Mixed Linear Regression Models Research Articles

Correlations of Corneal Endothelial Morphology and Corneal Thickness With Anterior Segment Parameters in Healthy Individuals.

The purpose of this study was to investigate the associations between central corneal endothelial cell density (ECD), endothelial morphology, and corneal thickness (central corneal thickness) with the anterior chamber depth, corneal volume (CV), white-to-white (WTW) distance, mean anterior chamber angle (CAmean), and gender in healthy individuals. This observational study included 136 healthy volunteers. The ECD, coefficient of variation of cell area, and hexagonal cell appearance ratio (%Hex) were measured by means of specular microscopy. The central corneal thickness, anterior chamber depth, CV, WTW distance, and the angle width of 12 points were taken by the Pentacam HR Scheimpflug anterior segment imaging. The arithmetical mean of the 12 points was considered as the CAmean. We used mixed effect linear regression model for the statistical analysis of the data. ECD was positively correlated with CV ( P = 0.028), while after adjusting for age, it was negatively correlated with age ( P < 0.001). Coefficient of variation of cell area was positively correlated with CAmean ( P = 0.036), while after adjusting for age, it was positively correlated with age ( P < 0.001) and CAmean ( P = 0.005). Hex was negatively correlated with WTW ( P = 0.023) and CAmean ( P = 0.025), and after adjusting for age, this correlation remained the same ( P = 0.029 when correlated with WTW and P = 0.035 with CAmean). There were significant changes in the morphology of the corneal endothelial cells in eyes with wider anterior chamber angle. Greater pleomorphism and polymegethism of the corneal endothelium was observed in healthy patients with wider CAmean. Deepening of the anterior chamber as myopia progresses could render the corneal endothelium more fragile and susceptible to mechanical stress, which is an area worthy of further study.

Weight loss outcomes with semaglutide based on diabetes severity using the individualized metabolic surgery score

Semaglutide demonstrated inferior weight loss responses in patients with type 2 diabetes (T2D) compared to patients with obesity without T2D. The individualized metabolic surgery (IMS) score was validated to predict T2D remission after bariatric surgery. The parameters of the IMS are HbA1c (<7%), insulin use, T2D medications and T2D duration. We aim to assess weight loss outcomes of semaglutide based on IMS score in patients with obesity and T2D. This is a retrospective multicentered cohort study of patients with T2D and BMI≥ 27kg/m2 taking ≥1mg of semaglutide recruited from January 2020 to December 2022. We excluded patients with a history of bariatric surgery or taking other anti-obesity medications. IMS was calculated at baseline and patients weight change was recorded at baseline, 3, 6, 9 and 12 months. IMS was classified as mild (0-24.9 points), moderate (25-94.9 points), and severe (95-180 points). Analysis was performed based on IMS score quartiles and combination of Mild-Moderate vs Severe categories. We performed mixed linear regression models including age, sex, and baseline weight to assess associations between IMS categories with total body weight loss percentage (TBWL%). We included 297 patients (42% female, mean age 62±12 years) in the analysis. At 12 months, there was a stepwise decrease in weight loss outcomes when comparing patients by IMS quartiles (LS mean TBWL%± SE): 8.8±0.8% vs 6.9±0.8% vs 5.7±0.9% vs 5.0±0.8%. In the mixed linear model, patients in the mild-moderate category achieved significantly superior weight loss outcomes (LS mean TBWL± SE:-8.3±0.7%) than patients in the severe category (-5.5±0.6%; difference:-2.9, 95% CI:-5.2 to-0.5, p=0.006) at 12 months. There was no significant difference in glycemic improvement regardless of IMS severity at baseline. In our cohort, lower IMS severity was associated with more weight loss in patients with obesity and T2D. Further studies are needed to understand T2D severity and its effect on semaglutide outcomes. Beyond payment to the research staff by Mayo Clinic, this research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Effect of age of first exposure on L2 contextual lexical semantic learning: an ERP investigation

Abstract Age of first exposure (AoFE) is an important factor that influences the quality of L2 acquisition. This study aims to investigate the AoFE effect on the contextual learning of L2 novel words at the neural level, as measured by the N400 component from event-related potentials (ERPs). Eighty-eight participants were recruited for the experiment of L2 pseudoword learning, which includes a learning session and a testing session. The participants’ EEG data were recorded from the testing session, and the N400 effect was derived from target words that were either congruous or incongruous with the context. The linear mixed model and multiple regression model revealed a positive AoFE effect on the N400 effect in discourses that were designed for testing retrieval of episodic and semantic memory even after accounting for the variance contributed by several confounding factors. In addition to AoFE, the effects of total L2 exposure, L2 proficiency and personality on the L2 novel word learning performance indicated by the N400 effect were also confirmed in the statistical results.

