Time-dependent Covariates Research Articles

A robust approach for regression analysis of panel count data with time-varying covariates

A robust approach for regression analysis of panel count data with time-varying covariates

Blood microRNA testing in participants with suspicious low-dose CT findings: follow-up of the BioMILD lung cancer screening trial

The proper management of suspicious radiologic findings is crucial to optimize the effectiveness of low-dose computed tomography (LDCT) lung cancer screening trials. In the BioMILD study, we evaluated the utility of combining a plasma 24-microRNA signature classifier (MSC) and LDCT to define the individual risk and personalize screening strategies. Here we aim to assess the utility of repeated MSC testing during annual screening rounds in 1024 participants with suspicious LDCT findings. The primary outcome was two-year lung cancer incidence in relation to MSC test results, reported as relative risk (RR) with 95% confidence interval (CI). Lung cancer incidence and mortality were estimated using extended Cox models for time-dependent covariates, yielding the respective hazard ratios (HR). Clinicaltrials.gov ID: NCT02247453. With a median follow-up of 8.5 years, the full study set included 1403 indeterminate LDCT (CTind) and 584 positive LDCT (CT+) results. A lung cancer RR increase in MSC+compared to MSC- participants was observed in both the CTind (RR: 2.5; 95% CI: 1.4-4.32) and CT+ (RR: 2.6; 95% CI: 1.81-3.74) groups and was maintained when considering stage I or resectable tumors only. A 98% negative predictive value in CTind/MSC- and a 30% positive predictive value in CT+/MSC+ lesions were recorded. At seven years' follow-up, MSC+ participants had a cumulative HR of 4.4 (95% CI: 3.0-6.4) for lung cancer incidence and of 8.1 (95% CI: 2.7-24.5) for lung cancer mortality. Our study shows that MSC can be reliably performed during LDCT screening rounds to increase the accuracy of lung cancer risk and mortality prediction and supports its clinical utility in the management of LDCT findings of uncertain malignancy. Italian Association for Cancer Research; Italian Ministry of Health; Horizon2020; National Cancer Institute (NCI); Gensignia LifeScience.

Prognosis and Risks for Probable Chronic Lung Allograft Dysfunction: A Prospective Multicenter Study.

Chronic lung allograft dysfunction (CLAD) hinders lung transplant success. A 2019 consensus refined CLAD diagnosis, introducing probable or definite CLAD based on persistence of lung function decline. Outcomes and risks for probable CLAD remain uncertain. Determine the prognosis and clinical risks for probable CLAD in a prospective multicenter cohort. Clinical Trials in Organ Transplantation-20 included 745 CLAD-eligible adult lung recipients at 5 centers and applied rigorous methods to prospectively adjudicate probable CLAD. The impact of probable CLAD on graft loss was determined using a Cox model that considered CLAD as a time-dependent covariate. Regularized Cox modeling with LASSO penalty was used to evaluate donor or recipient characteristics and the occurrence and timing of posttransplant events as probable CLAD risks. Similar analyses were performed for definite CLAD. Probable CLAD occurred in 29.7% of patients at 3 years posttransplant and conferred a marked increase in risk for graft loss (unadjusted HR 4.38, p<0.001). Most patients (80%) with probable CLAD progressed to definite CLAD. Cytomegalovirus infection and specifically late presence (>90 days posttransplant) of donor-specific alloantibodies, acute rejection, acute lung injury, or organizing pneumonia contributed the greatest independent information about probable CLAD risk. Definite CLAD risks were similar. Probable CLAD identifies patients at high risk for graft loss, supporting prospective identification of this condition for early initiation of CLAD-directed interventions. More effective strategies to prevent posttransplant cytomegalovirus, inhibit allospecific immunity, and reduce tissue injury are needed to reduce probable CLAD and improve lung recipient survival.

Sex differences in left ventricular dilatation in patients with significant aortic regurgitation: prognostic implications

