FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2896866045> ?p ?o ?g. }

• W2896866045 endingPage "342" @default.
• W2896866045 startingPage "341" @default.
• W2896866045 abstract "Central MessageLinear mixed models are powerful statistical methods in clinical research but require careful preparation and accurate planning.See Article page 327. Linear mixed models are powerful statistical methods in clinical research but require careful preparation and accurate planning. See Article page 327. In this issue of the Journal, Heo and colleagues1Heo W. Song S.W. Kim T.H. Lee J.S. Yoo K.J. Cho B.K. et al.Differential impact of intimal tear location on aortic dilation and reintervention after acute type I aortic dissection after total arch replacement.J Thorac Cardiovasc Surg. 2019; 158: 327-338.e2Abstract Full Text Full Text PDF Scopus (11) Google Scholar present an interesting study on the impact of the intimal tear on the aortic destiny after total arch replacement (TAR) in repairing type I aortic dissection. The hypothesis of this study was that the location of the intimal tear after surgery would represent the most important factor affecting both the dilatation of the aorta and the incidence of reintervention. To explore this hypothesis, the authors had a dataset of 85 patients who underwent TAR, and 40 of them have been investigated over time with serial computed tomography scans. Therefore, they had to analyze the same variables measured over multiple and different time points: This type of database is defined as a longitudinal dataset and requires a carefully planned statistical analysis. In fact, when evaluating a variable that changes over time, it has to be considered that the observations are not independent of each other and that the individual variance has an impact on the outcome, possibly representing a bias on the overall analysis. Statistical techniques that assume independent observations, such as linear regression, cannot be used in this context, and the need for addressing intra-individual correlation has triggered the development of alternative models.2Twisk J.W.R. Longitudinal data analysis. A comparison between generalized estimating equation and random coefficient analysis.Eur J Epidemiol. 2004; 19: 769-776Crossref PubMed Scopus (236) Google Scholar Linear mixed-effect models are very useful when analyzing longitudinal data, especially when there are missing values, there are more than 2 time points, or there is a need for adjusting for other confounding factors. In this approach, one can account for unobservable differences between individuals by specifying specific effects that can vary over subjects.3Liu X. Introduction to longitudinal data analysis in psychiatric research.Shanghai Arch Psychiatry. 2015; 27: 256-259Google Scholar “Mixed-effect modeling” means that there are 2 type of effects: a fixed effect, which is the one of interest for interpretation of the results, and a random effect, which in most cases is not of interest for interpretation of the results. Therefore, some parameters can vary between subjects (“random”), whereas others cannot change between subjects (“fixed”). For instance, in a longitudinal study, the individual subjects could be considered as a random effect or in a multicenter nonlongitudinal surgical study, the individual center (or even the individual surgeon) might be considered as a random effect. After specifying the random effect in the model, the differences between predicted and observed values of the outcome are considered conditionally independent. In a linear mixed-model effect, both the intercept and the slope can be considered as random: In a random intercept model, we account for baseline differences and assume that the effect of the variables of interest is going to be the same for each individual. In the random slope model, used in Heo and colleagues' study,1Heo W. Song S.W. Kim T.H. Lee J.S. Yoo K.J. Cho B.K. et al.Differential impact of intimal tear location on aortic dilation and reintervention after acute type I aortic dissection after total arch replacement.J Thorac Cardiovasc Surg. 2019; 158: 327-338.e2Abstract Full Text Full Text PDF Scopus (11) Google Scholar the subjects are allowed to have different intercepts, but also different slopes for the effect of the variable of interest. With this type of approach, the authors have been able to clearly define the factor affecting the fate of the aorta after TAR and have identified the importance of the residual tear location in the proximal descending thoracic artery. Their study is also a good example of integration of medical and statistical knowledge, for instance, they have decided to exclude the first follow-up computed tomography scans in the linear mixed model because these scans were performed after a short period of time after the surgery, probably too short to evaluate the stabilization of the intimal flap. This clinical rationale probably would have been missed by a pure statistician. As we can see, multiple different considerations have to be taken into account when planning a linear mixed model for longitudinal data. The correct mixture of random and fixed effect and the use of covariates have to be carefully dosed before introducing them into the final “recipe.” As James Bond would say: Shaken, not stirred. Differential impact of intimal tear location on aortic dilation and reintervention in acute type I aortic dissection after total arch replacementThe Journal of Thoracic and Cardiovascular SurgeryVol. 158Issue 2PreviewThe study objective was to evaluate the differential impact of intimal tear location on aortic dilation and reintervention after total arch replacement for acute type I aortic dissection. Full-Text PDF Open Archive" @default.