#376 Self-efficacy, social support and hemodiafiltration are independently associated with better Self-Reported Health Status over time

Abstract Background and Aims Patients with kidney disease receiving dialysis are a highly vulnerable population. Psychosocial determinants of Self-Efficacy (SE) and Social Support (SS) have been suggested as potentially important outcomes for people with end stage kidney disease. We investigated whether these variables could explain differences in Health-Related Quality of Life (HRQL) outcome domains over time independently or in interaction with treatment modality of high-flux hemodialysis (HD) versus high-dose hemodiafiltration (HDF). Method HRQL in 1264 participants from the international multi-centre, prospective, randomized, open label, controlled CONVINCE trial were assessed, using screening and quarterly follow-up data up to 24 months. Measures included the MOS Social Support Scale, General Self-Efficacy Scale, and PROMIS short forms for fatigue, physical function, pain interference and intensity, sleep disturbance, anxiety, cognitive function, depression, ability to participate in social roles. SE, SS and interactions with modality were included as independent variables and HRQL domains as dependent variables in a linear mixed effects model with random intercept and slope for SE and SS. Results Participants were 20-92 (M=62.33, SD=13.50) years old, 62.9% were male. Linear mixed effect regression models showed significant main effects for self-efficacy (β = −0.15-1.09, all p &amp;lt; .0001) on all HRQL domains. Significant SE × time interactions were found for all variables (β = −0.02-0.04, all p &amp;lt; .01) but pain intensity. Main effects for SS were found for cognition (β = 0.31, p = .00) and sleep disturbance (β = −0.21, p = .00). A significant SS × time interaction was found for anxiety (β = 0.01, p = .02) and sleep disturbance (β = 0.02, p = .00). Plots for these associations can be found in Figs 1 and 2. Marginal means were plotted and 25%, 50% and 75% quantiles were used to separate the different levels of SE and SS. There were no significant interactions between treatment modality (HDF vs. HD) and SE/SS. Conclusion Our results indicate that higher SE and better SS are independently associated with higher HRQL of dialysis patients over time, showing no significant interactions with HDF. SE had the largest effects on HRQL. This bears the chance to develop targeted psychosocial interventions to improve health outcomes when introducing new therapy options for patients. However, patients achieving higher HRQL over the course of the treatment may also experience higher SE and SS over time. Future research should include longitudinal assessments on these psychosocial determinants to evaluate their effect further.

#1703 Role of early AV fistula creation among elderly and its impact on estimated decline in GFR among pre dialysis patients

Abstract Background and Aims Recent evidence suggests that AVF creation may slow estimated glomerular filtration rate (eGFR) decline. The aim of the present study was to assess the impact of the pre dialysis AVF creation on eGFR decline and the need for dialysis, after controlling for key confounding factors among elderly population with advanced renal failure . Method A retrospective cohort study among patients attending Nephrology, Urology and General Surgery OPD between Jan 2019-Jan 2022 in Tertiary care hospital, Lucknow for AV fistula creation and assessed their stage of CKD and trajectory of fall in eGFR within a year's time and the time to initiation of dialysis from AV fistula creation was assessed. Results 34 men and 23 females (&amp;gt; 65 years ) of age who at time of AV fistula creation had an median eGFR at time of AVF creation of 15.5 (10.3–16.9) mL/min/1.73m2. The factors like frailty, diabetes, psychosocial issues were matched between the two genders . Males were more in age with better BMI .Baseline eGFRs were 14.4± 2.0 mL/min/1.73m2 in men and 13.8.0 ± 1.2 mL/min/1.73m2 in women. The crude annual eGFR decline rates were − 3.20 ± 2.50 mL/min/1.73 m2 pre- and − 2.45 ± 2.16 mL/min/1.73 m2 post-AVF. In a mixed effect linear regression model, monthly eGFR decline was − 0.43 mL/min/1.73m2/month.. The period after AVF creation was associated with a relatively higher eGFR . There was a significant association between follow-up time and the period pre/post AVF such that eGFR decline was more attenuated each month after AVF creation. There was better preservation of GFR from the pre AV fistula creation . Conclusion In this cohort, AVF creation was associated with a significant reduction of eGFR decline among elderly population . Further prospective studies are needed to confirm this association

#2969 One-year longitudinal changes in nutritional parameters associated with low-protein diets in patients starting with incremental PD

