In active surveillance of prostate cancer, cancer progression is interval-censored and the examination to detect progression is subject to misclassification, usually false negatives. Meanwhile, patients may initiate early treatment before progression detection, constituting a competing risk. We developed the Misclassification-Corrected Interval-censored Cause-specific Joint Model (MCICJM) to estimate the association between longitudinal biomarkers and cancer progression in this setting. The sensitivity of the examination is considered in the likelihood of this model via a parameter that may be set to a specific value if the sensitivity is known, or for which a prior distribution can be specified if the sensitivity is unknown. Our simulation results show that misspecification of the sensitivity parameter or ignoring it entirely impacts the model parameters, especially the parameter uncertainty and the baseline hazards. Moreover, specification of a prior distribution for the sensitivity parameter may reduce the risk of misspecification in settings where the exact sensitivity is unknown, but may cause identifiability issues. Thus, imposing restrictions on the baseline hazards is recommended. A trade-off between modelling with a sensitivity constant at the risk of misspecification and a sensitivity prior at the cost of flexibility needs to be decided.

Prostate cancer patients with biochemical recurrence (BCR) face a decision of whether to start salvage therapy (ST), which may reduce the probability of metastatic progression at the cost of side effects. To inform the decision to start ST at or after BCR, models that make counterfactual predictions incorporating treatment are highly desirable. However, estimation of such models using observational data requires care due to time-varying confounding by the longitudinal biomarker prostate-specific antigen (PSA). Moreover, a careful definition of the estimands of interest, referred to as "predictimands", is required due to the possibility of delayed initiation of treatment after biochemical recurrence. In this study, we utilize the framework of joint longitudinal and survival models to tackle these issues, estimating a model for pre-ST PSA trajectories and risk of metastasis that incorporates the effect of ST, from a dataset of 2075 patients with BCR. We define relevant predictimands for a new patient after BCR under three scenarios: Immediately treated, never treated, and treatment under a dynamic regime, where ST is started when PSA is observed to exceed a pre-specified threshold. We propose a Monte Carlo scheme for computing these predictimands, adapting previous work on dynamic predictions from joint models to account for treatment timing. This methodology is applied to an example patient and validated in a simulation study. This methodology could be adapted to a wide variety of applications requiring counterfactual predictions in the presence of time-varying treatments and biomarkers. Code to implement such analyses is available in the R package JMbayes2.

The experience sampling method (ESM) is a longitudinal research design where participants report their thoughts, emotional states and behaviours multiple times a day. Our work is motivated by such data collected by the GrowIt! app, which was released to investigate daily emotions among adolescents during the COVID-19 pandemic. Current procedures to analyse ESM data face various challenges. While standard statistical techniques may not scale well to a high-dimensional setting, machine learning procedures can give biased results due to selection bias introduced by missingness. In our motivating dataset, adolescents dropped out due to previous strong feelings of negative emotions. Hence, the implied missing data are of the missing-at-random type that standard machine learning procedures cannot accommodate. We develop a novel neural network architecture that generalises mixed effects models to deep learning to overcome these challenges. It allows semi-parametric and flexible modelling of data's mean and correlation structure through fixed and random effects. For estimation, we use an adaptation of variational auto-encoders and a Bayesian data augmentation algorithm. Through this approach, the model can accommodate longitudinal outcomes following generic distributions, scale well to high-dimensional settings and provide valid inference when data are missing-at-random. We applied the Deep Generalised Mixed Model to the GrowIt! study and various simulations. The results show potential for the Deep Generalised Mixed Model, yet suboptimal performance due to model instability.

Evaluating the performance of a prediction model is a common task in medical statistics. Standard accuracy metrics require the observation of the true outcomes. This is typically not possible in the setting with time-to-event outcomes due to censoring. Interval censoring, the presence of time-varying covariates, and competing risks present additional challenges in obtaining those accuracy metrics. In this study, we propose two methods to deal with interval censoring in a time-varying competing risk setting: a model-based approach and the inverse probability of censoring weighting (IPCW) approach, focusing on three key time-dependent metrics: area under the receiver-operating characteristic curve, Brier score, and expected predictive cross-entropy. The evaluation is conducted over a medically relevant time interval of interest, [ t , Δ t ) $[t, \Delta t)$ . The model-based approach includes all subjects in the risk set, using their predicted risks to contribute to the accuracy metrics. In contrast, the IPCW approach only considers the subset of subjects who are known to be event-free or experience the event within the interval of interest. We performed a simulation study to compare the performance of the two approaches with regard to the three metrics. Furthermore, we demonstrated the three metrics using the two approaches on an example prostate cancer surveillance cohort. Risk predictions were generated from a joint model handling the interval-censored cancer progression and the competing event, early treatment, and repeatedly measured biomarkers.

