Dynamic and concordance-assisted learning for risk stratification with application to Alzheimer's disease.

Dynamic prediction models capable of retaining accuracy by evolving over time could play a significant role for monitoring disease progression in clinical practice. In biomedical studies with long-term follow up, participants are often monitored through periodic clinical visits with repeat measurements until an occurrence of the event of interest (e.g. disease onset) or the study end. Acknowledging the dynamic nature of disease risk and clinical information contained in the longitudinal markers, we propose an innovative concordance-assisted learning algorithm to derive a real-time risk stratification score. The proposed approach bypasses the need to fit regression models, such as joint models of the longitudinal markers and time-to-event outcome, and hence enjoys the desirable property of model robustness. Simulation studies confirmed that the proposed method has satisfactory performance in dynamically monitoring the risk of developing disease and differentiating high-risk and low-risk population over time. We apply the proposed method to the Alzheimer's Disease Neuroimaging Initiative data and develop a dynamic risk score of Alzheimer's Disease for patients with mild cognitive impairment using multiple longitudinal markers and baseline prognostic factors.

Coronary flow capacity and survival prediction after revascularization: physiological basis and clinical implications.

BACKGROUND AND AIMS: Coronary flow capacity (CFC) is associated with an observed 10-year survival probability for individual patients before and after actual revascularization for comparison to virtual hypothetical ideal complete revascularization. METHODS: Stress myocardial perfusion (mL/min/g) and coronary flow reserve (CFR) per pixel were quantified in 6979 coronary artery disease (CAD) subjects using Rb-82 positron emission tomography (PET) for CFC maps of artery-specific size-severity abnormalities expressed as percent left ventricle with prospective follow-up to define survival probability per-decade as fraction of 1.0. RESULTS: Severely reduced CFC in 6979 subjects predicted low survival probability that improved by 42% after revascularization compared with no revascularization for comparable severity (P = .0015). For 283 pre-and-post-procedure PET pairs, severely reduced regional CFC-associated survival probability improved heterogeneously after revascularization (P < .001), more so after bypass surgery than percutaneous coronary interventions (P < .001) but normalized in only 5.7%; non-severe baseline CFC or survival probability did not improve compared with severe CFC (P = .00001). Observed CFC-associated survival probability after actual revascularization was lower than virtual ideal hypothetical complete post-revascularization survival probability due to residual CAD or failed revascularization (P < .001) unrelated to gender or microvascular dysfunction. Severely reduced CFC in 2552 post-revascularization subjects associated with low survival probability also improved after repeat revascularization compared with no repeat procedures (P = .025). CONCLUSIONS: Severely reduced CFC and associated observed survival probability improved after first and repeat revascularization compared with no revascularization for comparable CFC severity. Non-severe CFC showed no benefit. Discordance between observed actual and virtual hypothetical post-revascularization survival probability revealed residual CAD or failed revascularization.

The Time-Dependent Association Between Irritable Bowel Syndrome and All-Cause and Cause-Specific Mortality: A Prospective Cohort Study Within the UK Biobank.

INTRODUCTION: Irritable bowel syndrome (IBS) is one of the most common functional gastrointestinal disorders, but few studies have evaluated mortality risks among individuals with IBS. We explored the association between IBS and all-cause and cause-specific mortality in the UK Biobank. METHODS: We included 502,369 participants from the UK Biobank with mortality data through 2022. IBS was defined using baseline self-report and linkage to primary care or hospital admission data. We estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause and cause-specific mortality using multivariable Cox proportional hazards regression models within partitioned follow-up time categories (0-5, >5-10, and >10 years). RESULTS: A total of 25,697 participants (5.1%) had a history of IBS at baseline. After a median follow-up of 13.7 years, a total of 44,499 deaths occurred. Having an IBS diagnosis was strongly associated with lower risks of all-cause (HR = 0.70, 95% CI = 0.62-0.78) and all-cancer (HR = 0.69, 95% CI = 0.60-0.79) mortality in the first 5 years of follow-up. These associations were attenuated over follow-up, but even after 10 years of follow-up, associations remained inverse (all-cause: HR = 0.89, 95% CI = 0.84-0.96; all-cancer: HR = 0.87, 95% CI = 0.78-0.97) after full adjustment. Individuals with IBS had decreased risk of mortality from breast, prostate, and colorectal cancers in some of the follow-up time categories. DISCUSSION: We found that earlier during follow-up, having diagnosed IBS was associated with lower mortality risk, and the association attenuated over time. Additional studies to understand whether specific factors, such as lifestyle and healthcare access, explain the inverse association between IBS and mortality are needed.