Abstract Background Left ventricular (LV) dilatation is a strong indicator of adverse outcome in patients with aortic regurgitation (AR). Accordingly, current guidelines recommend the use of LV end-systolic diameter index (LVESDi) as one of the criteria to trigger aortic valve surgery (AVS) in patients with severe AR, but with the same threshold regardless of gender. Purpose To assess sex differences in LV remodeling using both volumetric and linear measurements in a large cohort of patients with significant AR (moderate or greater) and to investigate their prognostic implications. Methods A total of 1070 patients (56 ± 18 years, 691 men) were included. The primary outcome was all-cause mortality. Results Women presented with older age (58 ± 19 vs. 55 ± 17, p=0.023) and more advanced heart failure (HF) symptoms than men (NYHA class III-IV 12% vs. 9%, p=0.017). Men showed significantly larger LVESDi (21 ± 5 vs. 20 ± 5 mm/m2, p=0.013) and LV end-systolic volume index (LVESVi 37 ± 28 vs. 26 ± 17 ml/m2 p&amp;lt;0.001). During a median follow-up of 89 (IQR, 54-132) months, 168 patients died. Women had lower survival rates at 3, 5, and 10 years of follow-up compared to men (92% vs. 94.3%; 85% vs. 89.7% and 75.3% vs. 84.1%, p=0.005) (Figure 1). Spline curve analysis revealed that the threshold of LVESDi associated with an increased risk of mortality was 20 mm/m2 for both sexes. In turn, the threshold for LVESVi was 40 ml/m2 for women and 45 ml/m2 for men. Univariable Cox regression models were constructed separately for men and women. The variables associated with all-cause mortality were age, NYHA class III-IV, and AVS as time-dependent covariate in women, and age, diabetes, AVS as time-dependent covariate, and LAVI (left atrial volume index) in men. These parameters were used as a basal multivariable Cox regression model on top of which LVESDi &amp;gt;20 mm/m2 (Model 1), LVESVi &amp;gt;40 ml/m2 for women and 45 ml/m2 for men (Model 2) were added (Table 1A). LVESVi &amp;gt;40 ml/m2 was not a significant univariate predictor in men. In women, LVESDi &amp;gt;20 mm/m2 (HR: 2.046, 95%CI 1.244-3.419; p=0.006) and LVESVi &amp;gt;40 ml/m2 (HR 1.758, CI 1.095-2.825; p=0.020) showed an independent association with all-cause mortality among other factors (Table 1A). In men, death was independently associated with LVESDi &amp;gt;20 mm/m2 (HR 1.979, 95%CI 1.255-3.121; p=0.003) and LVESVi &amp;gt;45 ml/m2 (HR 2.388, 1.522-3.746; p&amp;lt;0.001), among other factors (Table 1A). Moreover, when added to a basal model that included clinical and echocardiographic variables associated with prognosis, LV dilatation based on volumetric LV enlargement resulted in a greater discriminatory power for both genders (Table 1B). Conclusion Men and women have a similar LVESDi threshold of 20 mm/m2, which is associated with an increased risk of death. However, their respective cut-off values for LVESVi differ, with the optimal threshold for women being 40 ml/m2 and for men 45 ml/m2.

Time-varying cox model for heart failure prediction in phospholamban p.(Arg14del)-positive individuals

Abstract Background Phospholamban (PLN) p.(Arg14del)-positive individuals are at risk for developing severe heart failure (HF), yet a comprehensive risk model for this outcome is missing. Our PLN registry contains longitudinal data which enables us to capture dynamic changes in covariables over time, while prior prediction models rely on baseline clinical information which may change over time. Using this longitudinal clinical information will enhance the accuracy of HF prediction and may aid patients selection for future (gene)therapy. Purpose The aim of this study is to develop a dynamic HF risk prediction model for PLN p.(Arg14del)-positive individuals. Method Data were collected of 594 PLN p.(Arg14del)-positive individuals who had no documented history of HF or myocardial infarction at time of the first echocardiography. We incorporated two time-dependent covariates in our time-varying Cox regression model: longitudinal individual predicted left ventricular ejection fraction (LVEF) and longitudinal individual predicted relative risk for major ventricular arrhythmia (VA). Predicted LVEF values were derived using observed LVEF values in a linear mixed-effect model, while the relative risk for major VA was predicted using a Cox regression model with major VA as the outcome. These time-dependent covariates were included in the time-varying Cox regression along with baseline covariables age at inclusion and previous major VA. The results were validated using 5-fold cross-validation. Results Over a median follow-up period of 3.6 years (interquartile range 1.4-6.8), 77 (13%) individuals developed HF, defined as a composite endpoint of HF hospitalisation, implantation of left ventricular assist device, heart transplantation and HF-related mortality. Our time-varying Cox regression model demonstrated a robust 5-year mean C-statistic of 0.904 (95% CI 0.901-0.908). The hazard ratio for time-dependent LVEF predictions was 0.89 per % LVEF increase (95% CI 0.87-0.91, p &amp;lt;0.001) and for time-dependent relative major VA risk predictions 3.59 (95% CI 1.49-8.69, p = 0.005). Regarding baseline covariates, age at inclusion had a hazard ratio of 1.01 (95% CI 0.99-1.03, p = 0.47) and previous major VA 0.48 (95% CI 0.21-1.1, p = 0.08). Combining the hazard ratios of time-dependent predicted relative risk for major VA and major VA in history results in a hazard ratio of 1.73. Conclusion Our study introduces a novel time-varying Cox regression model for the individual prediction of heart failure events in PLN p.(Arg14del)-positive individuals which may aid patient selection for future (gene)therapy. Results demonstrate robust predictive performance (C-statistic of 0.904, 95% CI 0.901-0.908), highlighting the significance of incorporating longitudinal data for enhanced prediction accuracy.

Aortic valve replacement and mortality with left ventricular hypertrophy in setting of normal LVEF