Abstract Background and Aims Starting peritoneal dialysis (PD) with an incremental approach, that is, a low dose of PD (1-2 dwells a day in continuous ambulatory PD and until 5 sessions a week in automated PD) based on residual kidney function, is a common strategy to increase the acceptance of dialysis for patients affected by advanced chronic kidney disease (CKD) requiring renal replacement therapy. However, no study has provided evidence on the optimal protein intake in patients treated with a low dose of PD. Indeed, patients starting dialysis require an increased protein intake to supply protein loss by dialysis. On the other hand, patients beginning with a low dose of dialysis and a high protein intake may lead to toxins accumulation and metabolic acidosis. Therefore, the maintenance of protein restriction could be indicated as long as malnutrition does not occur. Method The I-COPE study was designed to assess in 24 PD units in Italy the time to full-dose PD in patients undergoing integrated conservative therapy with low-dose PD. Patients who reached full-dose, switched to hemodialysis, received transplant, or died were censored. No specific recommendation on protein intake was included in the study protocol due to the lack of guidelines on this issue, so the prescription was left to decision of each participating center. This is an ad-interim analysis of the I-COPE study aimed at evaluating the longitudinal changes at one year of BMI and serum albumin. Two groups of nutritional interventions were identified a posteriori: non-low protein diet (NLPD, dietary protein intake &amp;gt; 0.6 g/Kg/day) vs low protein diet (LPD, dietary protein intake ≤ 0.6 g/Kg/day) on two visits after baseline (months 6 and 12). We used a mixed linear regression model to evaluate changes in variables over time in the two diet groups, assuming an unstructured covariance matrix; this was done to consider the correlation between repeated measures and missing points for patients not completing the analysis. Results The I-COPE study included 222 patients (age 63.6 ± 14.2 ys, males 65.3%, diabetes 29.2%). LPD was prescribed in 46.3%, while NLPD in 53.7%. At baseline, no intergroup difference was reported for age (P = 0.51), gender (P = 0.94), and diabetes (P = 0.11). 150 patients had at least an evaluation after baseline. The mixed regression model showed no difference in BMI (P = 0.668) and serum albumin (P = 0.297) adjusted for age, gender, diabetes, and GFR at the start of dialysis. As depicted in Fig. 1, serum albumin level was maintained at 12 months in either group. Conclusion We provide first-time observation that LPD may be safely associated with stable serum albumin and BMI over one year of incremental PD treatment. Therefore, in patients with advanced CKD requiring renal replacement therapy, LPD combined with a low dose of PD may provide a suitable nutritional approach to reduce waste metabolic products without increasing the risk of malnutrition.

#2715 Towards precise prediction of prognosis in ADPKD: a comprehensive biomarker analysis

Abstract Background and Aims Determining prognosis in patients with autosomal dominant polycystic kidney disease (ADPKD) heavily relies on the height-adjusted total kidney volume (HtTKV) Mayo classification. This classification predicts kidney outcomes on a group level but lacks precision in individual patients. Recently, several urine and blood biomarkers have demonstrated prognostic value in ADPKD, independent of HtTKV and baseline eGFR. This study compares the prognostic value of various biomarkers to aid selection when developing novel risk stratification tools. Method Included were participants of the DIPAK studies with a follow-up duration of &amp;gt;1 year and ≥3 eGFR measurements. We excluded patients with diabetes mellitus and patients using tolvaptan or somatostatin analogues. At baseline, urine samples were analyzed for albumin, urea, creatinine, β2-microglobulin (β2-MG), monocyte chemotactic protein-1 (MCP-1), TmP/GFR (tubular maximum reabsorption rate of phosphate to GFR), and blood was drawn to measure creatinine, urea (for assessment of urinary-to-plasma urea ratio; UP urea), glucagon, fibroblast growth factor 23 (FGF-23) and copeptin. Albumin, β2-MG and MCP-1 were divided by urinary creatinine concentration to account for concentration effects. The cross-sectional and longitudinal association with baseline eGFR and eGFR decline over time was analyzed using multivariate linear regression and repeated measures linear mixed model regression. Furthermore, logistic regression models were tested to identify patients with rapidly-progressive disease (eGFR slope ≤ −3.0 ml/min/1.73 m2/year). Lastly, multivariate Cox regression analysis was conducted to evaluate the time to the combined endpoint of 30% decline in eGFR, reaching an eGFR ≤ 15 ml/min/1.73 m2, initiation of dialysis or receiving a kidney transplant. Results 565 patients were included with a mean age of 47 ± 11 years, eGFR of 64 ± 27 ml/min/1.73 m2 and HtTKV of 858 [IQR: 540, 1305] ml/m. During a mean follow-up of 5.2 ± 2.2 years, the mean eGFR decline was −3.38 ± 2.37 mL/min/1.73 m2/year. Of these patients, 283 (50.1%) had rapidly progressive disease and 267 (47.3%) reached the combined kidney endpoint. All biomarkers except glucagon were associated with baseline eGFR and eGFR decline over time after adjustment for established clinical risk factors for disease progression (sex, PKD mutation and baseline age and eGFR). After adjustment for the same covariates, albumin/creatinine, MCP-1/creatinine, FGF-23, UP-urea and copeptin were predictive of rapid disease progression. The area under the curve (AUC) value for predicting rapidly progressive disease for the Mayo HtTKV classification alone was 0.694, which could be improved to 0.766 (P &amp;lt; .001) by a risk prediction model that also included the aforementioned predictive clinical risk factors. The introduction of several biomarkers further improved this clinical model, including albumin/creatinine (AUC 0.794, p = 0.03), MCP-1/creatinine (AUC 0.804, p = 0.04) and copeptin (AUC 0.789, p = 0.05). Similar results were seen in multivariate Cox regression models after adjustment for covariates, with independent predictors of the combined endpoint being again albumin/creatinine (P &amp;lt; .001), MCP-1 creatinine (P &amp;lt; .001), copeptin (p = 0.012), but now also UP-urea (p = 0.008) and FGF-23 (P &amp;lt; .001). These findings appeared robust in subgroup analyses of 200 patients with Mayo HtTKV class 1C and 122 patients with early disease (age ≤ 40 years and eGFR ≥ 60 ml/min/1.73 m2 at baseline), in whom predicting prognosis is most challenging. Conclusion In patients with ADPKD, albumin/creatinine, MCP-1/creatinine, FGF-23 and copeptin are consistent predictors of kidney outcomes after adjustment for established clinical risk factors. Our data suggest that these biomarkers should be prioritized when developing novel multimarker models to enhance precision of risk prediction in ADPKD.

