In modern science, computer models are often used to understand complex phenomena and a thriving statistical community has grown around analyzing them. This review aims to bring a spotlight to the growing prevalence of stochastic computer models—providing a catalogue of statistical methods for practitioners, an introductory view for statisticians (whether familiar with deterministic computer models or not), and an emphasis on open questions of relevance to practitioners and statisticians. Gaussian process surrogate models take center stage in this review, and these, along with several extensions needed for stochastic settings, are explained. The basic issues of designing a stochastic computer experiment and calibrating a stochastic computer model are prominent in the discussion. Instructive examples, with data and code, are used to describe the implementation of, and results from, various methods.
Healthcare workers (HCWs) are at the forefront of fighting against the COVID-19 pandemic. However, they are at high risk of acquiring the pathogen from infected patients and transmitting to other HCWs. We aimed to investigate risk factors for nosocomial COVID-19 infection among HCWs in a non-COVID-19 hospital yard. Retrospective data collection on demographics, lifestyles, contact status with infected subjects for 118 HCWs (including 12 COVID-19 HCWs) at Union Hospital of Wuhan, China. Sleep quality and working pressure were evaluated by the Pittsburgh Sleep Quality Index (PSQI) and The Nurse Stress Index (NSI), respectively. The follow-up duration was from Dec 25, 2019, to Feb 15, 2020. A high proportion of COVID-19 HCWs had engaged in night shift-work (75.0% vs. 40.6%) and felt working under pressure (66.7% vs. 32.1%) than uninfected HCWs. SARS-CoV-2 infected HCWs had significantly higher scores of PSQI and NSI than uninfected HCWs (P < 0.001). Specifically, scores of 5 factors (sleep quality, sleep time, sleep efficiency, sleep disorder, and daytime dysfunction) in PSQI were higher among infected HCWs. For NSI, its 5 subscales (nursing profession and work, workload and time allocation, working environment and resources, patient care, management and interpersonal relations) were all higher in infected than uninfected nurse. Furthermore, total scores of PSQI (HR = 2.97, 95%CI = 1.86−4.76; P <0.001) and NSI (HR = 4.67, 95%CI = 1.42−15.45; P = 0.011) were both positively associated with the risk of SARS-CoV-2 infection. Our analysis shows that poor sleep quality and higher working pressure may increase the risk of nosocomial SARS-CoV-2 infection among HCWs.
Observing system uncertainty experiments (OSUEs) have been recently proposed as a cost-effective way to perform probabilistic assessment of retrievals for NASA’s Orbiting Carbon Observatory-2 (OCO-2) mission. One important component in the OCO-2 retrieval algorithm is a full-physics forward model that describes the mathematical relationship between atmospheric variables such as carbon dioxide and radiances measured by the remote sensing instrument. This complex forward model is computationally expensive but large-scale OSUEs require evaluation of this model numerous times, which makes it infeasible for comprehensive experiments. To tackle this issue, we develop a statistical emulator to facilitate large-scale OSUEs in the OCO-2 mission with independent emulation. Within each distinct spectral band, the emulator represents radiances output at irregular wavelengths via a linear combination of basis functions and random coefficients. These random coefficients are then modeled with nearest-neighbor Gaussian processes with built-in input dimension reduction via active subspace. The proposed emulator reduces dimensionality in both input space and output space, so that fast computation is achieved within a fully Bayesian inference framework. Validation experiments demonstrate that this emulator outperforms other competing statistical methods and a reduced order model that approximates the full-physics forward model. Supplementary materials for this article are available online.
Development of strategies for mitigating the severity of COVID-19 is now a top public health priority. We sought to assess strategies for mitigating the COVID-19 outbreak in a hospital setting via the use of non-pharmaceutical interventions. We developed an individual-based model for COVID-19 transmission in a hospital setting. We calibrated the model using data of a COVID-19 outbreak in a hospital unit in Wuhan. The calibrated model was used to simulate different intervention scenarios and estimate the impact of different interventions on outbreak size and workday loss. The use of high-efficacy facial masks was shown to be able to reduce infection cases and workday loss by 80% (90% credible interval (CrI): 73.1–85.7%) and 87% (CrI: 80.0–92.5%), respectively. The use of social distancing alone, through reduced contacts between healthcare workers, had a marginal impact on the outbreak. Our results also indicated that a quarantine policy should be coupled with other interventions to achieve its effect. The effectiveness of all these interventions was shown to increase with their early implementation. Our analysis shows that a COVID-19 outbreak in a hospital's non-COVID-19 unit can be controlled or mitigated by the use of existing non-pharmaceutical measures.
1 August 2021 in Journal of Computational and Applied Mathematics