Abstract
We consider interval-valued time series, that is, series resulting from collecting real intervals as an ordered sequence through time. Since the lower and upper bounds of the observed intervals at each time point are in fact values of the same variable, they are naturally related. We propose modeling interval time series with space-time autoregressive models and, based on the process appropriate for the interval bounds, we derive the model for the intervals' center and radius. A simulation study and an application with data of daily wind speed at different meteorological stations in Ireland illustrate that the proposed approach is appropriate and useful.

Abstract
Background/Aims: Intradialytic hypotension (IDH) is a serious and frequent complication of hemodialysis (HD). Thus far, data are scarcely available to assess the impact of first versus subsequent HD sessions of the week in IDH. Therefore, the purpose of this work was to evaluate IDH risk in patients on thrice-weekly HD. Methods: We conducted an analysis of all blood pressure (BP) measurements obtained during 492 HD treatments given to 41 prevalent adult patients over a one month period. A logistic regression model for repeated binary observations was used to determine the association between hypotension and patient and dialysis factors. Results: The incidence of IDH was 32.5%. First dialysis session of the week was associated with a 9% higher risk of hypotension relatively to the second one. The risk was even higher from the first to the third session of the week (60%) and from the second to the third (50%). A higher hypotension odds ratio was also associated with age (1.03, 90% CI: 1.01-1.06), higher predialysis BP (1.04, 90% CI: 1.03-1.05) and higher phosphorus level (1.38, 90% CI: 1.07-1.76). The risk decreased 24.4% for each additional antihypertensive drug taken by the patient. Conclusions: The odds of hypotension occurrence decrease throughout dialysis sessions of the week. Minimizing modifiable risk factors may decrease IDH episodes. (C) 2016 The Author(s) Published by S. Karger AG, Basel

Abstract
Scand J Caring Sci; 2013; 27; 468474 Evaluation and cross-cultural adaptation of the Hendrich II Fall Risk Model to Portuguese Background: Several tools for the assessment of the risk of falling are used commonly by clinical nurses, but none have been validated in Portuguese. Aims: To adapt and evaluate the Hendrich II Fall Risk Model (HIIFRM) for use with elderly Portuguese inpatients. Method: We conducted a prospective study of 586 older inpatients in acute care hospitals, from November 2007 to May 2010. Results: The study involved 270 men and 316 women. The most frequent risk factor on admission and at discharge was a score 3 on the Get Up and Go' test. The adapted HIIFRM showed a sensitivity of 93.2% and 75.7%, and a specificity of 35% and 46.7%, on admission and at discharge, respectively. A positive predictive value of 17.2% on admission and 17% at discharge and a negative predictive value of 97.3% and 93%, respectively, were estimated. Conclusions: The HIIFRM was shown to be a useful tool in predicting falls by patients. Nevertheless, the research model suggested that only four risk factors affected the occurrence of falls significantly on admission and two risk factors at discharge. Further research is required in Portuguese hospital settings.

Abstract

Abstract
In time series analysis, Autoregressive Moving Average (ARMA) models play a central role. Because of the importance of parameter estimation in ARMA modeling and since it is based on aggregate time series so often, we analyze the effect of temporal aggregation on estimation accuracy. We derive the relationships between the aggregate and the basic parameters and compute the actual values of the former from those of the latter in order to measure and compare their estimation accuracy. We run a simulation experiment that shows that aggregation seriously worsens estimation accuracy and that the impact increases with the order of aggregation.

Abstract
Symptomatic intradialytic hypotension (IDH) continues to be an important complication of hemodialysis treatment. There is some evidence that besides an IDH episode, repeated episodes could represent an even more important independent risk factor for mortality in hemodialysis patients. A retrospective cross-sectional study was performed to study 18 dialysis treatments in 43 patients during 6 weeks. Relationships of IDH episodes with baseline variables were examined using a Poisson regression model (generalized linear model). IDH was frequent (93% of patients) and highly variable by patient (0%-100%). Multivariate analysis showed that patients who experienced frequent hypotensive episodes had a lower dry weight (90% confidence interval [CI]: 0.95-0.99), higher phosphorus levels (90% CI: 1.07-1.47), greater prevalence of diabetes mellitus (90% CI: 1.11-2.71), and hypertension (90% CI: 1.04-2.45). Dry weight, hypertension, and phosphorus levels are modifiable risk factors to possibly reduce the rate of IDH episodes. The potential protective role of phosphorus warrants further investigation.