Major Pancreatic Resection Increases Bone Mineral Density Loss, Osteoporosis, and Fractures.

To assess whether long-term survivors of pancreatic surgery show increased risk to develop impaired bone mineral density, osteoporosis, and vitamin D deficiency. Pancreatic resection poses a risk for malabsorption of fat-soluble vitamins and other micronutrients essential for bone mineralization. Here, we evaluated the long-term effects of pancreatic resection on bone mineral density (BMD) and its clinical sequelae. This was a two-pronged analysis of post-pancreatectomy patients with a follow-up period greater than 3 years comprising (1) a large, propensity score-matched, cohort study based on a multinational federated research network (FRN) and (2) a retrospective single institution review of clinical and radiographic patient data. In the FRN analysis, an initial cohort of 8,423 post-pancreatectomy patients were identified and propensity score-matched with normal controls. The primary endpoint was the 10-year risk of developing osteoporotic pathological fractures and secondary endpoints included diagnosis of osteoporosis, vitamin-D deficiency, and related therapies. The single institution retrospective analysis identified 224 patients who underwent pancreatic resection between 2005 and 2019. BMD was quantified in CT images acquired before and after surgery. BMD trends and related factors were assessed in a time-series mixed effect linear regression model. A total of 8,080 propensity score-matched pairs were included in the FRN analysis. The analysis revealed a 2.4-fold increase in pathological fractures (P<0.0001) and 1.4-1.5 fold increase in osteoporosis/osteomalacia (P<0.0001) and vitamin-D deficiency (P<0.0001) in post-pancreatectomy patients. Vitamin-D supplements were more common in the pancreatectomy group (OR 1.4, 95% CI 1.28-1.53, P<0.0001), as were specific osteoporosis/osteomalacia treatments such as calcitonin, denosumab, romosozumab, abaloparatide, and teriparatide (OR 2.24, 95%CI 1.69-2.95, P<0.0001). Retrospective analysis of CT imaging revealed that BMD declined more rapidly following pancreatic resection compared to normal historical controls (P=0.015). Older age, pancreatic cancer, and pancreaticoduodenectomy were associated with increased rates of BMD loss (P<0.05, each). After pancreatic resection, patients are at higher risk for BMD loss and subsequent fractures. As the cohort of pancreatic resection survivorship grows, attention will need to be paid to focused prevention efforts to reduce BMD loss, osteoporosis, and fractures in these vulnerable patients, with specific attention to the pancreatic cancer population.

Serologic responses to the MVA-based JYNNEOS mpox vaccine in a cohort of participants from the District of Columbia (D.C.)

We assessed early antibody responses after two doses of JYNNEOS (IMVANEX) mpox vaccine in the District of Columbia (D.C.) in persons at high risk for mpox without characteristic lesions or rash. Participants with PCR mpox negative specimens (oral swab, blood, and/or rectal swab) on the day of receipt of the first vaccine dose and who provided a baseline (day 0) serum sample and at least one serum sample at ∼28, ∼42–56 days, or 180 days post vaccination were included in this analysis. Orthopoxvirus (OPXV)-specific IgG and IgM ELISAs and neutralizing antibody titers were performed, and longitudinal serologic responses were examined. Based on participants’ IgG and IgM antibody levels at baseline, they were categorized as naïve or non-naïve. Linear mixed effects regression models were conducted to determine if IgG antibody response over time varied by age, sex, HIV status, and route of administration for both naïve and non-naïve participants. Among both naïve and non-naïve participants IgG seropositivity rates increased until day 42–56, with 89.4 % of naïve and 92.1 % of non-naïve participants having detectable IgG antibodies. The proportion of naive participants with detectable IgG antibodies declined by day 180 (67.7 %) but remained high among non-naïve participants (94.4 %). Neutralizing antibody titers displayed a similar pattern, increasing initially post vaccination but declining by day 180 among naïve participants. There were no significant serologic response differences by age, sex, or HIV status. Serologic response did vary by route of vaccine administration, with those receiving a combination of intradermal and subcutaneous doses displaying significantly higher IgG values than those receiving both doses intradermally. These analyses provide initial insights into the immunogenicity of a two-dose JYNNEOS PEP regimen in individuals at high risk of mpox exposure in the United States.