Telemental Health Use Is Associated With Lower Health Care Spending Among Medicare Beneficiaries With Major Depression.

BACKGROUND: Some policymakers are concerned that expanding telehealth coverage may increase Medicare expenditures. However, there is limited evidence on the association of telehealth use with utilization and spending among Medicare beneficiaries with major depression. OBJECTIVE: To examine the differences in spending and utilization among telemental health users and non-telemental health users with major depression. METHODS: We examined 2014-2019 traditional Medicare claims data for beneficiaries aged ≥50 years with major depression in Texas. Multivariable generalized linear models were used to assess the relationships between telemental health use and Medicare spending and utilization while adjusting for patient demographics and programmatic and clinical factors. RESULTS: In each of the years between 2014 and 2019, an average of 4.6% Medicare beneficiaries with major depression had at least 1 telemental health visit. Compared with beneficiaries without a telemental health visit, those who had a telemental health visit were significantly more likely to be enrolled in Medicaid, be Medicare eligible due to a disability, live in a lower income area or in a rural area, and have a higher comorbidity index. Beneficiaries utilizing telemental health services incurred higher unadjusted Medicare spending than those not receiving telemental health services. However, this difference appeared due to beneficiary and programmatic characteristics rather than telemental health use. Adjusting for model covariates, the telemental health group had lower overall per member per year predicted spending, inpatient admissions, and emergency department visits than non-telemental health users. CONCLUSION: Our findings suggest that telemental health care use may improve access to mental health care without increasing Medicare spending among telemental health users in Texas.

Addressing subject heterogeneity in time-dependent discrimination for biomarker evaluation.

Accurate discrimination has been the central goal in identifying biomarkers for monitoring disease progression and early detection. Acknowledging the fact that discrimination accuracy of biomarkers for a time-to-event outcome often changes over time, local measures such as the time-dependent receiver operating characteristic curve and its area under the curve (AUC) are used to assess time-dependent predictive discrimination. However, such measures do not address subject heterogeneity, although the impact of covariates including demographics, disease-related characteristics, and other clinical information on the discriminatory performance of biomarkers needs to be investigated before their clinical use. We propose the covariate-specific time-dependent AUC, a measure for covariate-adjusted discrimination. We develop a regression model on the covariate-specific time-dependent AUC to understand how and in what magnitude the covariates influence biomarker performance. Then we construct a pseudo partial-likelihood for estimation and inference. This is followed by our establishing the asymptotic properties of the proposed estimators and provide variance estimation. The simulation studies and application to the AIDS Clinical Trials Group 175 data demonstrate that the proposed method offers an informative tool for inferring covariate-specific and time-dependent predictive discrimination.

Rates and Predictors of Rapid Eye Movement Sleep Behavior Disorder Symptoms Among Post-9/11 Veterans.