Abstract Background Without symptoms, indications for aortic valve replacement (AVR) in severe aortic stenosis (AS) differ for individuals with normal vs. low/decreasing left ventricular ejection fraction (LVEF). Conceptually, the development of left ventricular hypertrophy (LVH) could also indicate ventricular injury and potential benefit with earlier AVR. However, little is known about the comparative effectiveness of AVR in individuals with LVH. As such, unlike low or dropping LVEF, development of LVH is not in itself an indication for AVR in clinical guidelines. Methods Initially, 1,232,492 echocardiography reports from 12 hospitals were identified from 1 January 2000 to 22 November 2022, and previously validated natural language processing modules were then applied to identify aortic gradients, LVEF, and presence of LVH. These reports were linked to mortality data, key comorbidities with a 2-year lookback covariate window, and date of any AVR. We then isolated echocardiograms demonstrating normal flow severe AS, LVEF ≥ 55% and LVH. Chart reviews were conducted to reduce missing data and outcomes were censored on the date of the chart review (October 4, 2023). To assess the association between AVR and mortality, Cox proportional hazards models considering treatment status (ie, AVR or not) were used with the index echocardiogram as time zero. Covariates used in risk-adjustment were chosen a priori based on conditions and laboratory values used in cardiac surgery mortality risk-adjustment, since those variables could plausibly confound the treatment decision. To address immortal time bias, AVR was considered a time-varying covariate. Multiple additional approaches with propensity scores were performed as sensitivity analyses. Results After application of inclusion and exclusion criteria, 4856 unique patients remained for analysis. Of those, 2043 (42.1%) received AVR with 841/2043 (41.2%) receiving surgical AVR and 1202/2043 (58.8%) receiving transcatheter AVR. Patients who received AVR had lower unadjusted mortality than those who did not receive AVR (33.6% vs. 47.1%, p &amp;lt; 0.001). After adjustment, and accounting for time-dependence of AVR, AVR was associated with reduced mortality (HR 0.81, 95% CI 0.73-0.89, p &amp;lt; 0.0001). The sensitivity analyses were all consistent with the primary analysis. Conclusions AVR was associated with reduced mortality for patients with severe AS, preserved LVEF, and LVH. This association persisted even after accounting for time-dependence, which tends to reduce the observed effectiveness of procedures in observational data. With over a million echocardiograms over more than 20 years, this is the largest known analysis of AVR outcomes in individuals with severe AS and LVH. Our findings suggest LVH may be a high-risk phenotype of individuals with AS, similar to dropping LVEF. As such, these results raise the potential of more proactive AVR clinical guidelines in individuals with LVH, even when LVEF is normal.Survival in AVR Vs No AVR

The association between heart failure diagnosis and the incidence of cancer: a longitudinal analysis of the United Kingdom Biobank

Abstract Background Heart failure (HF) and cancer represent prevalent chronic conditions with a high burden of disease worldwide. Several studies have shown a high frequency of coexistence of these two entities, highlighting a potential bidirectional association. However, few longitudinal studies have explored whether patients with HF have a significantly increased risk of cancer and if comorbidity profiles affect this risk. Purpose To explore the association between incident HF diagnosis and cancer development in UK Biobank participants after adjustment by potential confounders. Methods We conducted a retrospective cohort study in UK Biobank. HF and cancer diagnoses were determined using diagnoses and patient reports. We evaluated breast, liver, lung, prostate, colorectal, uterus, and hematological cancer. Individuals with a history of HF or cancer diagnosed before the baseline assessment were excluded. The primary exposure was incident HF diagnosis after enrollment, and the primary outcome was incident cancer during follow-up. Multivariate Cox proportional hazard regression models using time-varying covariates were used to calculate hazard ratios (HR) for the explored associations with overall cancer and by major cancer type. A Bonferroni-corrected p-value &amp;lt; 0.003 was considered statistically significant to control for family-wise error rate. Results 476,652 patients were analyzed (mean age: 56.3 years; 57.8% females). Over a median follow-up of 13.7 years, 12,461 (2.6%) developed HF, and 34,452 (7.2%) were diagnosed with cancer. Incident HF was significantly associated with increased overall cancer risk (HR 1.34, 95% CI 1.22–1.48. p&amp;lt;0.001), as well as lung (HR 1.30, 95% CI 1.18–1.43), and hematological cancers (HR 1.54, 95% CI 1.25–1.91. p&amp;lt;0.001) (Figure). Conclusion Despite sharing multiple comorbidities and risk factors, incident HF remained significantly associated with a higher cancer risk even after multivariable adjustment by relevant confounders. Nevertheless, confounding bias and especially detection bias, cannot be ruled out in the present analysis. Further research is necessary to validate these findings and elucidate the mechanisms underlying this association.Figure

Racial and ethnic disparities in the incidence of stroke and mortality in atrial fibrillation: a US nationwide study

