Continuous Outcome Variables: Regression-based Methods

Research has demonstrated erosion of empathy in students during medical education. Particularly, U.S. studies have demonstrated empathy declines during clinical training in the third and fourth year of traditional medical programs. Yet, studies conducted outside the U.S. have not confirmed this trend. Timing and extent of patient interactions have been identified as empathy-protective factors. The need to examine empathy within different learning contexts has been noted, as has the need for longitudinal and time-series research designs to analyze trajectories. Between fall 2010 and spring 2019, we assessed empathy longitudinally among six student cohorts (N = 493) at a U.S. medical school, where patient interaction occurs early and throughout an integrated curriculum. Empathy levels of students in each cohort were assessed at five time points utilizing the Jefferson Scale of Physician Empathy-Student version. We hypothesized empathy levels will not degrade by program end, and trajectories will not show patterns of decline in Years Three and Four. Analysis of Variance (ANOVA) and Linear Mixed Model (LMM) analyses were used to analyze differences at baseline and changes in empathy trajectories. ANOVA analyses revealed statistically significant differences at baseline by class cohort (F(5, 487) = [23.28], p < 0.001). LMM analyses indicated empathy was either significantly higher or not different at the end of the program (F(19, 1676) = [13.97], p < 0.001). Empathy trajectories varied among cohorts; yet, none resulted in an overall empathy decline by the end of the program. Findings demonstrate empathy in U.S. medical students can be unchanged or higher by the end of medical education. Outcomes are consistent with reports of non-declining medical student empathy outside the U.S. and support the notion of context-specificity. Results further support recent research, suggesting decreases in empathy during training can stabilize or increase by program end. These findings have important implications for future empathy research context and design considerations, as well as program planning.

Association mapping is an alternative to mapping in a biparental population. A key to successful association mapping is to avoid spurious associations by controlling for population structure. Confirming the marker/trait association in an independent population is necessary for the implementation of the marker in other genetic studies. Two independent soybean populations consisting of advanced breeding lines representing the diversity within maturity groups 00, 0, and I were screened in multi-site, replicated field trials to discover molecular markers associated with iron deficiency chlorosis (IDC), a major yield-limiting factor in soybean. Lines with extreme phenotypes were initially screened to identify simple sequence repeat (SSR) markers putatively associated with the IDC. Marker data collected from all lines were used to control for population structure and kinship relationships. Single factor analysis of variance (SFA) and mixed linear model (MLM) analyses were used to discover marker/trait associations. The MLM analyses, which include population structure, kinship or both factors, reduced the number of markers significantly associated with IDC by 50% compared with SFA. With the MLM approach, three markers were found to be associated with IDC in the first population. Two of these markers, Satt114 and Satt239, were also found to be associated with IDC in the second confirmation population. For both populations, those lines with the tolerance allele at both these two marker loci had significantly lower IDC scores than lines with one or no tolerant alleles.

This paper presents a Bayesian analysis of linear mixed models for quantile regression using a modified Cholesky decomposition for the covariance matrix of random effects and an asymmetric Laplace distribution for the error distribution. We consider several novel Bayesian shrinkage approaches for both fixed and random effects in a linear mixed quantile model using extended penalties. To improve mixing of the Markov chains, a simple and efficient partially collapsed Gibbs sampling algorithm is developed for posterior inference. We also extend the framework to a Bayesian mixed expectile model and develop a Metropolis–Hastings acceptance–rejection (MHAR) algorithm using proposal densities based on iteratively weighted least squares estimation. The proposed approach is then illustrated via both simulated and real data examples. Results indicate that the proposed approach performs very well in comparison to the other approaches.