OBJECTIVE: Posttraumatic stress disorder (PTSD) and traumatic brain injury (TBI), which are prevalent conditions among post-9/11 veterans, increase risks of rapid eye movement (REM) sleep behavior disorder (RBD) and degenerative synucleinopathy. Rates and predictors of RBD symptoms were investigated by screening post-9/11 veterans for RBD with a validated questionnaire. METHODS: In this cross-sectional analysis, consecutive patients in the Houston Translational Research Center for TBI and Stress Disorders (TRACTS) were screened with the English translation of the RBD Questionnaire-Hong Kong (RBDQ-HK). In addition to data from the standard TRACTS battery, systematic chart review was used to identify known sleep disorders mimicking or manifesting RBD. RESULTS: Of the 119 patients with available RBDQ-HK scores, 71 (60%) and 65 (55%) screened positive for RBD, when a total score ≥21 and a factor 2 score ≥8 were used as cutoff scores, respectively. Univariable analyses with both cutoffs showed consistent associations between a positive RBDQ-HK screen and global sleep quality, number of TBI exposures, and PTSD severity. Multivariable logistic regression with total score ≥21 as a cutoff indicated that PTSD severity (odds ratio=1.06, 95% CI=1.02-1.10) and number of TBIs (odds ratio=1.63, 95% CI=1.16-2.41) were independent predictors of a positive screen, whereas global sleep quality was no longer significant. Multivariable logistic regression with factor 2 score ≥8 as a cutoff showed similar results. CONCLUSIONS: Interdisciplinary parasomnia assessment, further validation of RBD screens, and standardized reporting of REM sleep without atonia could provide necessary information on the pathophysiological relationships linking PTSD, TBI, RBD symptoms, and ultimately synucleinopathy risk among post-9/11 veterans.

Improved mortality prediction for pediatric acute liver failure using dynamic prediction strategy.

OBJECTIVES: To develop and validate a prediction tool for pediatric acute liver failure (PALF) mortality risks that captures the rapid and heterogeneous clinical course for accurate and updated prediction. METHODS: Data included 1144 participants with PALF enrolled during three phases of the PALF registry study over 15 years. Using joint modeling, we built a dynamic prediction tool for mortality by combining longitudinal trajectories of multiple laboratory and clinical variables. The predictive performance for 7-day and 21-day mortality was assessed using the area under curve (AUC) through cross-validation and split-by-time validation. RESULTS: We constructed a prognostic joint model that combines the temporal trajectories of international normalized ratio, total bilirubin, hepatic encephalopathy, platelet count, and serum creatinine. Dynamic prediction using updated information improved predictive performance over static prediction using the information at enrollment (Day 0) only. In cross-validation, AUC increased from 0.784 to 0.887 when measurements obtained between Days 1 and 2 were incorporated. AUC remained similar when we used the earlier subset of the sample for training and the later subset for testing. CONCLUSIONS: Serial measurements of five variables in the first few days of PALF capture the dynamic clinical course of the disease and improve risk prediction for mortality. Continuous disease monitoring and updating risk prognosis are beneficial for timely and judicious medical decisions.

Congenital Heart Defects and Concurrent Diagnoses in Influenza Hospitalization in the Pediatric Health Information System Study, 2004-2019.

Influenza is associated with adverse outcomes in children, although modification by additional medical conditions is not well-documented. We aimed to compare outcomes in children with versus without congenital heart defects (CHDs) who were hospitalized for influenza. We retrospectively evaluated patients 1-18y hospitalized for influenza in the Pediatric Health Information (PHIS) database from 2004 to 2019. Outcomes were compared by CHD presence and then by CHD severity (minor biventricular, major biventricular, and single ventricle disease) using log-binomial regression adjusted for propensity scores accounting for age at admission, sex, and history of asthma. Outcomes included inpatient mortality, intensive care unit (ICU) admission, mechanical ventilation, and length of stay (LOS) > 12 days. To evaluate for effect modification by genetic diagnoses, analyses were repeated stratified by CHD and genetic diagnosis. Among 55,161 children hospitalized for influenza, 2369 (4.3%) had CHDs, including 963 with minor biventricular, 938 with major biventricular, and 468 with single ventricle CHDs. Adjusting for propensity scores, children with CHDs had higher mortality (4.1% versus 0.9%) compared to those without CHDs (risk ratio [RR] 2.5, 95% confidence interval [CI] 1.9-3.4). Children with CHDs were at higher risk of mechanical ventilation (RR 1.6, 95% CI 1.6-1.7), ICU admission (RR 1.9, 95% CI 1.8-2.1), and LOS > 12 days (RR 2.2, 95% CI 2.0-2.3). Compared to those with neither CHD nor genetic condition, children with both had significantly higher risk of all outcomes, with the largest difference for LOS > 12 days (RR 2.3, 95% CI 2.0-2.7). Children with CHDs hospitalized for influenza are particularly susceptible to adverse outcomes compared to those without CHDs. Future studies are needed to corroborate findings in light of influenza vaccination.