Abstract Background Atrial fibrillation (AF) is the most common arrhythmia globally and is associated with increased risks of cardiovascular mortality and ischemic stroke. Prior research has demonstrated racial/ethnic disparities in AF outcomes yet this work has been limited to small cohorts and did not adjust for anticoagulant therapy. Purpose We compared incidence of ischemic stroke and death by race/ethnicity for patients with AF in the Veterans Health Administration (VA), the largest integrated, low-cost, national healthcare system in the US. Methods In this retrospective cohort study we identified those with an outpatient diagnosis of AF and a confirmatory diagnosis ≦ 180 days of their index AF diagnosis enrolled in VA from January 1, 2014 to December 31, 2021, with follow-up for outcomes from the index AF diagnosis date until May 31, 2022. To create a cohort of incident AF cases, we excluded patients with an AF diagnosis or anticoagulant therapy in the 2 years prior to their index AF diagnosis. We also excluded patients with pre-existing valvular heart disease, prior stroke, receiving hospice care or who died within 180 days of index AF diagnosis. We categorized our independent variable as race (i.e., non-Hispanic American Indian / Alaska Native [AI/AN], Asian, Black, or White [reference group]), and ethnicity (i.e., Hispanic). Our primary outcomes were incidence of stroke or death at any point within the study follow-up period. To compare these outcomes by race/ethnicity we conducted a survival analysis, adjusting for anticoagulant use as a time-varying covariate along with sociodemographic factors (age, sex, region, area deprivation index), clinical factors (CHADS2VA2Sc stroke risk and HAS-BLED bleeding risk), and facility type. Results There were 187,080 patients in the final cohort including 0.5% AI/AN, 1% Asian, 9% Black, 4% Hispanic, 85% White; mean age was 73 years. Over a mean study follow-up period of 4.43 years, 65,375 (34.9%) patients died and 24,910 (13.3%) developed stroke. The adjusted hazard ratio [95% CI] for death was lower for Hispanic (0.84 [0.78, 0.91]) and Black (0.94 [0.91, 0.97]) patients than White patients. In contrast, the adjusted hazard ratio for stroke incidence was higher for Black (1.21 [1.16, 1.27]) and Hispanic (1.10 [1.03, 1.17]) patients than White patients. There was no difference observed between AI/AN or Asian and White patients in either outcome. Conclusions In a national cohort of patients with incident AF, we identified significantly higher rates of stroke in Black and Hispanic than White patients even controlling for myriad factors including anticoagulant use. Conversely, racial/ethnic disparities in mortality appear to be eliminated in VA, counter to prior data. Understanding the factors, beyond anticoagulation, driving disparities in stroke outcomes and identifying factors reducing mortality among minority groups is critical to improving overall AF care in the largest integrated US health system.Adjusted Hazard Ratios of Study Outcomes

Long-term prognostic value of microvascular obstruction by cardiac magnetic resonance in ST-segment elevation myocardial infarction

Abstract Background Microvascular obstruction (MVO) has been associated with a higher risk of cardiac remodelling and adverse events following ST-segment elevation myocardial infarction (STEMI). However, long-term prognostic implications of MVO in STEMI beyond five years of follow-up have not been assessed to date. Purpose To investigate the relation of MVO with all-cause mortality and recurrent ischemic events at long-term follow-up after STEMI. Methods This is a pooled analysis of three observational studies including 876 prospectively enrolled reperfused first STEMI patients, who underwent cardiac magnetic resonance imaging with late gadolinium enhancement, between 2003-2019, from four university medical centers in the Netherlands and Spain. Median follow-up duration was 6.3 (IQR 3.6-9.3) years. The primary outcome was all-cause mortality. The secondary outcome was a combined endpoint of all-cause mortality and recurrent ischemic events (i.e. myocardial infarction or stroke). We performed Cox regression analyses with a time-dependent covariate. Model 1 included correction for clinical covariates only, i.e. age, sex, smoking, diabetes mellitus, hypertension, and LAD culprit artery. Model 2 included imaging characteristics, i.e. infarct size and left ventricular ejection fraction, in addition to the clinical covariates. Results The study population consisted of 876 patients, mean age 59 years ± 12, 720 men (82%). MVO was present in 499 patients (58%). The presence of MVO was significantly associated with all-cause mortality up to six years post-STEMI, independent from clinical characteristics (Hazard Ratio [HR] 2.23, 95% CI 1.09-4.57, p=0.029), but not after six years post-STEMI (HR 0.98, 95% CI 0.45-2.12, p=0.958). Presence of MVO was not significantly associated with a combined endpoint of all-cause mortality and recurrent ischemic events before or after six years of follow-up (HR 1.27, 95% CI 0.81-1.99, p=0.294 and HR 0.68, 95% CI 0.35-1.31, p=0.244, respectively). All other traditional prognosticators, except age, also lacked predictive value at long-term follow-up beyond six years. Conclusions In STEMI patients, the presence of MVO by cardiac magnetic resonance imaging is associated with a more than two-fold higher risk of all-cause mortality up to six years after the index event. This relation seems to dissipate beyond this time period. Furthermore, presence of MVO was not associated with recurrent ischemic events at long-term follow-up post-STEMI.

Left ventricular dilatation in patients with significant aortic regurgitation: association with mortality