Background: Our aim was to examine the impact of early and late surveillance biopsy findings on kidney allograft functional outcome. Many inflamed biopsies with changes insufficient to be called T-Cell-mediated rejection are labeled “borderline “and its implication on graft outcome remains uncertain. Methods: We studied 158 renal transplant recipients between 2004 to 2012; mean follow up was 5.2 (standard deviation: 0.24) yrs. Early protocol biopsy (B1) and late protocol biopsy (B2) were performed between 3-12 and 13-24 months (m) respectively. There were 102 patients who had both biopsies. Biopsies were classified as: interstitial fibrosis and/or tubular atrophy only (IFTA), normal histology (NML), acute rejection (AR), borderline rejection (BDR). Baseline graft function by estimated glomerular filtration rate (eGFR) was fixed at 12m while changes were evaluated up to last follow up eGFR. Univariate and multivariable linear mixed effect models were developed to examine the effects of surveillance biopsies on changes in the eGFR. Random intercepts and random time effects were considered in linear mixed model analysis. All statistical analyses were carried out using SAS 9.3. Results: The eGFR rate declined from 12m to last visit was found to be -1.26 ± 0.40(standard error) per year. Multivariate analysis revealed that eGFR at 3 m, donor age, and delayed graft function were associated with baseline eGFR while B1 and B2 findings were significantly associated with both baseline eGFR and change in eGFR over time. Final multivariate analysis of both B1 and B2 taken together showed that the change in eGFR was found to be associated with findings of IFTA in B1 (p=0.0038), AR in B2 (p=0.0605) and BDR in B2 (p=0.049) when compared to NML. All three abnormal histology groups had lower eGFR (AR=25.65 ±14.28 ml/min; p=0.1029), (BDR=42.17±19.24; p=0.016), (IFTA 45.26±16.13; p=0.016), than those with (NML=51.77±16.96 ml/min) at last follow up (p<0.05). Conclusion: Early biopsy results correlate with baseline GFR at 12m; while abnormal late surveillance biopsy results significantly impact GFR decline over long-term. In addition, the diagnosis of borderline inflammation on surveillance biopsy affected graft functional status on long-term follow up.

In this study, the longitudinal relation between deliberate practice and performance in chess was examined using a linear mixed models analysis. The practice activities and performance ratings of young elite chess players, who were either in, or had dropped out of the Dutch national chess training, were analysed since they had started playing chess seriously. The results revealed that deliberate practice (i.e. serious chess study alone and serious chess play) strongly contributed to chess performance. The influence of deliberate practice was not only observable in current performance, but also over chess players' careers. Moreover, although the drop-outs' chess ratings developed more slowly over time, both the persistent and drop-out chess players benefited to the same extent from investments in deliberate practice. Finally, the effect of gender on chess performance proved to be much smaller than the effect of deliberate practice. This study provides longitudinal support for the monotonic benefits assumption of deliberate practice, by showing that over chess players' careers, deliberate practice has a significant effect on performance, and to the same extent for chess players of different ultimate performance levels. The results of this study are not in line with critique raised against the deliberate practice theory that the factors deliberate practice and talent could be confounded.

[This corrects the article DOI: 10.1098/rsos.182237.].

This chapter describes the facilities for estimation of variance components and analysis of linear mixed models using the method of residual maximum likelihood (REML), sometimes also known as restricted maximum likelihood. The REML algorithm estimates the treatment effects and variance components in a linear mixed model: that is, a linear model with both fixed and random effects. Like regression, REML can be used to analyse unbalanced data sets; but, unlike regression, it can also account for more than one source of variation in the data, providing an estimate of the variance components associated with the random terms in the model.

