Modelling Frontier Mortality Using Bayesian Generalised Additive Models

Mortality rates differ across countries and years, and the country with the lowest observed mortality has changed over time. However, the classic Science paper by Oeppen and Vaupel (2002) identified a persistent linear trend over time in maximum national life expectancy. In this article, we look to exploit similar regularities in age-specific mortality by considering for any given year a hypothetical mortality ‘frontier’, which we define as the lower limit of the force of mortality at each age across all countries. Change in this frontier reflects incremental advances across the wide range of social, institutional and scientific dimensions that influence mortality. We jointly estimate frontier mortality as well as mortality rates for individual countries. Generalised additive models are used to estimate a smooth set of baseline frontier mortality rates and mortality improvements, and country-level mortality is modelled as a set of smooth, positive deviations from this, forcing the mortality estimates for individual countries to lie above the frontier. This model is fitted to data for a selection of countries from the Human Mortality Database (2019). The efficacy of the model in forecasting over a ten-year horizon is compared to a similar model fitted to each country separately.

Optimising predictive modelling of Ross River virus using meteorological variables

Background Statistical models are regularly used in the forecasting and surveillance of infectious diseases to guide public health. Variable selection assists in determining factors associated with disease transmission, however, often overlooked in this process is the evaluation and suitability of the statistical model used in forecasting disease transmission and outbreaks. Here we aim to evaluate several modelling methods to optimise predictive modelling of Ross River virus (RRV) disease notifications and outbreaks in epidemiological important regions of Victoria and Western Australia. Methodology/Principal findings We developed several statistical methods using meteorological and RRV surveillance data from July 2000 until June 2018 in Victoria and from July 1991 until June 2018 in Western Australia. Models were developed for 11 Local Government Areas (LGAs) in Victoria and seven LGAs in Western Australia. We found generalised additive models and generalised boosted regression models, and generalised additive models and negative binomial models to be the best fit models when predicting RRV outbreaks and notifications, respectively. No association was found with a models’ ability to predict RRV notifications in LGAs with greater RRV activity, or for outbreak predictions to have a higher accuracy in LGAs with greater RRV notifications. Moreover, we assessed the use of factor analysis to generate independent variables used in predictive modelling. However, this approach did not result in as many best fit models than when not using this approach. Conclusions/Significance We demonstrate that models which are developed and used for predicting disease notifications may not be suitable for predicting disease outbreaks, or visa versa. Furthermore, poor predictive performance in modelling disease transmissions may be the result of inappropriate model selection methods. Our findings provide approaches and methods to facilitate the selection of the best fit statistical model for predicting mosquito-borne disease notifications and outbreaks used for disease surveillance.

Trend shifts in road traffic collisions: An application of Hidden Markov Models and Generalised Additive Models to assess the impact of the 20 mph speed limit policy in Edinburgh

Empirical study of road traffic collision (RTCs) rates is challenging at small geographies due to the relative rarity of collisions and the need to account for secular and seasonal trends. In this paper, we demonstrate the successful application of Hidden Markov Models (HMMs) and Generalised Additive Models (GAMs) to describe RTCs time series using monthly data from the city of Edinburgh (STATS19) as a case study. While both models have comparable level of complexity, they bring different advantages. HMMs provide a better interpretation of the data-generating process, whereas GAMs can be superior in terms of forecasting. In our study, both models successfully capture the declining trend and the seasonal pattern with a peak in the autumn and a dip in the spring months. Our best fitting HMM indicates a change in a fast-declining-trend state after the introduction of the 20 mph speed limit in July 2016. Our preferred GAM explicitly models this intervention and provides evidence for a significant further decline in the RTCs. In a comparison between the two modelling approaches, the GAM outperforms the HMM in out-of-sample forecasting of the RTCs for 2018. The application of HMMs and GAMs to routinely collected data such as the road traffic data may be beneficial to evaluations of interventions and policies, especially natural experiments, that seek to impact traffic collision rates.

Data presented by the UK government as lockdown was eased shows the transmission of COVID-19 had already increased

BackgroundCoronavirus disease 2019 (COVID-19) is an international emergency that has been addressed in many countries by changes in and restrictions on behaviour. These are often collectively labelled social distancing and lockdown. On the 23rd June 2020, Boris Johnson, the Prime Minister of the United Kingdom announced substantial easings of restrictions. This paper examines some of the data he presented.MethodsGeneralised additive models, with negative binomial errors and cyclic term representing day-of-week effects, were fitted to data on the daily numbers of new confirmed cases of COVID-19. Exponential rates for the epidemic were estimated for different periods, and then used to calculate R, the reproduction number, for the disease in different periods.ResultsAfter an initial stabilisation, the lockdown reduced R to around 0.81 (95% CI: 0.79, 0.82). This value increased to around 0.94 (95% CI 0.89, 0.996) for the fortnight from the 9th June 2020.ConclusionsOfficial UK data, presented as the easing of the lockdown was announced, shows that R was already more than half way back to 1 at that point. That suggests there was little scope for the announced changes to be implemented without restarting the spread of the disease.