A unified approach to variable selection for Cox’s proportional hazards model with interval-censored failure time data

Cox’s proportional hazards model is the most commonly used model for regression analysis of failure time data and some methods have been developed for its variable selection under different situations. In this paper, we consider a general type of failure time data, case K interval-censored data, that include all of other types discussed as special cases, and propose a unified penalized variable selection procedure. In addition to its generality, another significant feature of the proposed approach is that unlike all of the existing variable selection methods for failure time data, the proposed approach allows dependent censoring, which can occur quite often and could lead to biased or misleading conclusions if not taken into account. For the implementation, a coordinate descent algorithm is developed and the oracle property of the proposed method is established. The numerical studies indicate that the proposed approach works well for practical situations and it is applied to a set of real data arising from Alzheimer’s Disease Neuroimaging Initiative study that motivated this study.

Risk Factors for 30-Day Mortality in Neonatal Gram-Negative Bacilli Sepsis

Objective Multidrug-resistant gram-negative bacilli (MDR-GNB) have emerged globally as a serious threat and with a high case fatality rate (CFR). Study Design We performed a case–control study in a Thai neonatal intensive care unit to identify the risk factors for 30-day CFR of GNB sepsis between 1991 and 2017. The CFR was analyzed by Cox's proportional hazards model. Results For 27 years, the percentage of MDR-GNB from GNB sepsis was 66% (169/257). The medians (interquartile ranges) of gestational age and birth weight of the neonates with GNB sepsis were 33 (29–38) weeks and 1,817 (1,100–2,800) grams, respectively. The 30-day CFRs of the neonates with MDR-GNB and non-MDR-GNB sepsis were 33% (56/169) and 20% (18/88), respectively, (hazard ratio [HR] = 1.74; 95% confidence interval [CI]: 1.03–2.97; p = 0.04). Using Cox's proportional hazards model, nonsurvivors in GNB sepsis were more likely to have septic shock (adjusted HR [aHR] = 6.67; 95% CI: 3.28–13.57; p < 0.001) or no microbiological cure (aHR = 10.65; 95% CI: 4.98–22.76; p < 0.001) than survivors. Conclusion Neonates suspected of sepsis with septic shock need broad-spectrum empirical antimicrobial therapy until the second successive negative culture, especially in high MDR areas.