BackgroundFunctional decline associated with dementia, including in Alzheimer’s disease (AD), is not uniform across individuals, and respective heterogeneity is not yet fully explained. Such heterogeneity may in part be related to genetic variability among individuals. In this study, we investigated whether the SNP rs6859 in nectin cell adhesion molecule 2 (NECTIN2) gene (a major risk factor for AD) influences trajectories of cognitive decline in older participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI).MethodsWe retrospectively analyzed records on 1310 participants from the ADNI database for the multivariate analysis. We used longitudinal measures of Mini-Mental State Examination (MMSE) scores in participants, who were cognitively normal, or having AD, or other cognitive deficits to investigate the trajectories of cognitive changes. Multiple linear regression, linear mixed models and latent class analyses were conducted to investigate the association of the SNP rs6859 with MMSE.ResultsThe regression coefficient per one allele dose of the SNP rs6859 was independently associated with MMSE in both cross-sectional (-2.23, p < 0.01) and linear mixed models (-2.26, p < 0.01) analyses. The latent class model with three distinct subgroups (class 1: stable and gradual decline, class 2: intermediate and late decline, and class 3: lowest and irregular) performed best in the posterior classification, 42.67% (n = 559), 21.45% (n = 281), 35.88% (n = 470) were classified as class 1, class 2, and class 3. In the heterogeneous linear mixed model, the regression coefficient per one allele dose of rs6859 – A risk allele was significantly associated with MMSE class 1 and class 2 memberships and related decline; Class 1 (-2.28, 95% CI: -4.05, -0.50, p < 0.05), Class 2 (-5.56, 95% CI: -9.61, -1.51, p < 0.01) and Class 3 (-0.37, 95% CI: -1.62, 0.87, p = 0.55).ConclusionsThis study found statistical evidence supporting the classification of three latent subclass groups representing complex MMSE trajectories in the ADNI cohort. The SNP rs6859 can be suggested as a candidate genetic predictor of variation in modeling MMSE trajectory, as well as for identifying latent classes with higher baseline MMSE. Functional studies may help further elucidate this relationship.

BackgroundThe utilization of ultrapure dialysate has been shown to decrease dialysate contamination and mitigate inflammatory responses. The central dialysate delivery system (CDDS) has the potential to attain a level of purity similar to ultrapure dialysate. Nevertheless, there is limited research examining the impact of CDDS on inflammation in comparison to single-patient dialysis fluid delivery system(SPDDS). This study aims to investigate the effects of CDDS utilizing ultrapure dialysate on ameliorating the microinflammatory state in hemodialysis patients.MethodA retrospective cohort clinical study enrolled a total of 125 hemodialysis patients, with 58 patients from the CDDS unit and 67 patients from the SPDDS unit. Each participant was monitored for a period of 6 months, and the repeated measurement data was analyzed using a generalized linear mixed models (GLMM).ResultsThe average age of the studty cohort was 56.22 ± 12.64 years. The GLMM analysis showed a significant time*group interaction effect on hs-CRP changes over the follow-up period (β = -1.966, FTime* CDDS group = 13.389, P < 0.001). A linear mixed model analysis with random slope showed that a different slope was observed between CDDS group and SPDDS group (βCDDS =—0.793; βSPDDS = 0.791), indicating a decreased hs-CRP levels in CDDS group, while increased in the SPDDS group over the follow-up period. However, no significant time*group interaction effect were observed on albumin and β2-microglobulin levels during follow-up period(β2-microglobulin: β = -0.658, FTime* CDDS group = 1.228, P = 0.269; albumin: β = 0.012, FTime* CDDS group = 1.429, P = 0.233).ConclusionUsing ultrapure dialysate in the CDDS is associated with an improvement in hs-CRP levels compared to standard dialysate, which might confer long-term clinical advantages.

To examine income-related disparities in health-related quality of life (HRQOL) over a one-year period after surgery (radical prostatectomy) and its contributory factors in a longitudinal perspective. Evidence of associations between income and HRQOL among patients with prostate cancer (PCa) is sparse and their explanations still remain unclear. 246 males of two German hospitals filled out a questionnaire at the time of acute treatment, 6 and 12months later. Age, partnership status, baseline disease and treatment factors, physical and psychological comorbidities, as well as treatment factors and adverse effects at follow-up were additionally included in the analyses to explain potential disparities. HRQOL was assessed with the EORTC (European Organisation for Research and Treatment of Cancer) QLQ-C30 core questionnaire and the prostate-specific QLQ-PR25. A linear mixed model for repeated measures was calculated. The fixed effects showed highly significant income-related inequalities regarding the majority of HRQOL scales. Less affluent PCa patients reported lower HRQOL in terms of global quality of life, all functional scales and urinary symptoms. After introducing relevant covariates, some associations became insignificant (physical, cognitive and sexual function), while others only showed reduced estimates (global quality of life, urinary symptoms, role, emotional and social function). In particular, mental disorders/psychological comorbidity played a relevant role in the explanation of income-related disparities. One year after surgery, income-related disparities in various dimensions of HRQOL persist. With respect to economically disadvantaged PCa patients, the findings emphasize the importance of continuous psychosocial screening and tailored interventions, of patients' empowerment and improved access to supportive care.