Increasing evidence suggests that variability in longitudinal biomarkers, in addition to their mean trajectory, carries prognostic information for time-to-event outcomes. However, standard joint models typically capture only the expected value of the biomarker process, assuming constant residual variability across individuals and time. Fully joint extensions that model within-subject variability exist but are computationally demanding and require dedicated software packages. We propose a flexible two-step approach for incorporating biomarker variability into joint models. First, residuals (or their transformations) from a mixed-effects model are used to derive subject- and time-specific measures of variability. Second, these variability measures are included in a standard joint model, allowing their association with survival to be estimated alongside the mean biomarker trajectory. Our approach can also accommodate multiple biomarkers simultaneously and is readily implemented using existing joint modeling software without custom extensions. Through simulations, we show that our method provides reasonable performance for variability effects across a range of scenarios. We further illustrate our approach using longitudinal data of white blood cell counts from a large phase III glioblastoma trial, demonstrating that both mean levels and variability of hematological markers carry prognostic information for overall survival.

Dynamic predictions for longitudinal and time-to-event outcomes have become a versatile tool in precision medicine. Our work is motivated by the application of dynamic predictions in the decision-making process for primary biliary cholangitis patients. For these patients, serial biomarker measurements (e.g., bilirubin and alkaline phosphatase levels) are routinely collected to inform treating physicians of the risk of liver failure and guide clinical decision-making. Two popular statistical approaches to derive dynamic predictions are joint modelling and landmarking. However, recently, machine learning techniques have also been proposed. Each approach has its merits, and no single method exists to outperform all others. Consequently, obtaining the best possible survival estimates is challenging. Therefore, we extend the Super Learner framework to combine dynamic predictions from different models and procedures. Super Learner is an ensemble learning technique that allows users to combine different prediction algorithms to improve predictive accuracy and flexibility. It uses cross-validation and different objective functions of performance (e.g., squared loss) that suit specific applications to build the optimally weighted combination of predictions from a library of candidate algorithms. In our work, we pay special attention to appropriate objective functions for Super Learner to obtain the most optimal weighted combination of dynamic predictions. In our primary biliary cholangitis application, Super Learner presented unique benefits due to its ability to flexibly combine outputs from a diverse set of models with varying assumptions for equal or better predictive performance than any model fit separately.

Background: Joint models for longitudinal and time-to-event data are widely used in clinical research. However, the choice of functional form linking the biomarker trajectory to event risk is often treated as a technical detail, despite its importance for model assumptions and interpretation. Default specifications may fail to capture clinically relevant features of biomarker trajectories. Methods: We provide a structured overview of functional forms linking longitudinal and survival processes in joint models. We compare association structures including instantaneous effects (current value, slope, and acceleration), cumulative and change-based formulations, shared random effects, and variability-based associations. Using longitudinal white blood cell measurements and overall survival data from the MIRAGE glioblastoma trial, we illustrate how different functional forms capture distinct features of biomarker trajectories and define different biomarker-risk relationships. Results: Instantaneous forms capture the biomarker's current level or short-term dynamics, whereas cumulative and change-based forms reflect longer-term exposure or trends. Variability-based structures quantify instability in the biomarker trajectory as an alternative prognostic signal. Association parameters depend on the functional form, biomarker scale, and time scale, and effect sizes are therefore not directly comparable. In the MIRAGE application, alternative functional forms produced different effect interpretations and, in some cases, different conclusions regarding the biomarker-risk relationship. Conclusions: The choice of functional form is a key modelling decision in joint models and determines the interpretation of the biomarker-risk association. Aligning the functional form with the scientific question is essential for valid interpretation and transparent reporting.