Abstract Background In aortic regurgitation (AR), left ventricular (LV) dilatation occurs as a response to volume and pressure overload. Current guidelines recommend the use of LV end-systolic diameter index (LVESDi) to give indication for intervention, but recent studies suggested that volumetric measures may be more accurate in identifying LV remodeling in AR as compared to linear dimensions. Purpose To characterize LV dilatation in patients with significant AR (≥moderate), based on linear and volumetric dimensions, and to determine their prognostic value. Methods A total of 1070 patients (56 ± 18 years,65% male) were included. Cut-off values of 20 mm/m2 for LVESDi and 45 ml/m2 for end-systolic volume index (LVESVi) were used to identify the following LV remodeling patterns: No-significant LV dilatation (N=485), when both LVESDi and LVESVi were below the cut-off values; Discordant LV dilatation (N=279) if only one positive criterion was present; and Concordant LV dilatation (N=306) when both LVESDi and LVESVi were enlarged (Figure 1). The primary endpoint was all-cause mortality. Results Baseline characteristics differed across the 3 LV dilatation groups. Compared to patients with no-significant or discordant LV dilatation, those with concordant LV dilatation were more likely to be men, had a lower prevalence of systemic arterial hypertension and were more symptomatic. Concerning the echocardiographic characteristics, patients with concordant LV dilatation showed more severe AR, worse LV systolic function and larger left atrial volumes compared to the other two LV dilatation groups. During a median follow-up of 89 (IQR,54-132) months, 168 (16%) patients died and 484 (45%) underwent aortic valve surgery (AVS). Patients with concordant LV dilatation had worse 10-year survival (72%) compared to the groups (Figure 2), but benefited more from AVS based on a landmark analysis (p=0.001). In multivariable analysis, a significant association with the risk of all-cause mortality was observed for the presence of discordant (HR 1.645, 95%CI 1.077-2.513;p=0.021) or concordant LV dilatation (HR 2.695, 95% CI 1.710 to 4.249;p&amp;lt;0.001) after adjusting for the variables significant in the univariate analysis such as, age, male gender, NYHA class III-IV, AVS as a time-dependent covariate and LV ejection fraction (LVEF) ≤55%. Moreover, the addition of LV dilatation groups to a basal model including the above mentioned clinical and echocardiographic variables led to a significant increase in the predictivity of the model (Chi-square difference=21.1,p&amp;lt;0.001). Conclusion In patients with significant AR, LV dilatation detected by linear and/or volumetric measures was independently associated with increased mortality. Combining both methods for assessment of LV remodeling may improve risk stratification of these patients.

Pregnancy complications and long-term risk of cardiovascular events in women with heart disease

Abstract Background/Introduction Women with congenital heart or acquired heart disease are at increased risk of atherosclerotic cardiovascular disease. Pregnancy complications are established risk factors for cardiovascular disease in the general population, but their prognostic value for cardiovascular outcomes in the population with congenital or acquired heart disease is unknown. Purpose To determine the frequency and prognostic value of pregnancy complications for the risk of cardiovascular outcomes in women with congenital or acquired heart disease. Methods This nationwide registry-based cohort study included women in Sweden with heart disease (pulmonary arterial hypertension, congenital heart disease or acquired valvular heart disease) with singleton births registered in the national Medical Birth Register between 1973 and 2014. Exposures of pregnancy complications (preeclampsia/gestational hypertension, preterm birth or birth of an infant small for gestational age) were collected from the Medical Birth Register, and cardiovascular outcomes (mortality and hospitalizations) from the Cause of Death Register and the National Inpatient Register. Cox regression models were performed with time-dependent covariates, to determine the possible prognostic value of the studied pregnancy complications for cardiovascular outcomes including atherosclerotic cardiovascular disease, heart failure and arrythmias after adjustment for major confounding including smoking and body mass index. Results Among the total of 2,134,239 women included from the Medical Birth Register, 2,554 women with 5,568 singleton births were affected by heart disease. Preeclampsia/gestational hypertension affected 5.8% of pregnancies, preterm birth 9.7%, and birth of an infant small for gestational age 2.8%. Preterm birth (adjusted HR 1.91 (95% CI 1.38-2.64)) was associated with an increased risk of maternal all-cause mortality. Preeclampsia/gestational hypertension (adjusted HR 1.64 (95% CI 1.18-2.29)) and preterm birth (adjusted HR 1.56 (95% CI 1.19-2.04)) were associated with an increased risk of hospitalizations for atherosclerotic cardiovascular disease. Conclusions Pregnancy complications were frequent in women with congenital or acquired heart disease. With a median follow-up time of 22 years, preterm birth was associated with a higher risk of cardiovascular mortality, and preeclampsia/gestational hypertension and preterm birth were associated with cardiovascular morbidity. In women with heart disease, pregnancy complications may provide additional information for the risk assessment of future cardiovascular outcomes.Overview of study designPrevalence and risk estimation

Fat mass mediating effect on the association of worsening insulin resistance with increased left ventricular mass in 1595 adolescents: a 7-Year Longitudinal and Mediation Study

Abstract Background Fat mass-insulin resistance vicious cycle has been established in the young population and late adolescence has been identified as the critical time when the pathologic cycle begins. However, the independent and synergistic effect of fat mass in the relationship of insulin resistance with cardiac structure remains unexamined in a large cohort of adolescents, possibly due to the scarcity of repeated echocardiography measures. Mounting effective prevention strategies relies on strong evidence of the timing of the earliest cardiovascular alteration induced by insulin resistance. Purpose To investigate the longitudinal association of insulin resistance with left ventricular mass (LVM) and the mediating effect of fat mass in adolescents. Methods From the Avon Longitudinal Study of Parents and Children (ALSPAC), UK birth cohort, 1595 adolescents aged 17 years who had fasting glucose and insulin measured at 17- and 24-year clinic visits were included. Homeostatic model assessment for insulin resistance (HOMA-IR) was calculated from fasting insulin x fasting glucose/22.5 at ages 17 and 24 years. Repeated echocardiography measured LVM indexed for height2.7 (LVMI2.7) was available at baseline and follow-up. Fat mass was repeatedly measured with dual-energy Xray absorptiometry at both time points. Multivariable adjusted association was examined using generalized linear mixed-effect models and adjusted for sex, family history of hypertension/diabetes/high cholesterol/vascular disease, socioeconomic status, and time-varying covariates measured at both baseline and follow-up such as age, high-sensitivity C-reactive protein, heart rate, systolic blood pressure, fat mass, lean mass, smoking status, sedentary time, light physical activity, moderate-to-vigorous physical activity, low-density lipoprotein cholesterol, triglyceride, and high-density lipoprotein cholesterol. Mediation analysis was conducted with a structural equation model adjusting for same covariates, except fat mass. Results Among 1595 adolescents (mean [SD] age at baseline, 17.74 [0.38] years; 955 [59.8%] females]) median insulin resistance increased from 1.46 to 1.74, median fat mass increased from 15.6kg to 19.2kg and the mean LVMI increased from 35.4g/m2.7 to 38.1g/m2.7, over 7 years. In a fully adjusted model, increased insulin resistance from adolescence through young adulthood was significantly associated with increased LVMI (effect estimate 1.01g/m2.7 [95% CI, 0.63 – 1.57], p&amp;lt;0.001) across the 7-year observation period. Increased fat mass mediated (62.3% mediation effect) the longitudinal association between insulin resistance and LVMI from ages 17 – 24 years. Conclusion Worsening insulin resistance from adolescence through young adulthood was associated with increased cardiac mass, contributing 37% to the total increase in LVMI between ages 17 to 24 years but increased fat mass explained two-thirds of the relationship.

