between 18-45 years old were considered fathers and mothers of infants. Histamine2-receptor antagonists (H2RA) and proton pump inhibitors (PPI) were identified for infants. H2RA, PPIs, and anti-depressant (selective serotonin re-uptake inhibitors and serotonin-norepinephrine re-uptake inhibitors) were identified for parents. Sex, age (by month for child), race/ethnicity (White non-Hispanic, Black non-Hispanic, Hispanic, or other), region (Northeast, Midwest, South, and West), and poverty category were identified. Chi-squared and adjusted Wald tests were used to determine statistical significance with acid-suppressing medication use between 2010-2018. A multivariable logistic regression predicted any acid suppressing medication use with month of age and month of age2 as independent variables of use for the population between 2009-2018. The study was considered exempt by the OhioHealth Institutional Review Board. Survey weighting was included in all analyses. Results: The study included a total of 16,604 infants between 2002-2018. The rate of any acid suppressing medication increased between 2002-2004. H2RA use was more common than PPI use, especially after 2012. H2RA use appeared to increase after 2015. PPI use initially increased, but then remained relatively stable before appearing to decrease in 2015-2018. Ranitidine clearly had the most users of the H2RA, which was maintained throughout the study period. Lansoprazole had the most users among users of PPIs, but this decreased notably after 2010. Of the infants with a reported PPI, 34.0% (95%CI:23.6-46.3) also had an H2RA reported during the year. In total, 40.8% (95%CI:35.0-46.8) had only 1 reported fill of an acid suppresser. There were 8,075 infants under 2 years of age between 2010-2018. The rate of use was highest in infants between 4-11 months of age at the end of the survey year at 8.6% (95%CI:6.9-10.6). Acid-suppressing medication use was more common among infants in higher income families, White (non-Hispanic) race/ethnicity, infants with private health insurance, parents who reported acid-suppressing medication, maternal anti-depressant use, and certain regions of the country. Discussion: Between 6.9-10.6% of infants used an acid suppressing medication before their second birthday between 2010-2018, of which around 60% had multiple medication fills. It appears that there has been a small increases in use over the last few decades, of which ranitidine appears to be the medication driving the increase. Numerous socio-economic and demographic characteristics were associated with acid suppressing medication use. This study has numerous limitations, including potential under-reporting of acid-suppressing medications, an imperfect identification of parent, unreliable diagnoses (not included in the analysis), and a relatively small sample size. We opted not to run a multivariable logistic regression of socio-economic and demographic characteristics given concern for table 2 fallacy and relatively small sample size. Acid suppressing medication use was common in infants before their second birthday. Additional research should be conducted on efficacy and safety of these medications given the level of use and very low quality of available evidence.


2020 ◽  
Vol 21 ◽  
Author(s):  
Roberto Gabbiadini ◽  
Eirini Zacharopoulou ◽  
Federica Furfaro ◽  
Vincenzo Craviotto ◽  
Alessandra Zilli ◽  
...  

Background: Intestinal fibrosis and subsequent strictures represent an important burden in inflammatory bowel disease (IBD). The detection and evaluation of the degree of fibrosis in stricturing Crohn’s disease (CD) is important to address the best therapeutic strategy (medical anti-inflammatory therapy, endoscopic dilation, surgery). Ultrasound elastography (USE) is a non-invasive technique that has been proposed in the field of IBD for evaluating intestinal stiffness as a biomarker of intestinal fibrosis. Objective: The aim of this review is to discuss the ability and current role of ultrasound elastography in the assessment of intestinal fibrosis. Results and Conclusion: Data on USE in IBD are provided by pilot and proof-of-concept studies with small sample size. The first type of USE investigated was strain elastography, while shear wave elastography has been introduced lately. Despite the heterogeneity of the methods of the studies, USE has been proven to be able to assess intestinal fibrosis in patients with stricturing CD. However, before introducing this technique in current practice, further studies with larger sample size and homogeneous parameters, testing reproducibility, and identification of validated cut-off values are needed.


Author(s):  
Jonah T Hansen ◽  
Luca Casagrande ◽  
Michael J Ireland ◽  
Jane Lin

Abstract Statistical studies of exoplanets and the properties of their host stars have been critical to informing models of planet formation. Numerous trends have arisen in particular from the rich Kepler dataset, including that exoplanets are more likely to be found around stars with a high metallicity and the presence of a “gap” in the distribution of planetary radii at 1.9 R⊕. Here we present a new analysis on the Kepler field, using the APOGEE spectroscopic survey to build a metallicity calibration based on Gaia, 2MASS and Strömgren photometry. This calibration, along with masses and radii derived from a Bayesian isochrone fitting algorithm, is used to test a number of these trends with unbiased, photometrically derived parameters, albeit with a smaller sample size in comparison to recent studies. We recover that planets are more frequently found around higher metallicity stars; over the entire sample, planetary frequencies are 0.88 ± 0.12 percent for [Fe/H] < 0 and 1.37 ± 0.16 percent for [Fe/H] ≥ 0 but at two sigma we find that the size of exoplanets influences the strength of this trend. We also recover the planet radius gap, along with a slight positive correlation with stellar mass. We conclude that this method shows promise to derive robust statistics of exoplanets. We also remark that spectrophotometry from Gaia DR3 will have an effective resolution similar to narrow band filters and allow to overcome the small sample size inherent in this study.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shinya Hosokawa ◽  
Kyosuke Momota ◽  
Anthony A. Chariton ◽  
Ryoji Naito ◽  
Yoshiyuki Nakamura

AbstractDiversity indices are commonly used to measure changes in marine benthic communities. However, the reliability (and therefore suitability) of these indices for detecting environmental change is often unclear because of small sample size and the inappropriate choice of communities for analysis. This study explored uncertainties in taxonomic density and two indices of community structure in our target region, Japan, and in two local areas within this region, and explored potential solutions. Our analysis of the Japanese regional dataset showed a decrease in family density and a dominance of a few species as sediment conditions become degraded. Local case studies showed that species density is affected by sediment degradation at sites where multiple communities coexist. However, two indices of community structure could become insensitive because of masking by community variability, and small sample size sometimes caused misleading or inaccurate estimates of these indices. We conclude that species density is a sensitive indicator of change in marine benthic communities, and emphasise that indices of community structure should only be used when the community structure of the target community is distinguishable from other coexisting communities and there is sufficient sample size.


Sign in / Sign up

Export Citation Format

Share Document