Evaluating the ROC performance of markers for future events

Receiver operating characteristic (ROC) curves play a central role in the evaluation of biomarkers and tests for disease diagnosis. Predictors for event time outcomes can also be evaluated with ROC curves, but the time lag between marker measurement and event time must be acknowledged. We discuss different definitions of time-dependent ROC curves in the context of real applications. Several approaches have been proposed for estimation. We contrast retrospective versus prospective methods in regards to assumptions and flexibility, including their capacities to incorporate censored data, competing risks and different sampling schemes. Applications to two datasets are presented.     

Atmospheric Dispersion of Inflammable Substances for Estimating Vulnerable Zones in Hydrocarbon Industry

The present study utilizes an operational model as well as simple empirical relationships for estimating hazard zones due to fire, explosion, and toxic vapor cloud dispersion. The empirical relationships are based on giving appropriate weightage to each of the parameters on which the hazard in question (viz, fire, explosion, toxic vapour dispersion) is dependent. Results from these two approaches [i.e., an operational model FLAMCALC of U.K. Health and Safety Executive (HSE) and an empirical model named FIREX] have been compared with the data obtained from the Mexico City disaster in 1984. In general, results from the empirical approach and FLAMCALC are comparable to the observed effects.     

Estimation of Vulnerable Zones Due to Accidental Release of Toxic Materials Resulting in Dense Gas Clouds

Heavy gas dispersion models have been developed at IIT (hereinafter referred as IIT heavy gas models I and II) with a view to estimate vulnerable zones due to accidental (both instantaneous and continuous, respectively) release of dense toxic material in the atmosphere. The results obtained from IIT heavy gas models have been compared with those obtained from the DEGADIS model [Dense Gas Dispersion Model, developed by Havens and Spicer (1985) for the U.S. Coast Guard] as well as with the observed data collected during the Burro Series, Maplin Sands, and Thorney Island field trials. Both of these models include relevant features of dense gas dispersion, viz., gravity slumping, air entrainment, cloud heating, and transition to the passive phase, etc. The DEGADIS model has been considered for comparing the performance of IIT heavy gas models in this study because it incorporates most of the physical processes of dense gas dispersion in an elaborate manner, and has also been satisfactorily tested against field observations. The predictions from IIT heavy gas models indicate a fairly similar trend to the observed values from Thorney Island, Burro Series, and Maplin experiments with a tendency toward overprediction. There is a good agreement between the prediction of IIT Heavy Gas models I and II with those from DEGADIS, except for the simulations of IIT heavy gas model-I pertaining to very large release quantities under highly stable atmospheric conditions. In summary, the performance of IIT heavy gas models have been found to be reasonably good both with respect to the limited field data available and various simulations (selected on the basis of relevant storages in the industries and prevalent meteorological conditions performed with DEGADIS). However, there is a scope of improvement in the IIT heavy gas models (viz., better formulation for entrainment, modification of coefficients, transition criteria, etc.). Further, isotons (nomograms) have been prepared by using IIT heavy gas models for chlorine, which provide safe distance for various storage amounts for 24 meteorological scenarios prevalent in the entire year. These nomograms are prepared such that a nonspecialist can use them easily for control and management in case of an emergency requiring the evacuation of people in the affected region. These results can also be useful for siting and limiting the storage quantities.     

It Ain't Just for the 40 Dollars: Young Fathers' Responses to the Interview Experience

In-depth interviews were conducted with 13 young fathers to identify factors supporting their paternal involvement. Participants' responses to their interviews revealed that the qualitative experience was meaningful and encouraged their paternal involvement through (1) self-disclosure, (2) the opportunity to help others, and (3) personal engagement with the interviewer. Prior work has noted the general benefits of interviews for interviewees. This study addresses the less-explored, specific benefits of qualitative interviews for young fathers, along with recommendations, which include (1) conducting research that utilizes the qualitative interview as a therapeutic opportunity for young-father participants, (2) exploring fathers' behaviors in qualitative interviews as an untapped resource for engagement, and (3) conducting participatory action and mixed methodologies with young fathers.