Dynamic risk score modeling for multiple longitudinal risk factors and survival.

Modeling disease risk and survival using longitudinal risk factor trajectories is of interest in various clinical scenarios. The capacity to build a prognostic model using the trajectories of multiple longitudinal risk factors, in the presence of potential dependent censoring, would enable more informed, personalized decision making. A dynamic risk score modeling framework is proposed for multiple longitudinal risk factors and survival in the presence of dependent censoring, where both events depend on participants' post-baseline clinical progression and form a competing risks structure. The model requires relatively few random effects regardless of the number of longitudinal risk factors and can therefore accommodate multiple longitudinal risk factors in a parsimonious manner. The proposed method performed satisfactorily in extensive simulation studies. It is further applied to the motivating registry study on pediatric acute liver failure to model death using the trajectories of multiple clinical and biochemical markers. Once established, the model yields an easily calculable longitudinal risk score that can be used for disease monitoring among future patients.

A double-robust test for high-dimensional gene coexpression networks conditioning on clinical information.

It has been increasingly appealing to evaluate whether expression levels of two genes in a gene coexpression network are still dependent given samples' clinical information, in which the conditional independence test plays an essential role. For enhanced robustness regarding model assumptions, we propose a class of double-robust tests for evaluating the dependence of bivariate outcomes after controlling for known clinical information. Although the proposed test relies on the marginal density functions of bivariate outcomes given clinical information, the test remains valid as long as one of the density functions is correctly specified. Because of the closed-form variance formula, the proposed test procedure enjoys computational efficiency without requiring a resampling procedure or tuning parameters. We acknowledge the need to infer the conditional independence network with high-dimensional gene expressions, and further develop a procedure for multiple testing by controlling the false discovery rate. Numerical results show that our method accurately controls both the type-I error and false discovery rate, and it provides certain levels of robustness regarding model misspecification. We apply the method to a gastric cancer study with gene expression data to understand the associations between genes belonging to the transforming growth factor ß signaling pathway given cancer-stage information.

Post-Traumatic Stress Disorder and Risk of Degenerative Synucleinopathies: Systematic Review and Meta-Analysis.

OBJECTIVE: A systematic review was conducted to answer whether adult-onset post-traumatic stress disorder (PTSD) is associated with increased risk of Parkinson's disease (PD) and related synucleinopathies. DESIGN: A systematic search of Medline (Ovid), Embase (Elsevier), PsycInfo (Ovid), Cochrane Library (Wiley), and Web of Science (Clarivate) was performed using MeSH headings and equivalent terms for PTSD, PD, DLB, and related disorders. SETTING: No restrictions. PARTICIPANTS: Eligible articles were published in peer-reviewed journals, sampled adult human populations, and treated PTSD and degenerative synucleinopathies as exposures and outcomes, respectively. MEASUREMENTS: Extracted data included diagnostic methods, sample characteristics, matching procedures, covariates, and effect estimates. Bias assessment was performed with the Newcastle-Ottawa scale. Hazard ratios were pooled using the random effects model, and the Hartung-Knapp adjustment was applied due to the small number of studies. RESULTS: A total of six articles comprising seven unique samples (total n = 1,747,378) met eligibility criteria. The risk of PD was reported in three retrospective cohort studies and one case-control study. Risk of DLB was reported in one retrospective cohort, one case-control, and one prospective cohort study. No studies addressed potential relationships with multiple system atrophy or pure autonomic failure. Meta-analysis of hazard ratios from four retrospective cohort studies supported the hypothesis that incident PTSD was associated with PD and DLB risk (pooled HR 1.88, 95% C.I. 1.08-3.24; p = 0.035). CONCLUSIONS: The sparse literature to-date supports further investigations on the association of mid- to late-life PTSD with Parkinson's and related neurodegenerative disorders.