Heart failure (HF) is a severe potential complication of myocardial infarction(MI). However precise mechanisms underlying progression from MI to HF are not yet fully understood. Exploring how plasma proteomic profiles of post-MI patients relate to cardiac function during follow-up can give insights into pathophysiological processes that contribute to HF development. We measured 4587 circulating proteins in 246 patients hospitalized for a first anterior Q-wave MI at 1, 3 and 12 months post-MI. Echocardiographic measurements of left-ventricular (LV) end-diastolic volume (EDV), LV ejection fraction (EF), and left atrial volume (AV), were assessed at hospital discharge, 3, and 12 months. Associations between protein and echocardiographic variables were assessed using pair-wise multivariate linear mixed-effects models. Median (IQR) age was 56 (46, 69) years, and 19% were women. Twenty-eight proteins were associated with LVEDV (linked to cardiac remodeling, vascular dysfunction, oxidative stress), twelve with AV (coronary artery disease, atherosclerosis, immune system), and eight with LVEF (cardiac hypertrophy, fibrosis, inflammation). Trajectories of all three echocardiographic variables were associated with NT-proBNP and BNP. Our results give an overview of the most important mechanisms related to the deterioration of cardiac function after MI, with cardiac stress, cardiac remodeling, vascular dysfunction, and inflammation emerging as central mechanisms.

Background:Infliximab and adalimumab are effective anti-tumor necrosis factor (anti-TNF) therapies for the treatment of pediatric Crohn's disease (CD). The aim of this study was to compare the effectiveness of infliximab and adalimumab in a real-world cohort of children with CD. Methods:Data from biological-naïve children with luminal CD (age 3-18 years) who commenced anti-TNF and completed at least 1 year of follow-up were collected from the prospective multicenter observational PIBD-SETQuality study. The primary outcome was steroid-free clinical remission (SFCR), defined as clinical remission (weighted pediatric Crohn's disease activity index [wPCDAI] <12.5) without systemic steroids or luminal surgery at 1 year. The relative risk (RR) of SFCR was calculated using standardization, correcting for the following baseline covariates: age, upfront anti-TNF, C-reactive protein, erythrocyte sedimentation rate, albumin, leukocytes, disease behavior, wPCDAI, perianal disease, and concomitant immunomodulator use. Secondary outcomes included the durability of anti-TNF treatment without luminal surgery. Results:Between January 1, 2017, and June 14, 2024, 178 patients with anti-TNF were included (infliximab: n = 121 [68%], adalimumab: n = 57 [32%]). At 12 months, 34/56 (61%) patients treated with adalimumab and 66/120 (55%) patients treated with infliximab had reached SFCR. The RR of SFCR at 1 year with adalimumab compared to infliximab was 1.25 (95% confidence interval [CI] 0.94-1.66], P = .13. Adalimumab was associated with a significant lower adjusted hazard ratio (aHR) of treatment discontinuation than infliximab in patients with a concomitant immunomodulator (aHR 0.17 [95% CI 0.04-0.75], P = .020), adjusted for upfront anti-TNF. Conclusions:In this prospective cohort of children with CD, adalimumab and infliximab showed comparable clinical effectiveness 1 year after the start of anti-TNF treatment. Clinical trial registration:The ClincalTrials.gov ID of this study is NCT03571373.

Corticosteroids (CS) and exclusive enteral nutrition (EEN) are effective induction therapies for pediatric Crohn’s disease (CD), but comparative studies evaluating long-term outcomes in small bowel CD are lacking. Children (2–18 years) with newly diagnosed small bowel CD involving the ileum prospectively enrolled in the multicenter Canadian CIDsCaNN or European PIBD-SETQuality inception cohorts receiving CS or EEN induction treatment were evaluated longitudinally. The primary outcome was sustained steroid-free remission (SSFR) at 1 year. Secondary outcomes included changes in height z-scores, time-to-first-biologic and time-to-luminal-resection. Results were confirmed after propensity score matching (PSM). In total, 208 children (61

In medicine, multiple continuous outcomes are often repeatedly measured on each subject over time to assess disease severity. Usually, it is of interest to investigate the association between those outcomes, which may be measured at different time points, resulting in unbalanced data. The multivariate linear mixed-effects model (MLMM) is a popular framework for this analysis. It considers the unbalanced nature of the data and accounts for the association of the outcomes via the random effects, often assuming a multivariate normal distribution. However, measuring and understanding the degree of connection between longitudinal outcomes remains challenging. We propose to enhance the MLMM by incorporating various interpretable association structures. Specifically, we consider that multiple longitudinal outcomes are related to the primary outcome through their current value, cumulative effect (total or partial), or both. Our research is motivated by Pompe disease, a rare, inheritable, progressive metabolic myopathy. Clinically, it is important to investigate how patient-reported outcome measures (primary outcomes) are associated with physical outcomes to determine whether improvements in physical outcomes are accompanied by improvements in health-related quality of life and other patient experiences. We found a positive association between them. The proposed models are fitted under the Bayesian framework using Hamiltonian Monte Carlo.