Reciprocal Relationships Between Coping Strategies and Well-Being in Children and Adolescents

Abstract: Background: Not much is known about reciprocal prospective within-person effects between various coping strategies or between coping strategies and well-being in minors. Aims: Therefore, we used a random intercept cross-lagged panel model (RI-CLPM) with four-time points spaced 6 months apart to analyze how five coping strategies (seeking social support, problem-solving, avoidant coping, palliative emotion regulation, anger-related emotion regulation) were associated with each other and with well-being. Method: The sample comprised 702 minors. Their age at Time 1 was 10.30 years ( SD = 1.35; range: 8–14). Results: On the between-person level, problem-solving and palliative emotion regulation were positively related to well-being, whereas anger-related emotion regulation was negatively related. Among the coping strategies themselves, anger-related emotion regulation and problem-solving were positively correlated with seeking social support, and problem-solving and seeking social support were negatively correlated with avoidant coping. At the within-person level, avoidant coping was negative, and seeking social support was positively bidirectionally associated with well-being. Additionally, well-being positively influenced problem-solving, and seeking social support was positively and avoidant coping was negatively bidirectionally associated with problem-solving. Limitations: Due to the 6-month interval, statements are limited to only this time period. Furthermore, whereas we controlled for time-invariant covariates, we did not control for time-varying covariates. Conclusion: The findings underscore the importance of simultaneously examining different coping strategies along with well-being in a within-person design. Understanding the links between coping and well-being can contribute to the development and use of interventions promoting minors’ coping and well-being.

The impact of COVID-19 status and vaccine type following the first dose on acute heart disease: A nationwide retrospective cohort study in South Korea.

Recent studies have suggested an increased incidence of myocarditis and pericarditis following mRNA vaccination or COVID-19. However, the potential interaction effect between vaccine type and COVID-19 on heart disease risk remains uncertain. Our study aimed to examine the impact of COVID-19 status and vaccine type following the first dose on acute heart disease in the Korean population, using data from the National Health Insurance Service COVID-19 database (October 2018-March 2022). We sought to provide insights for public health policies and clinical decisions pertaining to COVID-19 vaccination strategies. We analysed heart disease risk, including acute cardiac injury, acute myocarditis, acute pericarditis, cardiac arrest, and cardiac arrhythmia, in relation to vaccine type and COVID-19 within 21days after the first vaccination date, employing Cox proportional hazards models with time-varying covariates. This study included 3,350,855 participants. The results revealed higher heart disease risk in individuals receiving mRNA vaccines than other types (adjusted HR, 1.48; 95% CI, 1.35-1.62). Individuals infected by SARS-CoV-2 also exhibited significantly higher heart disease risk than those uninfected (adjusted HR, 3.56; 95% CI, 1.15-11.04). We found no significant interaction effect between vaccine type and COVID-19 status on the risk of acute heart disease. Notably, however, younger individuals who received mRNA vaccines had a higher heart disease risk compared to older individuals. These results may suggest the need to consider alternative vaccine options for the younger population. Further research is needed to understand underlying mechanisms and guide vaccination strategies effectively.

GLP-1 analog use is associated with improved disease course in inflammatory bowel disease: a report from the Epi-IIRN.

The growing use of GLP-1 analogs for type 2 diabetes mellitus (DM2) and obesity necessitates studies about their use in patients with inflammatory bowel diseases (IBD). Data on patients with DM2 were retrieved from an Israeli nationwide cohort of patients with IBD (epi-IIRN), recording GLP-1 analog exposure for at least 6 months. Primary outcome was poor disease outcomes (i.e. composite of steroid-dependence, initiation of advanced IBD therapy, hospitalization, surgery, or death). Cox proportional hazard models with time-varying covariables were used to assess the impact of GLP-1 use on outcomes during follow-up. We included 3,737 patients (24,338 patient-years) with IBD and DM2 [(50.4% ulcerative colitis (UC)], of whom 633 were treated with GLP-1 analogs. Accounting for demographics IBD/DM2 related variables, medication use, and laboratory measurements, GLP-1 analog use was associated with reduced composite outcome in the full cohort (adjusted Hazard Ratio (aHR) 0.74, 95%CI 0.62-0.89) and in each subtype [UC (aHR 0.71, 95%CI 0.52-0.96) and Crohn's disease (aHR 0.78, 95%CI 0.62-0.99)]. Similar trends were seen in multivariate analyses of each individual outcome, although only hospitalization was significant (aHR 0.74, 95%CI 0.61-0.91). The protective effect of GLP-1 analogs was seen in patients with obesity (aHR 0.61, 95%CI 0.50-0.77), but not in non-obese (aHR 0.94, 95%CI 0.67-1.31). GLP-1 analogs are associated with improved outcomes in IBD, specifically in patients with obesity. The mechanisms of these effects require further investigation as well as their role in patients without DM2.