Primary Care Organizational Support for Nurse Practitioner Practice and Emotional Health Care Delivery.

Nurse practitioners (NPs) are key to delivery of primary care services. However, poor organizational support for independent NP practice, such as lack of access to clinic resources, may lead to prioritizing patient physical health over emotional health. We investigated the relationship between organizational support for independent NP practice and emotional health care delivery. This was a secondary analysis of cross-sectional survey data collected from 397 NPs in 2017. We measured organizational support for independent NP practice using the independent practice and support subscale of the NP Primary Care Organizational Climate Questionnaire. Emotional health care delivery was measured by asking NPs how frequently they addressed emotional concerns of patients. We utilized multilevel mixed effects linear regression models, adjusting for NP and practice covariates. Controlling for NP age, gender, marital status, race, and ethnicity, along with practice setting and size, as the independent practice and support score increased, NPs reported addressing emotional concerns of patients more frequently (beta = 0.34, 95% confidence interval = 0.02-0.66, P = 0.04). This indicates that as organizations provided more support for independent NP practice, NPs were able to more frequently address emotional concerns of patients. Organizational support for independent NP practice is associated with addressing emotional concerns of patients. To support NP practice, primary care organizations should ensure that NPs manage patients independently and have access to ancillary staff and support for care management.

Metabolic effects of switching to Biktarvy (B/F/TAF) in patients with HIV-1 treated with antiretroviral regimens that do not include tenofovir disoproxil fumarate (TDF) or tenofovir alafenamide (TAF): The Metabic study.

Studies on switching to tenofovir alafenamide (TAF)-based regimens raise concerns about a worse metabolic profile in people with HIV, even though most received tenofovir disoproxil fumarate (TDF) in their previous regimen. This study aims to evaluate changes in lipid fractions, glucose, and serum markers for hepatic steatosis (HS) after switching from a TDF- or TAF-sparing regimen to bictegravir/emtricitabine/TAF (B/F/TAF). We performed a retrospective cohort study of people with HIV who switched to B/F/TAF from TDF- or TAF-sparing regimens between January 2019 and May 2022 with at least 6 months of follow-up. The primary endpoint was the absolute change in lipid fractions at 6 months. Secondary outcomes were changes in lipid fractions at 12 months and changes in other metabolic parameters (glucose, creatinine, and HS based on the triglyceride-to-glucose [TyG] ratio at 6 and 12 months). Changes were analysed using mixed linear regression models with random intercept and time as a fixed effect. The study included 259 people with HIV (median age 55 [interquartile range (IQR) 47-60] years; 80% male; 88% Caucasian; CD4+ T-cell count 675 [IQR 450-880] cells/mm3; 84.3% HIV-RNA <50 copies/mL). In total, 63 patients (30%) had hypertension, 93 (44%) dyslipidaemia, 30 (14%) diabetes, and 45% obesity/overweight. Most (60%) switched from integrase inhibitor-based regimens, and 21% switched from a boosted regimen. At 6 months, significant reductions were observed in total cholesterol (-7.64 mg/dL [95% confidence interval (CI) -13.52 to -1.76; p = 0.002]), triglycerides (-23.4 [95% CI -42.07 to -4.65]; p = 0.003), and TyG ratio (-0.14 [95% CI -0.23 to -0.05]; p < 0.001). In our real-life cohort, the effect of switching TDF-/TAF-sparing regimens to triple therapy with B/F/TAF improved total cholesterol, triglycerides, and serum markers of HS at 6 months and was neutral for the remaining metabolic parameters at 12 months.

Psychological distress over 12 months post-diagnosis in an early inflammatory arthritis cohort.

People with inflammatory arthritis (IA) experience worsened mental wellbeing alongside disease progression. Using the National Early Inflammatory Arthritis Audit (NEIAA), we assessed trends in psychological distress during 12-months following IA diagnosis, mapping these against clinical outcomes to identify associations. This is a prospective study of people recruited to NEIAA receiving an IA diagnosis and completing the baseline patient survey. Patient reported outcomes (PROs) at baseline, 3-months and 12-months were collected, including psychological distress (assessed using Patient Health Questionnaire Anxiety and Depression Screener (PHQ4ADS)). Mixed effects linear regression models estimated associations between predictor variables with psychological distress at baseline and over time. Of 6,873 eligible patients, 3,451 (50.2%) showed psychological distress at baseline. Of those completing follow-ups, 30.0% and 24.1% were distressed at 3-months and 12-months, respectively. Higher psychological distress at diagnosis was more commonly reported by younger, female, and non-white patients. Clinical factors, including higher counts of comorbidities, prior depression, and higher disease activity at diagnosis were associated with higher distress. Higher distress at baseline was associated with poorer outcomes over time in quality of life, disability, work performance, disease activity, as well as reduced likelihood of achieving good treatment response by EULAR criteria. Half of patients with IA show significant mental health comorbidity at presentation, which associated with worse disease outcomes and quality of life. Screening for anxiety and depression should be a universal standard, and access to effective mood therapies alongside arthritis treatments is essential. Strategies should be culturally valid and consider multi-morbidities.

