[Multiple regression]

Regression Analyses and Their Particularities in Observational Studies

BACKGROUND

Regression analysis is a standard method in medical research. It is often not clear, however, how the individual components of regression models are to be understood and interpreted. In this article, we provide an overview of this type of analysis and discuss its special features when used in observational studies.

METHODS

Based on a selective literature review, the individual components of a regression model for differently scaled outcome variables (metric: linear regression; binary: logistic regression; time to event: Cox regression; count variable: Poisson or negative binomial regression) are explained, and their interpretation is illustrated with respect to a study on multiple sclerosis. The prerequisites for the use of each of these models, their applications, and their limitations are described in detail.

RESULTS

Regression analyses are used to quantify the relation between several variables and the outcome variable. In randomized clinical trials, this flexible statistical analysis method is usually lean and prespecified. In observational studies, where there is a need to control for potential confounders, researchers with knowledge of the topic in question must collaborate with experts in statistical modeling to ensure high model quality and avoid errors. Causal diagrams are an increasingly important basis for evaluation. They should be constructed in collaboration and should differentiate between confounders, mediators, and colliders.

CONCLUSION

Researchers need a basic understanding of regression models so that these models will be well defined and their findings will be fully reported and correctly interpreted.

Brewers' spent grain liquor as a feedstock for lactate production with Lactobacillus delbrueckii subsp. lactis

Brewers' spent grain (BSG) is a low-cost by-product of the brewing process. BSG liquor names the liquid components of BSG, mainly glucose, maltose, and long-chain α-1,4-glycosidic bond glucose oligomers. These substances should be separated in existing BSG biorefineries, as they might lead to an increased formation of microbe-inhibiting compounds in well-established hydrothermal/enzymatic saccharification processes. In most cases, this liquid fraction is discarded. The present study presents for the first time an optimized process with BSG liquor for the purpose of producing bulk chemicals (e.g., lactate) in relevant concentrations. The process comprises the application of yeast extract, produced from own brewing processes, as the sole supplemented complex constituent in a simultaneous fermentation and saccharification process. Kinetic parameters for the final optimized process conditions with the organism Lactobacillus delbrueckii subsp. lactis were: maximum specific growth rate µmax  =  0.47 h-1, maximum lactate concentration cLac, max  =  79.06 g L-1, process yield YPS  =  0.89 gLac gSugar -1, lactate production rate qP  =  4.18 gLac gCDW -1 h-1, and productivity P 15 h  =  4.93 gLac L-1 h-1. BSG liquor, linked with yeast extract from Brewers' yeast, can be a promising substrate for further bioprocess engineering tasks and contribute to a holistic and sustainable usage of Brewers' spent grain.

Data analysis of epidemiological studies: part 11 of a series on evaluation of scientific publications

INTRODUCTION

An important objective of epidemiological research is to identify risk factors for disease. Depending on the particular question being asked, cohort studies, case-control studies, or cross-sectional studies are conducted.

METHODS

Methods of data analysis in different types of epidemiological studies are illustrated through examples with fictive data. Important measures of frequency and effect will be introduced. Different regression models will be presented as examples of complex analytical methods.

RESULTS

Important frequency measures in cohort studies are incidence and mortality. Important effect measures such as the relative risk (RR), hazard ratio (HR), standardized incidence ratio (SIR), standardized mortality ratio (SMR), and odds ratio (OR) can also be calculated. In case-control or cross-sectional studies, the OR can be calculated as an effect measure. In cross-sectional studies, prevalence is the most important frequency measure. The interpretation of different frequency measures and effect measures will be discussed.

CONCLUSION

The measures to be calculated and the analyses to be performed in an epidemiological study depend on the research questions being asked, the study type, and the available data.

Psychosocial determinants of successful voice rehabilitation after laryngectomy

This study gives an overview of publications on factors that are associated with the outcome of voice rehabilitation after laryngectomy. A systematic literature review was conducted. Fifty-six manuscripts were analyzed regarding the parameters investigated, the number of participants included, the study designs used, the assessment instruments, and the results. A broad range of factors were considered to contribute to successful voice rehabilitation, whereby psychosocial attributes related to success are mentioned as often as medical and treatment-related characteristics. The results of the studies are mostly inconsistent. It can be concluded, however, that active communication behavior, employment status, type of alaryngeal speech and the general physical condition are associated with rehabilitation outcomes, whereas alcohol consumption is not. More comprehensive prospective studies are needed which analyze the impact of psychosocial factors with validated and standardized instruments. A large sample size would be necessary to calculate all possibly relevant factors and their interaction. Clinicians should be careful about considering their patients to be "unmotivated" if the rehabilitation fails; instead, they should encourage them to communicate actively and take part in social activities.