Effect of Prednisone Dosing on Mineralocorticoid-Related Side Effects With Abiraterone in Prostate Cancer.

Abiraterone use for prostate cancer can cause mineralocorticoid excess syndrome (MES; eg, hypertension and hypokalemia). Prednisone mitigates these effects; however, the optimal dose level is unclear. This study examines MES effects from abiraterone with 5 mg of prednisone once daily versus 5 mg twice daily. Data for 1,410 abiraterone-treated patients from 2011 to 2022 were identified from a large academic/community hospital system. Three hundred and fifty-three patients were excluded for missing medication data and use of an alternative steroid; 1,057 patients remained (5 mg once daily, n = 550, 5 mg twice daily, n = 507). Prednisone dose was treated as a time-varying covariate. Hypokalemia and hypertension incidence over 24 weeks after abiraterone initiation was analyzed via Cox proportional hazard models using Common Terminology Criteria for Adverse Events (v5.0) grading via direct clinical measurements and International Classification of Diseases (ICD)-10 code outcomes. Patients receiving 5 mg of prednisone twice daily had a statistically significant decrease in cumulative hazard for experiencing at least one MES event (hypertension and/or hypokalemia) via direct clinical measurement (hazard ratio [HR], 0.79 [CI, 0.68 to 0.91]; P = .002) and by ICD-10 code (HR, 0.65 [CI, 0.54 to 0.79]; P < .001) analysis. This finding was durable with individual end point analysis of hypertension and hypokalemia. There were no changes to BMI or hyperglycemia (>140 mg/dL) between the cohorts. This retrospective analysis shows a decrease in risk for the development of at least one episode of hypertension or hypokalemia with abiraterone using 5 mg twice-daily prednisone in the study population. Assessments of metabolic impacts (BMI, hyperglycemia) did not show differences with prednisone dosing. These findings may merit consideration when determining an optimal prednisone dosing regimen.

Lung cancer survival trends and prognostic factors: A 26-year population-based study in Girona Province, Spain

BackgroundLung cancer (LC) is Europe’s primary cause of cancer-related mortality largely due to its historically low survival rates. The aim of this study was to analyze 26-year survival trends in the province of Girona, Spain, and to identify key prognostic factors. MethodsPopulation-based study of LC cases collected between 1994 and 2019, with follow-up until December 31, 2021. Variables included date of diagnosis, sex, age, histology, and tumor stage (the latter since 2010). Diagnosis dates were categorized into three periods (1994–2002, 2003–2011, and 2012–2019). Multivariate flexible parametric models, incorporating age as a non-linear, time-varying covariate, were used to analyze net survival (NS) and trends. Annual absolute change in survival (AAC_S) was calculated using 3-year NS. ResultsThe analysis of 9,113 LC cases showed a NS improvement between the first and last period (7.1 months (95 %CI: 6.5;7.6) to 8.5 months (95 %CI: 7.9;9.1)). Squamous cell carcinoma (NSC-SCC) showed the greatest improvement with an AAC_S of 0.32 % (95 % CI: 0.21; 0.43), while survival for non-small cell lung cancer not otherwise specified declined (AAC_S of −0.19 % (95 %CI: −0.26; −0.12)). Prognostic analysis of the 3,642 cases (2010–2019) indicated a lower LC death risk for adenocarcinoma and NSC-SCC compared to LC not otherwise specified (HR 0.52 and 0.62, respectively). Increasing tumor stage correlated with higher LC mortality risk (1.8-, 4.0-, and 10.1-fold increase for stage II, III, and IV, respectively, compared to stage I). ConclusionsLC survival has notably improved, particularly for NSC-SCC. Survival is influenced by sex, age, date of diagnosis, tumor histology and especially by stage, underscoring comprehensive data collection’s importance.

Is a Driver's history of drunk driving and other traffic offenses linked to subsequent drunk driving and traffic crashes? A case study of Taiwan.

