Abstract
Objective: to identify diet quality time trends in Portuguese elderly households and the association of diet quality with sociodemographic characteristics. Design: Set of four cross-sectional studies. Setting: Portuguese population. Subjects: nationally representative samples from Household Budget Surveys (1989-1990, 1994-1995, 2000-2001 and 2005-2006 versions). Households with members aged >= 65 years were categorized as solitary elderly female, solitary elderly male, or elderly couple (composed of one elderly female and one elderly male) and compared with adult households with the same composition. Methods: Diet quality was assessed through a revised version of the Healthy Diet indicator (HDIr). Univariate and multiple backward linear regression models were used to study the association with sociodemographic characteristics. Results: Mean values of dietary index were low and the proportion of elderly households with low diet quality (HDIr <= 4) was high (between 47.4% and 68.4%). However, the frequency of HDIr inadequacy for adult household was even higher (P<0.05). in general, adjusted coefficients for survey year, educational level of the household head, and eating out expenses were inversely associated with HDIr; whilst semi-urban and rural location of the households predicted higher HDIr values. exceptions were found in elderly male households where the educational level of the household head was positively associated with HDIr values. Conclusions: overall, diet quality was low and decreased over time but lonely elderly female and elderly couple households had higher values of HDIr. adherence to a healthier diet was associated with lower educational level of the household head and location of the household in less urbanized areas.

Abstract
To analyze the simultaneous effects of sociodemographic variables and time on each food group contribution to total Portuguese elderly household food availability. Four cross sectional Portuguese Household Budget Surveys were used. A multivariate analysis of variance (MANOVA), using a general linear model (GLM), was applied to analyze the simultaneous effects of sociodemographic variables and time. Portuguese population. Nationally representative samples of households with members aged a parts per thousand yen 65 years were selected and categorized as solitary elderly female, solitary elderly male, or couple (one elderly female and one elderly male). Samples included 1,967 households in 1989-1990, 2,219 households in 1994-1995, 2,533 households in 2000-2001 and 2,441 households in 2005-2006. The simultaneous effects of sociodemographic variables and time were significant for all food groups (P < 0.001). The highest contribution for the total household food availability was found for cereals, potatoes, alcoholic beverages, nonalcoholic beverages and fruits. The effects were large for "household food availability" and medium for "elderly household type", "urbanization degree", "income", "food expenses" and "eating out expenses". Solitary elderly male households had the highest proportion of cereals and alcoholic beverages, whilst solitary elderly female households had higher availability of milk/milk products and fruits. Households located in urban areas had higher contribution of milk/milk products while rural, had higher contribution of potatoes. The simultaneous effect of the studied variables on food group contribution to total household food availability can be considered when addressing dietary recommendation for providing an insight into the motivations associated with food purchases.

Abstract
[abstract]

Abstract
[abstract]

Abstract
[abstract]

Abstract
Purpose: The paper aims to analyse the engagement in sustainability reporting assurance (SRA) by a sample of Portuguese firms between 2008 and 2011. Design/methodology/approach: Bivariate and multivariate non-parametric statistics is used to analyse some factors that influence the decision to have sustainability reports assured. Findings: Results indicate that size, leverage, profitability, listing status and industrial affiliation are determinants of SRA, whereas type of ownership is not. A downward trend in sustainability reporting and its assurance was also detected. Research limitations/implications: The sample is small. Originality/value: It adds to the scarce research on SRA by providing new empirical data in a context of crisis and extends prior research by analysing the effects of listing status and type of ownership. © Emerald Group Publishing Limited.