Joint models for longitudinal and time-to-event data are widely used in many disciplines. Nonetheless, existing model comparison criteria do not indicate whether a model adequately fits the data or which components may be misspecified. We introduce a Bayesian posterior predictive checks framework for assessing a joint model's fit to the longitudinal and survival processes and their association. The framework supports multiple settings, including existing subjects, new subjects with only covariates, dynamic prediction at intermediate follow-up times, and cross-validated assessment. For the longitudinal component, goodness-of-fit is assessed through the mean, variance, and correlation structure, while the survival component is evaluated using empirical cumulative distributions and probability integral transforms. The association between processes is examined using time-dependent concordance statistics. We apply these checks to the Bio-SHiFT heart failure study, and a simulation study demonstrates that they can identify model misspecification that standard information criteria fail to detect. The proposed methodology is implemented in the freely available R package JMbayes2.

Importance: Optimal data availability for secondary use is crucial for continuous improvement in healthcare. At the same time, it is imperative to uphold patients’ rights to be informed, to control the use of their health data and to protect their privacy. To balance these two needs, we investigated which consent procedure (opt-in or opt-out) would be most supportive of data availability.Objective: This study explores an opt-in procedure versus an opt-out procedure as a consent procurement method for secondary use of routinely recorded health data, images and tissues for scientific research purposes.Design/Setting: A randomized controlled trial was performed in Erasmus Medical Center, a large tertiary hospital in the Netherlands. New, first time patients were recruited from 16 outpatient clinics and randomily assigned to either the opt-in (intervention group) or the opt-out procedure (control group), until the equally balanced sample size of 2228 was reached.Results: Patient inclusion spanned from December 2022 to September 2023. The opt-out procedure resulted in higher consent rates compared to the opt-in procedure. Differences were found for gender, socioeconomic status and country of birth.Conclusions: An opt-out procedure appears to be more effective in ensuring optimal data availability with less bias for the secondary use of health data compared to opt-in. To uphold patient control over data, it is pivotal that patients are well-informed about the consent procedure.

BACKGROUND:Implications of relabeling Grade Group 1 prostate cancer as noncancer will depend on the recommended active surveillance strategy. Whether relabeling should prompt deintensifying, prostate-specific antigen (PSA)-based active monitoring approaches is unclear. We investigated outcomes of biopsy-based active surveillance strategies vs PSA-based active monitoring for Grade Group 1 diagnoses under different patient adherence rates. METHODS:We analyzed longitudinal PSA levels and time to Grade Group 2 or higher reclassification among 850 patients with a diagnosis of Grade Group 1 disease from the Canary Prostate Active Surveillance Study (2008-2013). We then simulated 20 000 patients over 12 years, comparing Grade Group 2 or higher detection under biennial biopsy against 3 PSA-based strategies: (1) PSA (biopsy for PSA change ≥20% per year), (2) PSA plus magnetic resonance imaging (magnetic resonance imaging for PSA change ≥20% per year and biopsy for Prostate Imaging Reporting & Data System ≥3), and (3) predicted risk (biopsy for predicted upgrading risk ≥10%). RESULTS:Under biennial biopsies and 20% dropout to active treatment, 17% of patients had a 2-year or longer delay in Grade Group 2 or higher detection. The PSA strategy reduced the number of biopsies by 39% but delayed detection in 32% of patients. The PSA plus magnetic resonance imaging strategy reduced the number of biopsies by 52%, with a 34% delay. The predicted risk strategy reduced the number of biopsies by 31%, with only an 8% delay. These findings are robust to biopsy sensitivity and confirmatory biopsy. CONCLUSIONS:Prostate-specific antigen-based active monitoring could substantially reduce biopsy frequency; however, a precision strategy based on an individual upgrading risk is most likely to minimize delays in detection of disease progression. This strategy may be preferred if active surveillance is deintensified under relabeling, provided patient adherence remains unaffected.

Active surveillance (AS), where biopsies are conducted to detect cancer progression, has been acknowledged as an efficient way to reduce the overtreatment of prostate cancer. Most AS cohorts use fixed biopsy schedules for all patients. However, the ideal test frequency remains unknown, and the routine use of such invasive tests burdens the patients. An emerging idea is to generate personalized biopsy schedules based on each patient's progression-specific risk. To achieve that, we propose the interval-censored cause-specific joint model (ICJM), which models the impact of longitudinal biomarkers on cancer progression while considering the competing event of early treatment initiation. The underlying likelihood function incorporates the interval-censoring of cancer progression, the competing risk of treatment, and the uncertainty about whether cancer progression occurred since the last biopsy in patients that are right-censored or experience the competing event. The model can produce patient-specific risk profiles until a horizon time. If the risk exceeds a certain threshold, a biopsy is conducted. The optimal threshold can be chosen by balancing two indicators of the biopsy schedules: the expected number of biopsies and expected delay in detection of cancer progression. A simulation study showed that our personalized schedules could considerably reduce the number of biopsies per patient by 34%-54% compared to the fixed schedules, though at the cost of a slightly longer detection delay.

