Context matters: using reinforcement learning to develop human-readable, state-dependent outbreak response policies.

The number of all possible epidemics of a given infectious disease that could occur on a given landscape is large for systems of real-world complexity. Furthermore, there is no guarantee that the control actions that are optimal, on average, over all possible epidemics are also best for each possible epidemic. Reinforcement learning (RL) and Monte Carlo control have been used to develop machine-readable context-dependent solutions for complex problems with many possible realizations ranging from video-games to the game of Go. RL could be a valuable tool to generate context-dependent policies for outbreak response, though translating the resulting policies into simple rules that can be read and interpreted by human decision-makers remains a challenge. Here we illustrate the application of RL to the development of context-dependent outbreak response policies to minimize outbreaks of foot-and-mouth disease. We show that control based on the resulting context-dependent policies, which adapt interventions to the specific outbreak, result in smaller outbreaks than static policies. We further illustrate two approaches for translating the complex machine-readable policies into simple heuristics that can be evaluated by human decision-makers. This article is part of the theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: epidemic forecasting and control'. This theme issue is linked with the earlier issue 'Modelling infectious disease outbreaks in humans, animals and plants: approaches and important themes'.

Body mass index and health-related quality of life.

OBJECTIVE: There are conflicting data regarding the association between body mass index (BMI) and health-related quality of life (HRQoL), especially among certain population subgroups and for mental and physical health domains. METHODS: This study analysed the relationship between BMI and HRQoL (Patient-Reported Outcomes Measurement Information System mental and physical health scales) using ordinary least squares regression. Each model allowed for the possibility of a non-linear relationship between BMI and the outcome, adjusting for age, gender, comorbidities, diet and physical activity. RESULTS: A total of 10,133 respondents were predominantly female (71.7%), White (84.1%), median age of 52.1 years (interquartile range 37.2-63.3) and median BMI of 27.9 (interquartile range 24.0-33.2). In adjusted models, BMI was significantly associated with physical and mental HRQoL (p < 0.001). For physical HRQoL, there was a significant interaction with age (p = 0.02). For mental HRQoL, there was a significant interaction with sex (p = 0.0004) but not age (p = 0.7). CONCLUSIONS: This study demonstrates a non-linear association of variable clinical relevance between BMI and HRQoL after adjusting for demographic factors and comorbidities. The relationship between BMI and HRQoL is nuanced and impacted by gender and age. These findings challenge the idea of obesity as a main driver of reduced HRQoL, particularly among women and with respect to mental HRQoL.