Cross-classification between self-rated health and health status: longitudinal analyses of all-cause mortality and leading causes of death in the UK

Risk stratification is an important public health priority that is central to clinical decision making and resource allocation. The aim of this study was to examine how different combinations of self-rated and objective health status predict all-cause mortality and leading causes of death in the UK. The UK Biobank study recruited > 500,000 participants between 2006 and 2010. Self-rated health was assessed using a single-item question and health status was derived from medical history, including data on 81 cancer and 443 non-cancer illnesses. Analyses included > 370,000 middle-aged and older adults with a median follow-up of 11.75 (IQR = 1.4) years, yielding 4,320,270 person-years of follow-up. Compared to individuals with excellent self-rated health and favourable health status, individuals with other combinations of self-rated and objective health status had a greater mortality risk, with hazard ratios ranging from HR = 1.22 (95% CI 1.15–1.29, PBonf. < 0.001) for individuals with good self-rated health and favourable health status to HR = 7.14 (95% CI 6.70–7.60, PBonf. < 0.001) for individuals with poor self-rated health and unfavourable health status. Our findings highlight that self-rated health captures additional health-related information and should be more widely assessed. The cross-classification between self-rated health and health status represents a straightforward metric for risk stratification, with applications to population health, clinical decision making and resource allocation.

Accuracy of Non-Exercise Estimated Cardiorespiratory Fitness in Japanese Adults

Cardiorespiratory fitness (CRF) is an independent predictor of morbidity and mortality. In Japan, annual physical exams are mandatory in workplace settings, and most healthcare settings have electronic medical records (EMRs). However, in both settings, CRF is not usually determined, thereby limiting the potential for epidemiological investigations using EMR data. PURPOSE: To estimate CRF (mL/kg/min) using variables commonly recorded in EMRs. METHODS: Participants were 5293 Japanese adults (11.7% women) who completed an annual physical exam at a large gas company in Tokyo, Japan, in 2004. The mean age was 48.3 ± 8.0 years. Estimated CRF (eCRF) was based on age, measured body mass index, resting heart rate, systolic and diastolic blood pressure, and smoking. Measured CRF was determined by a submaximal cycle ergometer graded exercise test. RESULTS: Regression models were used for males and females to calculate Pearson’s correlation and regression coefficients. Cross-classification of measured CRF and eCRF was conducted using the lowest quintile, quartile, and tertile as the unfit categories. R’s for eCRF were 0.61 (MD 4.41) for men and 0.64 (MD 4.22) for women. The overall accuracy level was reasonable and consistent across models, yet the unfit lower tertile model provided the best overall model when considering the positive predictive value and sensitivity. CONCLUSION: eCRF may provide a useful method for conducting investigations using data derived from EMRs or datasets devoid of CRF or physical activity measures.

ei.Datasets: Real Data Sets for Assessing Ecological Inference Algorithms

Ecological inference models aim to infer individual-level relationships using aggregate data. They are routinely used to estimate voter transitions between elections, disclose split-ticket voting behaviors, or infer racial voting patterns in U.S. elections. A large number of procedures have been proposed in the literature to solve these problems; therefore, an assessment and comparison of them are overdue. The secret ballot however makes this a difficult endeavor since real individual data are usually not accessible. The most recent work on ecological inference has assessed methods using a very small number of data sets with ground truth, combined with artificial, simulated data. This article dramatically increases the number of real instances by presenting a unique database (available in the R package ei.Datasets) composed of data from more than 550 elections where the true inner-cell values of the global cross-classification tables are known. The article describes how the data sets are organized, details the data curation and data wrangling processes performed, and analyses the main features characterizing the different data sets.

SUMS OF POLYNOMIAL-TYPE EXCEPTIONAL UNITS MODULO

Let $f(x)\in \mathbb {Z}[x]$ be a nonconstant polynomial. Let $n\ge 1, k\ge 2$ and c be integers. An integer a is called an f-exunit in the ring $\mathbb {Z}_n$ of residue classes modulo n if $\gcd (f(a),n)=1$ . We use the principle of cross-classification to derive an explicit formula for the number ${\mathcal N}_{k,f,c}(n)$ of solutions $(x_1,\ldots ,x_k)$ of the congruence $x_1+\cdots +x_k\equiv c\pmod n$ with all $x_i$ being f-exunits in the ring $\mathbb {Z}_n$ . This extends a recent result of Anand et al. [‘On a question of f-exunits in $\mathbb {Z}/{n\mathbb {Z}}$ ’, Arch. Math. (Basel)116 (2021), 403–409]. We derive a more explicit formula for ${\mathcal N}_{k,f,c}(n)$ when $f(x)$ is linear or quadratic.

MVPAlab: A Machine Learning decoding toolbox for multidimensional electroencephalography data

MVPAlab is a MATLAB-based and very flexible decoding toolbox for multidimensional electroencephalography and mag-netoencephalography data. The MVPAlab Toolbox implements several machine learning algorithms to compute multivari-ate pattern analyses, cross-classification, temporal generalization matrices and feature and frequency contribution anal-yses. It also provides access to an extensive set of preprocessing routines for, among others, data normalization, data smoothing, dimensionality reduction and supertrials generation. To draw statistical inferences at the group level, MVPAlab includes a non-parametric cluster-based permutation approach. This toolbox has been designed to include an easy-to-use and very intuitive graphic user interface and data representation software, which makes MVPAlab a very convenient tool for those users with few or no previous coding experience. However, MVPAlab is not for beginners only, as it implements several high and low-level routines allowing more experienced users to design their own projects in a highly flexible manner.