• W2896866045 created "2018-10-26" @default.
• W2896866045 creator A5018195873 @default.
• W2896866045 creator A5022606080 @default.
• W2896866045 date "2019-08-01" @default.
• W2896866045 modified "2023-10-18" @default.
• W2896866045 title "Commentary: Linear mixed-effect models in longitudinal data analysis: Shaken not stirred" @default.
• W2896866045 cites W2069764417 @default.
• W2896866045 cites W2897538171 @default.
• W2896866045 doi "https://doi.org/10.1016/j.jtcvs.2018.10.003" @default.
• W2896866045 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30446279" @default.
• W2896866045 hasPublicationYear "2019" @default.
• W2896866045 type Work @default.
• W2896866045 sameAs 2896866045 @default.
• W2896866045 citedByCount "1" @default.
• W2896866045 countsByYear W28968660452020 @default.
• W2896866045 crossrefType "journal-article" @default.
• W2896866045 hasAuthorship W2896866045A5018195873 @default.
• W2896866045 hasAuthorship W2896866045A5022606080 @default.
• W2896866045 hasBestOaLocation W28968660451 @default.
• W2896866045 hasConcept C114614502 @default.
• W2896866045 hasConcept C126838900 @default.
• W2896866045 hasConcept C141071460 @default.
• W2896866045 hasConcept C2775862295 @default.
• W2896866045 hasConcept C2779980429 @default.
• W2896866045 hasConcept C2779993142 @default.
• W2896866045 hasConcept C2780757906 @default.
• W2896866045 hasConcept C2781285907 @default.
• W2896866045 hasConcept C33923547 @default.
• W2896866045 hasConcept C71924100 @default.
• W2896866045 hasConceptScore W2896866045C114614502 @default.
• W2896866045 hasConceptScore W2896866045C126838900 @default.
• W2896866045 hasConceptScore W2896866045C141071460 @default.
• W2896866045 hasConceptScore W2896866045C2775862295 @default.
• W2896866045 hasConceptScore W2896866045C2779980429 @default.
• W2896866045 hasConceptScore W2896866045C2779993142 @default.
• W2896866045 hasConceptScore W2896866045C2780757906 @default.
• W2896866045 hasConceptScore W2896866045C2781285907 @default.
• W2896866045 hasConceptScore W2896866045C33923547 @default.
• W2896866045 hasConceptScore W2896866045C71924100 @default.
• W2896866045 hasIssue "2" @default.
• W2896866045 hasLocation W28968660451 @default.
• W2896866045 hasLocation W28968660452 @default.
• W2896866045 hasOpenAccess W2896866045 @default.
• W2896866045 hasPrimaryLocation W28968660451 @default.
• W2896866045 hasRelatedWork W2006810533 @default.
• W2896866045 hasRelatedWork W2012639740 @default.
• W2896866045 hasRelatedWork W2033152741 @default.
• W2896866045 hasRelatedWork W2121972189 @default.
• W2896866045 hasRelatedWork W2260504642 @default.
• W2896866045 hasRelatedWork W2332519438 @default.
• W2896866045 hasRelatedWork W2409978114 @default.
• W2896866045 hasRelatedWork W2432765816 @default.
• W2896866045 hasRelatedWork W2914276787 @default.
• W2896866045 hasRelatedWork W4283320624 @default.
• W2896866045 hasVolume "158" @default.
• W2896866045 isParatext "false" @default.
• W2896866045 isRetracted "false" @default.
• W2896866045 magId "2896866045" @default.
• W2896866045 workType "article" @default.