Associations Between Midlife Anticholinergic Medication Use and Subsequent Cognitive Decline: A British Birth Cohort Study.

Anticholinergic medication use is associated with cognitive decline and incident dementia. Our study, a prospective birthcohortanalysis, aimed to determine if repeated exposure to anticholinergic medicationswas associated with greater decline, and whether decline was reversed with medication reduction. From the Medical Research Council (MRC) National Survey of Health and Development, a British birth cohort with all participants born in a single week of March 1946, we quantified anticholinergic exposure between ages 53 and 69 years using the Anticholinergic Cognitive Burden Scale (ACBS). We used multinomial regression to estimate associations with global cognition, quantified bythe Addenbrooke's Cognitive Examination, 3rd Edition (ACE-III). Longitudinal associations between ACBS and cognitive test results (Verbal memory quantifiedby theWord Learning Test [WLT], and processing speed quantified by theTimed Letter Search Task [TLST]) at three time points (age 53, 60-64 and 69) were assessed using mixed and fixed effects linear regression models. Analyses were adjusted for sex, childhood cognition, education, chronic disease count and severity, and mental health symptoms. Anticholinergic exposure was associated cross-sectionally with lower ACE-III scores at age 69, with the greatest effects in those with high exposure at ages 60-64 (mean difference - 2.34, 95% confidence interval [CI] - 3.51 to - 1.17). Longitudinally, both mild-moderate and high ACBS scores were linked to lower WLT scores, again with high exposure showing larger effects (mean difference with contemporaneous exposure - 0.90, 95% CI - 1.63 to - 0.17; mean difference with lagged exposure - 1.53, 95% CI - 2.43 to - 0.64). Associations remained in fixed effects models (mean difference with contemporaneous exposure -1.78, 95% CI -2.85 to - 0.71; mean difference with lagged exposure - 2.23, 95% CI - 3.33 to - 1.13). Associations with TLST were noted only in isolated contemporaneous exposure (mean difference - 13.14, 95% CI - 19.04 to - 7.23; p<0.01). Anticholinergic exposure throughout mid and later life was associated with lower cognitive function. Reduced processing speed was associated only with contemporaneous anticholinergic medication use, and not historical use. Associations with lower verbal recall were evident with both historical and contemporaneous use of anticholinergic medication, and associations with historical use persisted in individuals even when their anticholinergic medication use decreased over the course of the study.

Occurrence and characteristics of suicidal ideation in psychiatrically healthy individuals based on ecological momentary assessment.

Decedents with no known mental disorder comprise 5-40% of suicides, suggesting that suicide ideation (SI) and behavior may occur in the psychiatrically healthy with important implications for suicide risk screening. Healthy Volunteers (HV) and patients with Major Depressive Disorder (MDD) provided 7 days of Ecological Momentary Assessment (EMA) data about SI and stressors. Longitudinal mixed effects logistic regression models compared HV and patient SI and stressors. Mixed effects linear regression models compared HVs' and patients' SI score change from the previous epoch's SI score when each stressor occurred. HVs (n = 42) reported less frequent (p < 0.001) and less intense SI (p < 0.003) than patients (n = 80), yet did endorse SI and/or SI-related items in 44% of EMA epochs, endorsing SI items in 25% of epochs with non-zero SI scores. For 7 of 8 stressors, patients reported stressors more often than HVs (all p < 0.001) responding to them with increased SI (0.0001 < p < 0.0472). HVs were relatively resilient to stressors, reporting SI increases only in response to neglect (p < 0.0147). Although SI and SAs are documented among psychiatrically healthy individuals, scientific attention to these observations has been scant. Real-time SI measurement showed that HVs' SI was less pronounced than MDD patients', but was endorsed, nonetheless. Patients were more likely to report stressors than HVs, perhaps due to greater sensitivity to the environment, and reported SI in response to stressors, which was less common in HVs. Both MDD patients and HVs most often manifested passive SI (viz, "decreased wish to live"). However, passive SI (viz, "desire for death"), may predict suicide, even absent SI per se (thinking about killing yourself). This study validates the utility of real-time SI assessment, showing that HVs endorse SI items in 11% of epochs, which implies that suicide risk screening focused on those with mental disorders may be too narrow an approach.

Activated clotting time value as an independent predictor of postoperative bleeding and transfusion.