Conditional concordance-assisted learning under matched case-control design for combining biomarkers for population screening.

Incorporating promising biomarkers into cancer screening practices for early-detection is increasingly appealing because of the unsatisfactory performance of current cancer screening strategies. The matched case-control design is commonly adopted in biomarker development studies to evaluate the discriminative power of biomarker candidates, with an intention to eliminate confounding effects. Data from matched case-control studies have been routinely analyzed by the conditional logistic regression, although the assumed logit link between biomarker combinations and disease risk may not always hold. We propose a conditional concordance-assisted learning method, which is distribution-free, for identifying an optimal combination of biomarkers to discriminate cases and controls. We are particularly interested in combinations with a clinically and practically meaningful specificity to prevent disease-free subjects from unnecessary and possibly intrusive diagnostic procedures, which is a top priority for cancer population screening. We establish asymptotic properties for the derived combination and confirm its favorable finite sample performance in simulations. We apply the proposed method to the prostate cancer data from the carotene and retinol efficacy trial (CARET).

Reimbursement for HPV Vaccine Cost in the Private Sector: A Comparison Across Specialties.

The human papillomavirus (HPV) vaccine is the most expensive of all routinely recommended pediatric vaccines. Adequate cost reimbursement by 3rd-party payers is a critical enabling factor for clinicians to continue offering vaccines. This study found that net returns from HPV vaccine cost reimbursements are lowest for family physicians ($0.34/dose) and highest for pediatricians ($5.08/dose). Furthermore, a $1 increment in return was associated with an increase in HPV vaccine doses administered (highest for family physicians; 0.08% per dollar). Reimbursement for HPV vaccine costs by private payers is adequate; however, return margins are small for non-pediatric specialties.

Semiparametric regression modeling of the global percentile outcome.

When no single outcome is sufficient to capture the multidimensional impairments of a disease, investigators often rely on multiple outcomes for comprehensive assessment of global disease status. Methods for assessing covariate effects on global disease status include the composite outcome and global test procedures. One global test procedure is the O'Brien's rank-sum test, which combines information from multiple outcomes using a global rank-sum score. However, existing methods for the global rank-sum do not lend themselves to regression modeling. We consider sensible regression strategies for the global percentile outcome (GPO), under the transformed linear model and the monotonic index model. Posing minimal assumptions, we develop estimation and inference procedures that account for the special features of the GPO. Asymptotics are established using U-statistic and U-process techniques. We illustrate the practical utilities of the proposed methods via extensive simulations and application to a Parkinson's disease study.

A flexible-hazards cure model with application to patients with soft tissue sarcoma.

In medical research, it is often of great interest to have an accurate estimation of cure rates by different treatment options and for different patient groups. If the follow-up time is sufficiently long and the sample size is large, the proportion of cured patients will make the Kaplan-Meier estimator of survival function have a flat plateau at its tail, whose value indicates the overall cure rate. However, it may be difficult to estimate and compare the cure rates for all the subsets of interest in this way, due to the limit of sample sizes and curse of dimensionality. In the current literature, most regression models for estimating cure rates assume proportional hazards (PH) between different subgroups. It turns out that the estimation of cure rates for subgroups is highly sensitive to this assumption, so more flexible models are needed, especially when this PH assumption is clearly violated. We propose a new cure model to simultaneously incorporate both PH and non-PH scenarios for different covariates. We develop a stable and easily implementable iterative procedure for parameter estimation through maximization of the nonparametric likelihood function. The covariance matrix is estimated by adding perturbation weights to the estimation procedure. In simulation studies, the proposed method provides unbiased estimation for the regression coefficients, survival curves, and cure rates given covariates, while existing models are biased. Our model is applied to a study of stage III soft tissue sarcoma and provides trustworthy estimation of cure rates for different treatment and demographic groups.

