A statistical model for hospital admissions caused by seasonal diseases

Moriña, David; Puig, Pedro; Ríos, José; Vilella, Anna; Trilla, Antoni

doi:10.1002/sim.4336

Cited by 53 publications

(47 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Nine studies predicted the timing of the epidemic peak or incidence at the peak; all performed validation using at least some forecasts made at least 4 weeks before the actual peak [10]–[13], [16]–[18], [29], [31]. The facility-based forecasting studies used 1-step-ahead [37]–[39] or n -step-ahead [40] predictions of visit counts over step sizes of 1 day [40] to 1 month [39]. The regional or global pandemic spread forecasting studies used early data from the 2009 influenza A(H1N1)pdm09 pandemic to predict outcomes at national level across countries, including pandemic arrival, and peak incidence and time of peak.…”

Section: Resultsmentioning

confidence: 99%

Influenza Forecasting in Human Populations: A Scoping Review

et al. 2014

View full text Add to dashboard Cite

Forecasts of influenza activity in human populations could help guide key preparedness tasks. We conducted a scoping review to characterize these methodological approaches and identify research gaps. Adapting the PRISMA methodology for systematic reviews, we searched PubMed, CINAHL, Project Euclid, and Cochrane Database of Systematic Reviews for publications in English since January 1, 2000 using the terms “influenza AND (forecast* OR predict*)”, excluding studies that did not validate forecasts against independent data or incorporate influenza-related surveillance data from the season or pandemic for which the forecasts were applied. We included 35 publications describing population-based (N = 27), medical facility-based (N = 4), and regional or global pandemic spread (N = 4) forecasts. They included areas of North America (N = 15), Europe (N = 14), and/or Asia-Pacific region (N = 4), or had global scope (N = 3). Forecasting models were statistical (N = 18) or epidemiological (N = 17). Five studies used data assimilation methods to update forecasts with new surveillance data. Models used virological (N = 14), syndromic (N = 13), meteorological (N = 6), internet search query (N = 4), and/or other surveillance data as inputs. Forecasting outcomes and validation metrics varied widely. Two studies compared distinct modeling approaches using common data, 2 assessed model calibration, and 1 systematically incorporated expert input. Of the 17 studies using epidemiological models, 8 included sensitivity analysis. This review suggests need for use of good practices in influenza forecasting (e.g., sensitivity analysis); direct comparisons of diverse approaches; assessment of model calibration; integration of subjective expert input; operational research in pilot, real-world applications; and improved mutual understanding among modelers and public health officials.

show abstract

Section: Resultsmentioning

confidence: 99%

Influenza Forecasting in Human Populations: A Scoping Review

et al. 2014

View full text Add to dashboard Cite

show abstract

“…Integer-valued time series arise in contexts such as modelling monthly traffic fatalities (Neal and Subba Rao 2007) or the number of patients in a hospital at a sequence of time points (Moriña et al 2011). Consider the following integer-valued autoregressive model of order p , known as INAR( p ): where Z t for t >1 are independent and identically distributed integer-valued random variables with , with the Z t assumed to be independent of the X t .…”

Section: Examplesmentioning

confidence: 99%

Piecewise Approximate Bayesian Computation: fast inference for discretely observed Markov models using a factorised posterior distribution

2013

View full text Add to dashboard Cite

Many modern statistical applications involve inference for complicated stochastic models for which the likelihood function is difficult or even impossible to calculate, and hence conventional likelihood-based inferential techniques cannot be used. In such settings, Bayesian inference can be performed using Approximate Bayesian Computation (ABC). However, in spite of many recent developments to ABC methodology, in many applications the computational cost of ABC necessitates the choice of summary statistics and tolerances that can potentially severely bias the estimate of the posterior.We propose a new “piecewise” ABC approach suitable for discretely observed Markov models that involves writing the posterior density of the parameters as a product of factors, each a function of only a subset of the data, and then using ABC within each factor. The approach has the advantage of side-stepping the need to choose a summary statistic and it enables a stringent tolerance to be set, making the posterior “less approximate”. We investigate two methods for estimating the posterior density based on ABC samples for each of the factors: the first is to use a Gaussian approximation for each factor, and the second is to use a kernel density estimate. Both methods have their merits. The Gaussian approximation is simple, fast, and probably adequate for many applications. On the other hand, using instead a kernel density estimate has the benefit of consistently estimating the true piecewise ABC posterior as the number of ABC samples tends to infinity. We illustrate the piecewise ABC approach with four examples; in each case, the approach offers fast and accurate inference.

show abstract

“…After non-adaptive regression, autoregressive models are the second most popular option [17,4,2,20,18,1,56,44,55,68,69,50,74,28]. This type of model is commonly used in the Box-Jenkins methodology [7] and allows the use of the data correlation structure.…”

Section: Methodsmentioning

confidence: 99%

“…However, ARIMA requires a large number of observations and certain assumptions about the time series. Hence, alternative autoregressive approaches are also used [50,74,28]. For instance, integer autoregressive models yield good results for a short morbidity series [50].…”

Section: Methodsmentioning

confidence: 99%

Long-Term Forecasting of Influenza-Like Illnesses in Russia

Kondratyev

Цыбалова²

2013

Int. J. of Pure and Appl. Math.

View full text Add to dashboard Cite

This paper compares the feasible methods for the long-term forecasting of the incidence rates of influenza-like illnesses (ILI) and acute respiratory infections (ARI), which is important for strategic management. A literature survey shows that the most appropriate techniques for long-term ILI & ARI morbidity projections are the following well-known statistical methods: simple averaging of observations, point-to-point linear estimates, Serfling-type regression models, autoregressive models such as autoregressive integrated moving average (ARIMA) models, and generalized exponential smoothing using the Holt-Winters approach. Using these methods and official data on the total number of ILI & ARI cases per week in 2000-2012 in Moscow, St. Petersburg, Novosibirsk, Yekaterinburg, Nizhny Novgorod and Yakutsk, we developed oneyear projections and evaluated their accuracy. Different methods yielded the best results, depending on the time series. Generally, it is preferable to use the Serfling model. The Serfling model forecasts almost matched the pointto-point linear estimates. In certain cases, ARIMA outperformed the Serfling model. Simple averaging can ensure a fairly good prediction when the ILI & ARI time series do not exhibit a trend. The results of exponential smoothing were poorer than those of other techniques.

show abstract

A statistical model for hospital admissions caused by seasonal diseases

Cited by 53 publications

References 20 publications

Influenza Forecasting in Human Populations: A Scoping Review

Influenza Forecasting in Human Populations: A Scoping Review

Piecewise Approximate Bayesian Computation: fast inference for discretely observed Markov models using a factorised posterior distribution

Long-Term Forecasting of Influenza-Like Illnesses in Russia

Contact Info

Product

Resources

About