Activated Clotting Time (ACT) is commonly used to monitor anticoagulation during cardiac surgeries. Final-ACT values may be essential to predict postoperative bleeding and transfusions, although ideal values remain unknown. Our aim was to evaluate the utility of ACT as a predictor of postoperative bleeding and transfusion use. Retrospective study (722 patients) submitted to surgery between July 2018-October 2021. We compared patients with Final-ACT<basal-ACT and Final-ACT≥basal-ACT; and Final-ACT < 140s with ≥140 s. Continuous variables were analyzed with Wilcoxon Rank-Sum test; categorical variables using Chi-square or Fisher's exact test. A Linear Mixed Regression model was used to analyze bleeding in patients with Final-ACT < 140s and ≥140 s.Independent variables were analyzed with Binary Logistic Regression models to investigate their association with bleeding and transfusion. Patients with Final-ACT ≥ 140s presented higher postoperative bleeding than Final-ACT < 140s at 12 h (P = 0.006) and 24 h (**P = 0.004). CPB time (OR 1.009,1.002-1.015,95% CI) and masculine sex (OR 2.842,1.721-4.821,95% CI) were significant predictors of bleeding. Patients with Final-ACT ≥ 140s had higher risk of UT (OR 1.81, 1.13-2.89, 95% CI; P = 0.0104), compared to Final-ACT < 140s. CPB time (OR 1.019,1.012-1.026,95% CI) and Final-ACT (OR 1.021,1.010-1.032,95% CI) were significant predictors of transfusion. Female sex was a predictor of UT, with a probability for use of 27.23% (21.84-33.39%,95% CI) in elective surgeries, and 60.38% (37.65-79.36%,95% CI) in urgent surgeries, higher than in males. Final-ACT has a good predictive value for the use of transfusion. Final-ACT ≥ 140s correlate with higher risk of transfusion and increased bleeding. The risk of bleeding and transfusion is higher with longer periods of CPB. Males have a higher risk of bleeding, but females have a higher risk of transfusion.

Approximate inferences for Bayesian hierarchical generalised linear regression models

SummaryGeneralised linear mixed regression models are fundamental in statistics. Modelling random effects that are shared by individuals allows for correlation among those individuals. There are many methods and statistical packages available for analysing data using these models. Most require some form of numerical or analytic approximation because the likelihood function generally involves intractable integrals over the latents. The Bayesian approach avoids this issue by iteratively sampling the full conditional distributions for various blocks of parameters and latent random effects. Depending on the choice of the prior, some full conditionals are recognisable while others are not. In this paper we develop a novel normal approximation for the random effects full conditional, establish its asymptotic correctness and evaluate how well it performs. We make the case for hierarchical binomial and Poisson regression models with canonical link functions, for hierarchical gamma regression models with log link and for other cases. We also develop what we term a sufficient reduction (SR) approach to the Markov Chain Monte Carlo algorithm that allows for making inferences about all model parameters by replacing the full conditional for the latent variables with a considerably reduced dimensional function of the latents. We expect that this approximation could be quite useful in situations where there are a very large number of latent effects, which may be occurring in an increasingly ‘Big Data’ world. In the sequel, we compare our methods with INLA, which is a particularly popular method and which has been shown to be excellent in terms of speed and accuracy across a variety of settings. Our methods appear to be comparable to theirs in terms of accuracy, while INLA was faster, for the settings we considered. In addition, we note that our methods and those of others that involve Gibbs sampling trivially handle parameters that are functions of multiple parameters, while INLA approximations do not. Our primary illustration is for a three‐level hierarchical binomial regression model for data on health outcomes for patients who are clustered within physicians who are clustered within particular hospitals or hospital systems.

Cross-sectional and longitudinal associations of Iron biomarkers and cardiovascular risk factors in pre- and postmenopausal women: leveraging repeated measurements to address natural variability

BackgroundThe association between iron biomarkers and cardiovascular disease risk factors (CVD-RFs) remains unclear. We aimed to (1) evaluate the cross-sectional and longitudinal associations between iron biomarkers (serum ferritin, transferrin saturation (TSAT), transferrin) and CVD-RFs among women, and (2) explore if these associations were modified by menopausal status.MethodCross-sectional and longitudinal analyses including 2542 and 1482 women from CoLaus cohort, respectively. Multiple linear regression and multilevel mixed models were used to analyse the associations between Iron biomarkers and CVD-RFs. Variability of outcomes and iron markers between surveys was accessed using intraclass correlation (ICC).ResultsAfter multivariable adjustment, elevated serum ferritin levels were associated with increased insulin and glucose levels, while higher transferrin levels were linked to elevated glucose, insulin and total cholesterol, and systolic and diastolic blood pressure (p < 0.05). No association was observed between CVD-RFs and TSAT (p > 0.05). Iron biomarkers demonstrated low reliability across reproductive stages but exhibited stronger associations in the perimenopausal group. In longitudinal analysis, we found association only for transferrin with lower glucose levels [β = − 0.59, 95% CI (− 1.10, − 0.08), p = 0.02] and lower diastolic blood pressure [β = − 7.81, 95% CI (− 15.9, − 0.56), p = 0.04].ConclusionIn cross-sectional analysis, transferrin was associated with several CVD-RFs, and the associations did not change according to menopausal status. Conversely, in the longitudinal analyses, changes in transferrin were associated only with lower glucose and diastolic blood pressure levels. These differences might stem from the substantial longitudinal variation of iron biomarkers, underscoring the need for multiple iron measurements in longitudinal analyses.

