Research and analysis

Economic theory often provides shape restrictions on functions of interest in applications, such as monotonicity, convexity, non-increasing (non-decreasing) returns to scale, or the Slutsky inequality of consumer theory; but economic theory does not provide finite-dimensional parametric models. This motivates nonparametric estimation under shape restrictions. Nonparametric estimates are often very noisy. Shape restrictions stabilize nonparametric estimates without imposing arbitrary restrictions, such as additivity or a single-index structure, that may be inconsistent with economic theory and the data. This paper explains how to estimate and obtain an asymptotic uniform confidence band for a conditional mean function under possibly nonlinear shape restrictions, such as the Slutsky inequality. The results of Monte Carlo experiments illustrate the finite-sample performance of the method, and an empirical example illustrates its use in an application.

Older adults' negative beliefs about ageing are related to their health and functioning, but little is known about how perceptions of ageing (POA) relate to frailty status. This study aimed to explore the relationship between POA and frailty. Secondary analysis of data used were from the English Longitudinal Study of Ageing Waves 2 and 5. A POA score was based on participants' responses to 12 statements using a five-point Likert scale at baseline, and a Frailty Index (FI) score was calculated for each participant for both waves. Multiple linear regression models were conducted to assess the relationship between POA and frailty cross-sectionally and longitudinally in models controlled for age, gender, depression symptoms, and socioeconomic status. Older adults with more negative POA had greater frailty (β = .12, p < .001). Negative POA predicted greater frailty 6 years later (β = .03, p < .05). Future work regarding the mechanisms of this relationship is needed to identify ways of intervening to improve health.

Repeated exposure to systemic inflammation and risk of new depressive symptoms among older adults

OBJECTIVES: To determine whether hearing loss is associated with incident physician-diagnosed dementia in a representative sample.

BACKGROUND: Frailty is a common syndrome in older adults characterised by increased vulnerability to adverse health outcomes as a result of decline in functional and physiological measures. Frailty predicts a range of poor health and social outcomes and is associated with increased risk of hospital admission. The health benefits of sport and physical activity and the health risks of inactivity are well known. However, less is known about the role of sports clubs and physical activity in preventing and managing frailty in older adults. The objective of this study is to examine the role of membership of sports clubs in promoting physical activity and reducing levels of frailty in older adults.

Pedestrian crossings in the UK and US require people to walk at 1.2 m/s to cross the road in time; however a large proportion of older people do not walk this fast, potentially discouraging walking or putting older people at risk of injury. We use longitudinal data to investigate changes in walking speed, and ability to cross the road in time, at older ages. 31,015 walking speed measurements were taken from 10,249 men and women aged 60+ years in waves 1-7 of the English Longitudinal Study of Ageing (2002-2014). Growth curve analyses were used to model how walking speed changes with increasing age, and predicted probabilities of being able to cross the road in time were estimated. 10% of measured walking speeds were fast enough to cross the road in time. Walking speed declined with age (-5.7×10-3m/s/yr (95% CI -7.6×10-3, -3.9×10-3)), and the decline accelerated with increasing age (-0.3 ×10-3m/s/yr (-0.4 ×10-3, -0.3 ×10-3)). Female, less wealthy and less healthy older people had slower walking speeds. For instance, predicted probability of crossing the road in time at age 60 was 14.8% (10.1, 18.5) and 2.7% (1.5, 3.8) for the richest and poorest men and 8.4% (6.0, 1.1) and 1.5% (0.9, 2.2) for the richest and poorest women, and at age 80 they were 7.1% (3.6, 10.5) and 1.0% (0.3, 1.7) for the richest and poorest men and 3.7% (1.6, 5.9) and 0.5% (0.1, 0.9) for the richest and poorest women. Most older people do not walk fast enough to cross the road in time. Even the majority of the wealthiest and healthiest people aged 60 years and older do not walk fast enough to cross pedestrian crossings in the allocated time. Crossing times should be increased to allow for older peoples' slower walking speeds or other policies considered to improve walkability, and to help avoid injuries and social isolation.

AIMS: To validate the Leicester Self-Assessment score using a representative English dataset for detecting prevalent non-diabetic hyperglycaemia or undiagnosed Type 2 diabetes (defined as HbA1c ≥6.0%) and for identifying those who may go on to develop Type 2 diabetes within 10 years. METHODS: Data were taken from the English Longitudinal Study of Ageing, a nationally representative dataset of people aged ≥50 years. The area under the receiver-operator curve and performance metrics for the score at the recommended score threshold (≥16), were calculated for the outcomes of HbA1c ≥42 mmol/mol (6.0%) at baseline and self-reported Type 2 diabetes within 10 years in those aged 50-75 years at baseline. RESULTS: A total of 3203 individuals had a baseline HbA1c measurement, of whom 247 (7.7%) had an HbA1c concentration ≥ 42 mmol/mol (6.0%). The area under the receiver-operator curve was 69.4% (95% CI 66.0-72.9) for baseline HbA1c ≥42 mmol/mol. A total of 3550 individuals had diabetes status recorded at 10 years, of whom 324 (9.1%) were diagnosed with Type 2 diabetes within this time; the area under the receiver-operator curve for this outcome was 74.9% (95% CI 72.4-77.5). The score threshold of ≥16 had a sensitivity of 89.2% (95% CI 85.3-92.4) and a specificity of 42.3% (95% CI 40.5-44.0) for Type 2 diabetes within 10 years. CONCLUSIONS: The Leicester Self-Assessment score is validated for use across England to identify people with non-diabetic hyperglycaemia or undiagnosed Type 2 diabetes. Those with a high score are at high risk of developing diabetes in the future. This article is protected by copyright. All rights reserved.

There is substantial evidence of a significant relationship between parents’ income and sons’ earnings in the UK, and that this relationship has strengthened over time. We extend this by exploring a broader measure of net family income as an outcome.

This paper provides a framework for identifying preferences in a large network where links are pairwise stable.

Bayesian deconvolution: an R vinaigrette

Quantile regression 40 years on

Rebayes: an R package for empirical bayes mixture methods

Testing for homogeneity in mixture models

An informational intervention in Western Kenya

Cancer and heart attack survivors’ expectations of employment status: results from the English Longitudinal Study of Ageing

BACKGROUND: Depression independently increases the risk of falls in older people, but the mechanism for this relationship, as well as the specific falls type involved, remains unclear. Accidental falls (AFs) are due to slips or trips, while the cause of unexplained falls (UFs) is not immediately apparent and can include unrecognised syncope.

This paper studies inference in randomized controlled trials with covariate-adaptive randomization when there are multiple treatments. More specifically, we study in this setting inference about the average effect of one or more treatments relative to other treatments or a control. As in Bugni et al. (2017), covariate-adaptive randomization refers to randomization schemes that first stratify according to baseline covariates and then assign treatment status so as to achieve "balance" within each stratum. In contrast to Bugni et al. (2017), however, we allow for the proportion of units being assigned to each of the treatments to vary across strata.