AimsTo compare the surgical outcomes of exotropia patients with cerebral palsy (CP) and controls without CP.Methods30 patients with exotropia and CP and 60 age-matched controls without CP who underwent bilateral...

Transplant centers hesitate to transplant patients with cognitive impairment. It is unclear if pre-kidney transplant (KT) cognitive screening can predict post-KT cognitive function. We evaluated pre- to post-KT cognitive function with theMontreal Cognitive Assessment (MoCA) in a cohort of 108 patients. We used an adjusted logistic regression model to assess pre- to post-KT changes in cognitive status (continuous variable) and a linear mixed model to assess changes in MoCA scores (categorical variable) pre- to post- KT. The average pre- and post-KT MoCA scores were 25.3±3.0 and 26.4±2.8, respectively. Final pre-KT score did not predict post-KT cognitive status (OR=1.08; 95% CI: .92-1.26; P=.35). 32% of the patients with a final pre-KT score ≥26 had at least one post-KT score<26. Conversely, 61% of the patients with a final pre-KT score<26 had at least one post KT score ≥26. In the linear mixed model analysis, the final pre-KT score was associated with a small, clinically insignificant (β=.34; 95% CI: .19-.49; P<.001) effect on the post-KT score. A low pre-KT MoCA score is not a strong independent predictor of post-KT cognitive function and should not preclude patients from receiving a KT.

Longitudinal study of impact of medication for opioid use disorder on Hamilton Depression Rating Scale

Multi-level safety climate associations with safety behaviors in the fire service

BackgroundIntegrase strand transfer inhibitor (INSTI) associated weight gain has been observed in a number of recent studies but with limited data on bictegravir. Here we examine weight change associated with the switch to co-formulated bictegravir/emtricitabine/tenofovir alafenamide (B/F/TAF).MethodsWe performed a retrospective analysis of consecutive PWH attending an academic outpatient clinic who received a prescription for B/F/TAF from 02/07/2018-02/07/2019 and had a baseline HIV RNA < 200 copies/mL prior to starting B/F/TAF. Baseline demographics and clinical parameters were obtained from chart review. Parameters of interest were collected for one year (at a minimum) before and one year after starting B/F/TAF. Linear mixed model analyses were conducted for PWH before/ after switch. Separate analyses were performed examining factors associated with ≥ 10% BMI increase versus < 10% increase.Results156 PWH switching to B/F/TAF were identified, of whom 107 (69%) identified as men, 105 (67%) were African American. Median age was 49 years (IQR 35-57), weight 184 lb (IQR 153-208), and BMI 27.5 (IQR 23-32.3). At time of switch, 3% were underweight, 31% normal weight, 24% overweight, and 41% obese. 74% switched from INSTI-based regimen, 17% from NNRTI- and 16% from PI-based regimens. Of the INSTI, elvitegravir (54.3%) or dolutegravir (41.7%) were most frequently used. 50% were on TAF pre-switch with 28% on tenofovir disoproxil. The mixed model analysis indicated that there was not a significant shift in the mean BMI (P=0.2017) or BMI rate of change over time (P=0.792) after participants switched. 19.2% had ≥ 10% increase in BMI; and when compared to those with < 10% increase, younger age (42.8±13.8 vs. 48.9±13.2 years, P=0.036), switch from a non-PI based regimen (P=0.004), and switch from a TDF containing regimen (36.4% vs. 12.6%, P < 0.001) were associated with greater weight gain.ConclusionOverall, there were no significant changes in BMI between pre and post switch to B/F/TAF time periods; however the majority of PWH switched from an INSTI-based regimen. Analysis of PWH who experienced ≥ 10% increase compared to < 10% BMI increase, indicated that factors including younger age, switch from a non-PI containing regimen and switch from TDF were associated with greater weight gain.DisclosuresAll Authors: No reported disclosures