Crashes involving alcohol-impaired driving or driving under influence (DUI) are more likely to increase crash probability and severity. Many countries have adopted increasingly stringent policies in curbing DUI. Nevertheless, more and more studies show that while the total number of DUIs has reduced as a whole, DUI recidivism remains challenging. As such, this study seeks to research into whether it is possible to identify frequent DUI recidivists based on their traffic offense history, so that effective countermeasure could be put in place and in time. This study proposes using the drivers' traffic offense history and length of duration between two DUIs, the duration of re-offending, to identify frequent DUI recidivists. This approach is not only widely adopted in public health, but is also flexible in accommodating many modeling issues such as data censoring, recurrent events, and the inclusion of time-varying covariates to address questions like whether the probability of recidivism increases or decreases with subsequent DUI offenses or other traffic offenses or violations. Our major results show that: (1) For all drivers caught for a DUI, 10 % of them would be caught for another DUI within a year; (2) In contrast, the same one-year recidivism probability for those who accumulated two DUIs and two run-the-red-light could be as high as 17 %; (3) Each subsequent DUI increased the probability of a further DUI offense by 57 %; and (4) Each additional DUI offense was associated with 45 % increase in probability of being involved in a crash involving DUI. Overall, there are clear links between a driver's history of traffic offenses, DUI recidivism, and crash involvement, which could provide valuable information for authorities to profile potential recidivists and apply preventative measures in advance to reduce DUI-related crashes.

A longitudinal study of the psychological predictors of mental health and stress among university students during the COVID-19 pandemic in Germany

AbstractRepeated nationwide lockdowns and contact restrictions during the COVID-19 pandemic in Germany massively affected students’ life.There is a lack of longitudinal studies on the change in students’ mental health problems and the influencing factors across the pandemic phases. The aims of the longitudinal study were to investigate the change in students’ depression, anxiety and stress across a one-and-a-half-year study period of the pandemic in Germany and the association between protective (mindfulness, coping strategies, self-efficacy) and risk factors (financial and academic worries) and students’ mental health issues and stress. A subsample of 911 university students was included in the longitudinal analyses. Students were surveyed at four points during the first and second years of the pandemic. Linear mixed-effects modelling was used to investigate our research questions. The models investigated the protective and risk factors as time-invariant and time-varying covariates. The results showed that depression, anxiety and stress decreased slightly over the pandemic. Further, mindfulness and active coping remained equally important over time for reducing depression, anxiety and stress. However, self-efficacy and acceptance became more critical for reducing mental health issues or stress over time. Although financial worries remained equally important for depression, anxiety and stress throughout the pandemic, academic worries became more important for mental health issues over time.The study provides insights into protective and risk factors throughout the pandemic. The results highlight the importance of investigating time-invariant and time-varying predictors of mental health issues.

AI Survival Prediction Modeling: The Importance of Considering Treatments and Changes in Health Status over Time

Background and objectives: Deep learning (DL)-based models for predicting the survival of patients with local stages of breast cancer only use time-fixed covariates, i.e., patient and cancer data at the time of diagnosis. These predictions are inherently error-prone because they do not consider time-varying events that occur after initial diagnosis. Our objective is to improve the predictive modeling of survival of patients with localized breast cancer to consider both time-fixed and time-varying events; thus, we take into account the progression of a patient’s health status over time. Methods: We extended four DL-based predictive survival models (DeepSurv, DeepHit, Nnet-survival, and Cox-Time) that deal with right-censored time-to-event data to consider not only a patient’s time-fixed covariates (patient and cancer data at diagnosis) but also a patient’s time-varying covariates (e.g., treatments, comorbidities, progressive age, frailty index, adverse events from treatment). We utilized, as our study data, the SEER-Medicare linked dataset from 1991 to 2016 to study a population of women diagnosed with stage I–III breast cancer (BC) enrolled in Medicare at 65 years or older as qualified by age. We delineated time-fixed variables recorded at the time of diagnosis, including age, race, marital status, breast cancer stage, tumor grade, laterality, estrogen receptor (ER), progesterone receptor (PR), and human epidermal receptor 2 (HER2) status, and comorbidity index. We analyzed six distinct prognostic categories, cancer stages I–III BC, and each stage’s ER/PR+ or ER/PR− status. At each visit, we delineated the time-varying covariates of administered treatments, induced adverse events, comorbidity index, and age. We predicted the survival of three hypothetical patients to demonstrate the model’s utility. Main Outcomes and Measures: The primary outcomes of the modeling were the measures of the model’s prediction error, as measured by the concordance index, the most commonly applied evaluation metric in survival analysis, and the integrated Brier score, a metric of the model’s discrimination and calibration. Results: The proposed extended patients’ covariates that include both time-fixed and time-varying covariates significantly improved the deep learning models’ prediction error and the discrimination and calibration of a model’s estimates. The prediction of the four DL models using time-fixed covariates in six different prognostic categories all resulted in approximately a 30% error in all six categories. When applying the proposed extension to include time-varying covariates, the accuracy of all four predictive models improved significantly, with the error decreasing to approximately 10%. The models’ predictive accuracy was independent of the differing published survival predictions from time-fixed covariates in the six prognostic categories. We demonstrate the utility of the model in three hypothetical patients with unique patient, cancer, and treatment variables. The model predicted survival based on the patient’s individual time-fixed and time-varying features, which varied considerably from Social Security age-based, and stage and race-based breast cancer survival predictions. Conclusions: The predictive modeling of the survival of patients with early-stage breast cancer using DL models has a prediction error of around 30% when considering only time-fixed covariates at the time of diagnosis and decreases to values under 10% when time-varying covariates are added as input to the models, regardless of the prognostic category of the patient groups. These models can be used to predict individual patients’ survival probabilities based on their unique repertoire of time-fixed and time-varying features. They will provide guidance for patients and their caregivers to assist in decision making.