Tdap-HPV vaccination bundling in the USA: Trends, predictors, and implications for vaccine series completion.

The Centers for Disease Control and Prevention (CDC) promotes taking a 'bundling approach' (i.e., administering Tetanus, diphtheria toxoids, and acellular pertussis [Tdap] and human papillomavirus [HPV] vaccines in the same way and on the same day) for adolescent vaccinations. Recent trends and patterns in Tdap-HPV vaccination bundling in the USA remain undocumented. In addition, the implications of bundling Tdap-HPV vaccination for HPV vaccine series completion remain unknown. To address these critical knowledge gaps, we performed a retrospective study using a nationwide sample of privately insured adolescents (Optum's de-identified Clinformatics® Data Mart Database). Tdap-HPV vaccination bundling (per 100 Tdap vaccination encounters) during 2014-2018 was estimated overall, for 50 states, and by adolescents' age, sex, and provider specialties. Survival model estimated the likelihood of series completion among 9-14-year-old adolescents. From 2014 to 2018, 560,806 adolescents received a Tdap vaccine of which 172,604 (30.8%) received the HPV vaccines on the same day. Tdap-HPV vaccination bundling (per 100 Tdap vaccinations) increased nationally, from 22.9 in 2014 to 39.1 in 2018 (Ptrend < 0.001); bundling was lowest in New York and New Jersey. The likelihood of receiving the Tdap and HPV vaccines bundled was higher for young and female adolescents. Adolescents who received their first HPV vaccine bundled with the Tdap vaccine were more likely to complete the series compared to those who received it alone (Hazards Ratio = 1.45; 1.43-1.48). HPV vaccination bundling has increased in recent years in the USA. The increased likelihood of HPV vaccine series completion provides important evidence supporting the adoption of same-day Tdap-HPV vaccine administration in clinical practice to boost HPV vaccination coverage.

Estimation and inference of predictive discrimination for survival outcome risk prediction models.

Accurate risk prediction has been the central goal in many studies of survival outcomes. In the presence of multiple risk factors, a censored regression model can be employed to estimate a risk prediction rule. Before the prediction tool can be popularized for practical use, it is crucial to rigorously assess its prediction performance. In our motivating example, researchers are interested in developing and validating a risk prediction tool to identify future lung cancer cases by integrating demographic information, disease characteristics and smoking-related data. Considering the long latency period of cancer, it is desirable for a prediction tool to achieve discriminative performance that does not weaken over time. We propose estimation and inferential procedures to comprehensively assess both the overall predictive discrimination and the temporal pattern of an estimated prediction rule. The proposed methods readily accommodate commonly used censored regression models, including the Cox proportional hazards model and the accelerated failure time model. The estimators are consistent and asymptotically normal, and reliable variance estimators are also developed. The proposed methods offer an informative tool for inferring time-dependent predictive discrimination, as well as for comparing the discrimination performance between candidate models. Applications of the proposed methods demonstrate enduring performance of the risk prediction tool in the PLCO study and detected decaying performance in a study of liver disease.

On the time-varying predictive performance of longitudinal biomarkers: Measure and estimation.

In many biomedical studies, participants are monitored at periodic visits until the occurrence of the failure event. Biomarkers are often measured repeatedly during these visits, and such measurements can facilitate updated disease prediction. In this work, we propose a two-dimensional incident dynamic area under curve (AUC), to capture the variability due to both the biomarker assessment time and the prediction time to comprehensively quantify the predictive performance of a longitudinal biomarker. We propose a pseudo partial-likelihood to achieve consistent estimation of the AUC under two realistic scenarios of visit schedules. Variance estimation methods are designed to facilitate inferential procedures. We examine the finite-sample performance of our method through extensive simulations. The methods are applied to a study of chronic myeloid leukemia to evaluate the predictive performance of longitudinally collected gene expression levels.