82 Relationships between performance traits and osteochondrosis of the hock in Standardbred racehorses

Abstract Equine osteochondrosis (OC) is a developmental orthopedic disease that results in abnormal articular cartilage and often the formation of osteochondral bone chips (osteochondrosis dissecans). Standardbreds, a breed predominately used in harness racing, are prone to the development of OC. This study aimed to quantify the relationship between OC and key performance traits, fastest mile time, annual earnings, and consistency in race times over career, using a cohort of 347 North American Standardbred horses. Radiographic examination of the tarsi was previously performed to identify horses as OC-affected (n = 155) or healthy (n = 192). Performance data was acquired from the United States Trotting Association (USTA) detailing horse identification, performance, and lifetime race record. Linear mixed multiple regression models were constructed to evaluate the relationship between OC and performance. A total of 24 performance related variables (e.g., age, gait, track) were evaluated for inclusion in the model through subset variable selection and goodness of fit as determined by Schwarz’s Bayesian Information Criterion (SBC). Preliminary results indicate that healthy horses exhibit significantly faster mile times (P &amp;lt; 0.001) compared with their OC-affected counterparts. However, no significant association was detected between OC status and annual earnings (P &amp;gt; 0.2) or consistency in race times over career (P &amp;gt; 0.2). Ongoing analyses are currently being conducted to assess additional predictors and performance variables for a potential association with osteochondrosis. This study emphasizes the importance of implementing effective strategies for OC management and improving breeding selection practices when producing offspring with desired performance characteristics.

Exercise does not cause post-exertional malaise in Veterans with Gulf War Illness: A randomized, controlled, dose–response, crossover study

Chronic multisymptom illnesses (CMI) such as Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, long-COVID, and Gulf War Illness (GWI) are associated with an elevated risk of post-exertional malaise (PEM), an acute exacerbation of symptoms and other related outcomes following exercise. These individuals may benefit from personalized exercise prescriptions which prioritize risk minimization, necessitating a better understanding of dose–response effects of exercise intensity on PEM. MethodsVeterans with GWI (n = 40) completed a randomized controlled crossover experiment comparing 20 min of seated rest to light-, moderate-, and vigorous-intensity cycling conditions over four separate study visits. Symptoms, pain sensitivity, cognitive performance, inflammatory markers (C-reactive protein and plasma cytokines) were measured before and within 1 h after exercise and seated rest. Physical activity behavior was measured ≥ 7 days following each study visit via actigraphy. Linear mixed effects regression models tested the central hypothesis that higher intensity exercise would elicit greater exacerbation of negative outcomes, as indicated by a significant condition-by-time interaction for symptom, pain sensitivity, cognitive performance, and inflammatory marker models and a significant main effect of condition for physical activity models. ResultsSignificant condition-by-time interactions were not observed for primary or secondary measures of symptoms, pain sensitivity, cognitive performance, and a majority of inflammatory markers. Similarly, a significant effect of condition was not observed for primary or secondary measures of physical activity. ConclusionsUndesirable effects such as symptom exacerbation were observed for some participants, but the group-level risk of PEM following light-, moderate-, or vigorous-intensity exercise was no greater than seated rest. These findings challenge several prior views about PEM and lend support to a broader body of literature showing that the benefits of exercise outweigh the risks.

The brain insulin receptor gene network and associations with frailty index.

To investigate longitudinal associations between variations in the co-expression-based brain insulin receptor polygenic risk score and frailty, as well as change in frailty across follow-up. This longitudinal study included 1605 participants from the Helsinki Birth Cohort Study. Biologically informed expression-based polygenic risk scores for the insulin receptor gene network, which measure genetic variation in the function of the insulin receptor, were calculated for the hippocampal (hePRS-IR) and the mesocorticolimbic (mePRS-IR) regions. Frailty was assessed in at baseline in 2001-2004, 2011-2013 and 2017-2018 by applying a deficit accumulation-based frailty index. Analyses were carried out by applying linear mixed models and logistical regression models adjusted for adult socioeconomic status, birthweight, smoking and their interactions with age. The FI levels of women were 1.19%-points (95% CI 0.12-2.26, P =0.029) higher than in men. Both categorical and continuous hePRS-IR in women were associated with higher FI levels than in men at baseline (P < 0.05). In women with high hePRS-IR, the rate of change was steeper with increasing age compared to those with low or moderate hePRS-IR (P < 0.05). No associations were detected between mePRS-IR and frailty at baseline, nor between mePRS-IR and the increase in mean FI levels per year in either sex (P > 0.43). Higher variation in the function of the insulin receptor gene network in the hippocampus is associated with increasing frailty in women. This could potentially offer novel targets for future drug development aimed at frailty and ageing.