Development and external validation of the “Flower-FFQ”: a food frequency questionnaire designed for the Lifelines Cohort Study

Objective: Food frequency questionnaires (FFQs) assess habitual dietary intake and are relatively inexpensive to process, but may take up to 60 minutes to complete. This article describes the validation of the Flower-FFQ, which consists of four short FFQs measuring the intake of energy and macronutrients or specific (micro)nutrients/foods that can be merged into one complete daily assessment using predefined algorithms. Design: Participants completed the Flower-FFQ and validated regular-FFQ (n=401). Urinary nitrogen (n=242) and potassium excretions (n=361) were measured. We evaluated: 1) group-level bias, 2) correlations, and 3) cross-classification. Setting: Observational study. Participants: Dutch adults, 54±11(mean±SD) years. Results: Flower-FFQ1, Flower-FFQ2, Flower-FFQ3, and Flower-FFQ4 were completed in ±24, 9, 8 and 9 minutes (±50 minutes total), respectively. The regular-FFQ was completed in ±43 minutes. Mean energy (flower vs. regular: 7953 vs. 8718 kJ/day) and macronutrient intakes (carbohydrates: 204 vs. 222 g/day; protein: 75 vs. 76 g/day; fat: 74 vs. 83 g/day; ethanol: 8 vs. 12 g/day) were comparatively similar. Spearman correlations between Flower-FFQ and regular-FFQ ranged from 0.60-0.80 for macronutrients and from 0.40-0.80 for micronutrients and foods. For all micronutrients and foods, ≥78% of the participants classified in the same/adjacent quartile. The flower-FFQ underestimated urinary nitrogen and potassium excretions by 24% and 18%; 75% and 73% of the participants ranked in the same/adjacent quartile. Conclusion: Completing the Flower-FFQ required 50 minutes with a maximum of 25 minutes per short FFQ. The Flower-FFQ has a moderate to good ranking ability for most nutrients and foods and performs sufficiently to study diet-disease associations.

A Curious Identity on Multiple Sums over Fields with Applications

Let [Formula: see text] be a field, and let [Formula: see text] be integers such that [Formula: see text] and [Formula: see text]. We show that for any subset [Formula: see text], the curious identity [Formula: see text] holds with [Formula: see text] being the set of nonnegative integers. As an application, we prove that for any subset [Formula: see text] with [Formula: see text] being the finite field of [Formula: see text] elements and [Formula: see text] being integers such that [Formula: see text] and [Formula: see text], [Formula: see text] Using this identity and providing an extension of the principle of cross-classification that slightly generalizes the one obtained by Hong in 1996, we show that if [Formula: see text] is an integer with [Formula: see text], then for any subset [Formula: see text] we have [Formula: see text] This implies [Formula: see text].

Cross-classification between self-rated health and health status: longitudinal analyses of all-cause mortality and leading causes of death in the UK

BackgroundRisk stratification is an important public health priority that is central to clinical decision making and resource allocation. The aim of the present study was to examine how different combinations of self-rated and objective health status predict (i) all-cause mortality and (ii) cause-specific mortality from leading causes of death in the UK.MethodsThe UK Biobank study recruited >500,000 participants, aged 37-73, between 2006–2010. The health cross-classification examined incorporated self-rated health (poor, fair, good or excellent) and health status derived from medical history and current disease status, including 81 cancer and 443 non-cancer illnesses. We examined all-cause mortality and six specific causes of death: ischaemic heart disease, cerebrovascular disease, influenza and pneumonia, dementia and Alzheimer’s disease, chronic lower respiratory disease and malignant neoplasm.ResultsAnalyses included >370,000 middle-aged and older adults with a median follow-up of 11.75 (IQR = 1.4) years, yielding 4,320,270 person years of follow-up. Compared to excellent self-rated health and favourable health status, all other levels of the health cross-classification were associated with a greater risk of mortality, with hazard ratios ranging from 1.22 (95% CI 1.15-1.29, pBonf. < 0.001) for good self-rated health and favourable health status to 7.14 (95% CI 6.70-7.60, pBonf. < 0.001) for poor self-rated health and unfavourable health status.ConclusionsOur findings highlight that self-rated health captures additional health-related information and should be more widely assessed across settings. The cross-classification between health status and self-rated health represents a straightforward metric for risk stratification, with applications to population health, clinical decision making and resource allocation.

Estimation of trip generation rates for residential land use in Uyo, Nigeria

Change in trip rates affects a transportation system and could lead to the redesign of the transport infrastructure in order to satisfy the new demand. This study estimates trip generation rates for residential land use in Uyo using cross classification method. Five (5) residential estates were considered and household survey carried out to collect trip data from 500 households on purpose and mode of travel through household interview and their response recorded in questionnaire. Four independent variables (household size, household income, car ownership, number of employed persons) were used for the study based on the prevailing conditions of theresidential land use. Cross-classification trip rates were developed from the most significant variables; household size, household income and car ownership. The analysis indicated that work trip produced the highest reported trip rates of 29.6% followed by religious trip of 24.7%. Similarly, private car trips contributed 42.8% of trips made by mode of travel as the highest trip. It was found that household size is the strongest socio-economic variable that influence trip generation in residential land use in Uyo. The cross-classification trip rates developed in this study could provide basis for the estimation of trip generation in residential land use in Uyo. Keywords: Trip generation; analysis of variance, ANOVA; cross classification