Evaluation of competing risks prediction models using polytomous discrimination index.

For competing risks data, it is often important to predict a patient's outcome status at a clinically meaningful time point after incorporating the informative censoring due to competing risks. This can be done by adopting a regression model that relates the cumulative incidence probabilities to a set of covariates. To assess the performance of the resulting prediction tool, we propose an estimator of the polytomous discrimination index applicable to competing risks data, which can quantify a prognostic model's ability to discriminate among subjects from different outcome groups. The proposed estimator allows the prediction model to be subject to model misspecification and enjoys desirable asymptotic properties. We also develop an efficient computation algorithm that features a computational complexity of O(n log n). A perturbation resampling scheme is developed to achieve consistent variance estimation. Numerical results suggest that the estimator performs well under realistic sample sizes. We apply the proposed methods to a study of monoclonal gammopathy of undetermined significance.

Lorsque des données comportent des risques concurrents, il est souvent important de prédire l'issue pour un patient à un temps significatif d'un point de vue clinique après avoir incorporé la censure informative due aux risques concurrents, ce qui peut être fait en adaptant un modèle de régression qui lie les probabilités cumulatives d'incidence à un ensemble de covariables. Pour évaluer la performance des outils de prévision qui en découlent, les auteurs proposent un estimateur de l'indice de discrimination polytomique applicable aux données avec des risques concurrents, et qui permet de quantifier l'habileté d'un modèle de pronostic à discriminer entre les sujets de différents groupes de dénouement. L'estimateur proposé permet au modèle de prévision un certain nombre de mauvaises classifications et exhibe des propriétés asymptotiques souhaitables. Les auteurs développent un algorithme efficace dont la complexité de calcul est d'ordre O(n log n). Ils proposent un schéma de rééchantillonnage pour obtenir une estimation convergente de la variance et présentent des résultats numériques qui suggèrent que leur estimateur offre de bonnes performances avec des tailles d'échantillons réalistes. Ils appliquent finalement leur méthode à une étude de gammopathie monoclonale dont la significativité n'est pas établie.

Quantile association regression on bivariate survival data.

The association between two event times is of scientific importance in various fields. Due to population heterogeneity, it is desirable to examine the degree to which local association depends on different characteristics of the population. Here we adopt a novel quantile-based local association measure and propose a conditional quantile association regression model to allow covariate effects on local association of two survival times. Estimating equations for the quantile association coefficients are constructed based on the relationship between this quantile association measure and the conditional copula. Asymptotic properties for the resulting estimators are rigorously derived, and induced smoothing is used to obtain the covariance matrix. Through simulations we demonstrate the good practical performance of the proposed inference procedures. An application to age-related macular degeneration (AMD) data reals interesting varying effects of the baseline AMD severity score on the local association between two AMD progression times.

L'association entre les temps jusqu'à deux événements revêt une importance scientifique dans plusieurs domaines. Il est intéressant de pouvoir observer à quel point leur degré d'association local dépend de différentes caractéristiques d'une population lorsque celle-ci exhibe de l'hétérogénéité. Les auteures adoptent une nouvelle mesure d'association locale basée sur les quantiles et proposent un modèle conditionnel de régression quantile permettant aux covariables d'avoir un effet sur l'association locale de deux temps de survie. Elles construisent les équations d'estimation pour les coefficients du modèle à partir de la relation entre cette mesure d'association quantile et la copule conditionnelle. Elles dérivent rigoureusement les propriétés asymptotiques des estimateurs résultants et utilisent un lissage induit afin d'obtenir la matrice de covariance. À l'aide de simulations, les auteures démontrent les bonne performances pratiques des procédures d'inférence proposées. Elles présentent une application à des données de dégénérescence maculaire liées à l'âge (DMA) qui montrent des effets variables du score de sévérité de base de la DMA sur l'association locale entre deux temps de progression de la DMA.