Abstract Background Immune-mediated diseases (IMIDs), like Crohn’s disease (CD), ulcerative colitis (UC), psoriatic arthritis (PsA) and type I diabetes (T1D), have common features, including intestinal microbial dysbiosis. However, it is unclear whether dysbiosis contributes to IMID pathogenesis and whether the host immune system perceives dysbiotic commensals. In CD, high IgG and T-cell responses to Lachnospiraceae-derived flagellins associate with complex disease, while Lachnospiraceae fecal abundance is decreased in CD, suggesting that individual variation in the host-microbial mutualism is important. We studied IgG responses to dysbiotic microbial species across IMIDs, and questioned whether changes in these IgG responses are homogeneous within each IMID and shared among IMIDs. Methods Using shotgun metagenomic sequencing of feces from 6 IMID cohorts (n=5650), dysbiosis was extrapolated to species level. Plasma IgG reactivity to lysates of 71 dysbiotic species was measured in adult CD (aCD, n=50), adult UC (aUC, n=50), PsA (n=100), rheumatoid arthritis (RA, n=74), T1D (n=75), healthy controls (aHC, n=97), pediatric CD (pCD, n=103), pediatric UC (pUC, n=48), pediatric HC (pHC, n=58), pediatric celiac disease (CeD, n=103) and no-CeD HC (n=68). Multiple ordinal regression analysis was performed corrected for age, sex and multiple testing. Results Anti-microbial IgG responses compared to age-matched HC were increased in aCD (2/71); pCD (46/71); PsA (3/71); RA (1/71); decreased in T1D (10/71) and pUC (1/71), and not different in CeD and aUC. Shared increased responses occurred to: Klebsiella oxytoca and Roseburia inulinivorans in aCD and pCD, Streptococcus parasanguinis and Streptococcus vestibularis in PsA and pCD, and Acidaminococcus intestini in both arthritic diseases PsA and RA. As changes were heterogeneous within each IMID, hierarchical clustering of all anti-microbial responses across adult IMIDs was performed. Clusters had a mix of IMIDs, uncovering shared response patterns across IMIDs. Although aCD and aUC patients had disease-specific responses, their overall anti-microbial response pattern was not different from other IMIDs. In contrast, pCD patients were clearly distinct from pUC and pHC, with high number of significantly increased anti-microbial IgG responses and separate hierarchical clustering, demonstrating overall increased anti-microbial responses in pCD versus aCD. Interestingly, clustering grouped pCD patients with similar clinicopathological parameters, arguing that anti-microbial IgG response patterns may relate to disease pathogenesis. Conclusion We show that the immune system perceives dysbiotic commensals and uncover shared and non-shared anti-microbial IgG responses among adult and pediatric IBD and other IMID patients.

BACKGROUND:Dorsal preservation is a recently popularized technique to lower the nasal dorsum without opening the cartilaginous vault. Improved nasal breathing has been reported after lowering an intact dorsum using preservation techniques, suggesting that septal deprojection opens the internal nasal valves. The goal of this study was to evaluate the effect of dorsal preservation on internal nasal valve dimensions in noses with an overprojected cartilaginous septum. METHODS:Ten postmortem human specimen heads with a tension nose deformity were imaged using ultra-high-resolution photon-counting detector computed tomography, after which a low-strip let-down technique was performed on each specimen. Following dorsal lowering, scans were repeated and internal nasal valve angle and area of pre- and postoperative scans were measured by three assessors. Differences in pre- and postoperative measurements were assessed using a linear mixed-effects model. RESULTS:A significant increase in both internal nasal valve angle (4.28 degrees, 95% CI: 3.11-5.46) and area (8.86 mm2, 95% CI: 7.11-10.61) was demonstrated after dorsal lowering. Interrater reliability among the three assessors was high, with ICCs ranging from 0.839 to 0.985. CONCLUSIONS:This study provides morphological evidence that the internal nasal valve widens after mobilizing the dorsum and lowering the septum, without alterations to the cartilaginous vault itself. Although these results suggest that low-strip dorsal preservation may be effective in treating the functionally impaired tension nose, clinical studies are necessary to substantiate these findings in live tissue. LEVEL OF EVIDENCE:NA Laryngoscope, 135:2359-2366, 2025.

