Multivariate survival models induced by genetic frailties, with application to linkage analysis

Reliability of repairable rolls used in Rolling Mills was modeled taking to survival modeling route to address the presence of recurrent failure events, censoring event and terminal event processes observed longitudinally in rolls. All the event processes were influenced by measured and unmeasured covariates. Prior to fitting appropriate model, Archimedean Gumbel and Clayton Copula analyses confirmed that the measured covariates had no significant dependence structure. Since the censoring events were “informative terminations”, joint shared frailty multivariate survival models involving Log-Normal, Gamma and Log Gamma frailty distributions were fitted to recurrent and terminal events data where, the ‘frailty' parameter represented the effect of unmeasured covariates related to condition of rolling operation. Gaussian quadrature method helped in estimating the model parameters. Statistical significance of the frailty parameter and its variance in all the models confirmed existence of heterogeneity across the recurrent failure events within and between rolls on account of unmeasured covariates. The statistically significant positive association between hazard functions of both the recurrent failure events and the terminal events justified joint modeling approach to the recurrent events and the terminal events processes observed in rolls. The joint lognormal shared frailty multivariate survival models were considered appropriate for analyzing the reliability of rolls.

Download Full-text

Multivariate Survival Models Incorporating Hidden Truncation

Distributions With Given Marginals and Statistical Modelling ◽

10.1007/978-94-017-0061-0_2 ◽

2002 ◽

pp. 9-19

Author(s):

Barry C. Arnold ◽

Robert J. Beaver

Keyword(s):

Survival Models ◽

Multivariate Survival ◽

Hidden Truncation

Download Full-text

Risk assessment of progression to severe conditions for patients with COVID-19 pneumonia: a single-center retrospective study

10.1101/2020.03.25.20043166 ◽

2020 ◽

Cited By ~ 12

Author(s):

Lijiao Zeng ◽

Jialu Li ◽

Mingfeng Liao ◽

Rui Hua ◽

Pilai Huang ◽

...

Keyword(s):

Inflammatory Responses ◽

Early Stage ◽

Survival Models ◽

Multivariate Survival ◽

Risk Of Progression ◽

Baseline Information ◽

Severe Group ◽

Progression Risk ◽

Improved Performance ◽

Mean Time

AbstractBackgroundManagement of high mortality risk due to significant progression requires prior assessment of time-to-progression. However, few related methods are available for COVID-19 pneumonia.MethodsWe retrospectively enrolled 338 adult patients admitted to one hospital between Jan 11, 2020 to Feb 29, 2020. The final follow-up date was March 8, 2020. We compared characteristics between patients with severe and non-severe outcome, and used multivariate survival analyses to assess the risk of progression to severe conditions.ResultsA total of 76 (31.9%) patients progressed to severe conditions and 3 (0.9%) died. The mean time from hospital admission to severity onset is 3.7 days. Age, body mass index (BMI), fever symptom on admission, co-existing hypertension or diabetes are associated with severe progression. Compared to non-severe group, the severe group already demonstrated, at an early stage, abnormalities in biomarkers indicating organ function, inflammatory responses, blood oxygen and coagulation function. The cohort is characterized with increasing cumulative incidences of severe progression up to 10 days after admission. Competing risks survival model incorporating CT imaging and baseline information showed an improved performance for predicting severity onset (mean time-dependent AUC = 0.880).ConclusionsMultiple predisposition factors can be utilized to assess the risk of progression to severe conditions at an early stage. Multivariate survival models can reasonably analyze the progression risk based on early-stage CT images that would otherwise be misjudged by artificial analysis.Funded by Sanming Project of Medicine in Shenzhen (SZSM201812058), China.

Download Full-text