Background The striking link between Cushing syndrome, the metabolic syndrome (MetS), and cardiovascular disease suggests that long-term exposure to high glucocorticoid levels catalyzes cardiometabolic deterioration. However, the relation of subtle variations in long-term glucocorticoid levels with MetS remains poorly understood. Specifically, little is known about potential moderating roles of age, sex, and mental health status in this association.Design We investigated the association of long-term glucocorticoid levels with MetS using data of 1405 participants (73.4% women) of the Netherlands Study of Depression and Anxiety. Predictors included hair cortisol and cortisone levels. Outcomes were MetS presence, number of MetS components, and individual component (ie, diastolic blood pressure, waist circumference, and fasting glucose, high-density lipoprotein cholesterol, and triglycerides). We investigated potential interactions with age, sex, and mental health status.Results Hair glucocorticoid levels were positively associated with MetS presence (OR = 1.27; 95% CI = 1.11-1.44, and OR = 1.32; 95% CI = 1.14-1.52 for hair cortisol and cortisone, respectively), number of MetS components, waist circumference, and triglyceride levels. Hair cortisol, but not cortisone, was in trend associated with diastolic blood pressure and high-density lipoprotein cholesterol levels. No associations were seen with blood glucose. Of note, the relationship of hair cortisone with MetS was stronger among younger compared to older individuals (OR = 1.95; 95% CI = 1.50-2.54 vs OR = 1.14; 95% CI = .96-1.35 in younger vs older participants).Conclusion Long-term biological stress, measured through hair glucocorticoid levels, is associated with MetS presence, especially among younger individuals. Prospective studies need to evaluate the extent to which hair cortisol and cortisone add to standard risk factors when predicting future cardiometabolic diseases.

Predicting mortality in COVID-19 ARDS may support ICU clinical decision-making. Biomarkers of innate immunity, coagulation, endothelial injury, and fibroproliferation have been studied as predictors. We aimed to examine whether trends in plasma biomarkers predict ICU mortality and to explore underlying biological processes through pathway analysis. Additionally, we explored whether HDS changes biomarker trajectories in COVID-19 ARDS. In this observational study, we included patients with COVID-19 ARDS admitted to the ICU of an academic hospital in Rotterdam between February 2020 and February 2022. In repeated plasma samples, 64 biomarkers were measured. Joint modeling assessed the association between biomarker changes and ICU mortality, adjusting for age, sex, BMI, and HDS. Protein–protein interaction and gene ontology enrichment analyses were performed using STRING, Cytoscape, and DAVID EASE. Biomarker trajectories were compared between HDS-treated and non-treated patients, adjusting for timing, SOFA score, and tocilizumab. One hundred and sixty-two patients were included and 43 died during ICU stay. A doubling in the values of 26 biomarkers over the next day was predictive of ICU mortality (HRs 0.16–8.56; q < 0.05). Gene ontology enrichment analysis identified 19 overrepresented biological processes (FDR ≤ 0.05), with highest fold enrichment for macrophage chemotaxis, negative regulation of bone resorption, and leukocyte cell–cell adhesion. Forty-eight patients received HDS at a median of 6 ICU days. HDS significantly changed the trajectories of four mortality-associated biomarkers: Albumin and lactoferrin decreased, while CRP and VEGF increased. In COVID-19 ARDS, repeated biomarker measurements demonstrate a systemic inflammatory state associated with mortality. HDS changed trends of several biomarkers, but did not reduce those associated with fatal outcomes. Trial registration: ClinicalTrials.gov NCT05403359; https://clinicaltrials.gov/ct2/show/NCT05403359

Patient monitoring is routinely used to detect disease aggravation in many chronic conditions. We propose an adaptive scheduling strategy based on dynamic individual risk predictions that can improve the efficiency of monitoring programs that incorporate multiple longitudinal measurements and competing events. It is motivated by stable chronic heart failure (CHF) patients who are periodically seen to assess the risk of disease aggravation based on multiple patient characteristics and circulating marker protein levels such as NT-proBNP and troponin. We assess the performance of the adaptive strategy versus fixed schedule alternatives using a simulation study based on the Bio-SHiFT study, a cohort of stable CHF patients.