Linear Random Effects Research Articles

Modelling a preference-based index for EQ-5D-3L and EQ-5D-3L + Sleep using a Bayesian framework.

Conventionally, frequentist approach has been used to model health state valuation data. Recently, researchers started to explore the use of Bayesian methods in this area. This paper presents an alternative approach to modelling health state valuation data of the EQ-5D-3L and EQ-5D-3L + Sleep descriptive systems, using a Bayesian framework, and demonstrates its superiority to conventional frequentist methods. The valuation study is composed of 18 EQ-5D-3L health states and 18 EQ-5D-3L + Sleep health states valued by 160 members of the general public in South Yorkshire, UK, using the time tradeo-ff technique. Three different models were developed for EQ-5D-3L and EQ-5D-3L + Sleep accordingly using Bayesian Markov chain Monte Carlo simulation methods. Bayesian methods were applied to models fitted included a linear regression, random effect and random effect with covariates. The models are compared based on their predictive performance using mean predictions, root mean squared error (RMSE) and deviance information criterion (DIC). All analyses were performed using Bayesian Markov chain Monte Carlo simulation methods. The random effects with covariates model performs best under all criterions for the two preference-based measures, with RMSE (0.037) and DIC (637.5) for EQ-5D-3L and RMSE (0.019), DIC (416.4) for EQ-5D + Sleep. Compared with models previously estimated using frequentist approach, the Bayesian models reported in this paper provided better predictions of observed values. Bayesian methods provide a better way to model EQ-5D-3L valuation data with and without a sleep 'bolt-on' and provide a more flexible in characterizing the full range of uncertainty inherent in these estimates.

Variation Source Identification in Manufacturing Processes Using Bayesian Approach With Sparse Variance Components Prior

We present a Bayesian linear random effects’ model for variation source identification in multistage manufacturing processes with a prior for sparse variance components. A modified horseshoe+ (HS+) prior is used to tackle high-dimensional problems with low sample size and sparse variation sources. Furthermore, we introduce the informed HS+ (IHS+) prior that incorporates the likelihood information of possible variation sources. To estimate the variations from the IHS+ prior, a specially designed Gibbs sampler is established. Through a series of numerical experiments and case study, we showed that the proposed IHS+ outperforms the existing prior distributions when variation sources are sparse. Note to Practitioners —Economic globalization brings intense competition among manufacturing enterprises. The key to success in this competitive climate is a prompt response to rapidly changing market demands with high-quality products. Variation reduction is essential for improving process efficiency and product quality. Existing variation source identification approaches generally assume that the amount of measurement data is larger than that of the possible variation sources. We propose a method to identify the process variation sources with limited measurement capability. Moreover, we provide a tool to apply the domain knowledge on the possible variations in the manufacturing processes. The proposed method is flexible and can be adjusted according to the confidence level of the practitioners’ empirical knowledge. The experiments have shown that our method possesses advantageous features in detecting variation sources and estimating the size of variations.

Periadrenal Volume is a Better Predictor of Prolonged Operative Time in Laparoscopic Retroperitoneal Adrenalectomy than BMI.

Retroperitoneal laparoscopic adrenalectomy is gaining traction as a minimally invasive technique. One of the purported relative contraindications is BMI given the smaller working space. We hypothesize that other anthropometric measurements may be better predictors of operative time. An IRB-approved, single-institution, retrospective study of 83 patients who underwent laparoscopic retroperitoneal adrenalectomy evaluated the association of anthropometric measurements taken from cross-sectional imaging and the primary outcome of operative time. Descriptive statistics were performed with Wilcoxon rank-sum test for continuous variables (median; IQR) and Chi-square (n; %) for categorical variables. A linear random effects model was used to model operative time. The majority of the patients were white (40; 48.2%) women (46; 55.4%) with a median age of 54 with interquartile range (IQR) of 43-63 and a median BMI of 27.8 (IQR 21.2-38.6). On univariable analysis, factors that led to longer operative time included right-sided operation (p = 0.04), male gender (p < 0.01), clinical diagnosis (p < 0.01), waist area (p < 0.01), waist/hip ratio (p < 0.01), periadrenal volume (p < 0.01), posterior adiposity index (PAI) (p < 0.01) and BMI (p < 0.01). Only side, order of operation, and periadrenal fat volume (p < 0.01, p = 0.02 and p < 0.01, respectively) remained independent predictors of increased operative time on multivariable analysis. This study demonstrates that anthropometric measurements, specifically periadrenal fat volume, and side of operation, are better predictors for increased operative time in laparoscopic retroperitoneal adrenalectomies than BMI. This information can help facilitate appropriate patient selection for this operative approach.

Notes on Comparison of Covariance Matrices of BLUPs Under Linear Random-Effects Model with Its Two Subsample Models

A general linear random-effects model that includes both fixed and random effects, and its two subsample models are considered without making any restrictions on correlation of random effects and any full rank assumptions. Predictors of joint unknown parameter vectors under these three models have different algebraic expressions. Because of having different properties and performances under these models, it is one of the main focuses to make comparison of predictors. Covariance matrices of best linear unbiased predictors (BLUPs) of unknown parameters are used as a criterion to compare with other types predictors due to their definition of minimum covariance matrices structure. The comparison problem of covariance matrices of BLUPs under the models is considered in the study. We give a variety of equalities and inequalities in the comparison of covariance matrices of BLUPs of a general linear function of fixed effects and random effects under the models by using an approach consisting matrix rank and inertia formulas.

Differences in health-related quality of life by implant strategy: Analyses from the Interagency Registry for Mechanically Assisted Circulatory Support

Midterm change in health-related quality of life (HRQOL) by left ventricular assist device (LVAD) implant strategy is unknown. The purpose of this study was to examine HRQOL by pre-operative implant strategy from before to 2 years after surgery. Adult patients in the Interagency Registry for Mechanically Assisted Circulatory Support were stratified into 3 groups based on pre-implant device strategy: destination therapy (DT) (n = 2,901), bridge to transplant (BTT) (n = 2,209), and bridge to candidacy (BTC) (n = 3,076).HRQOL data were collected before and 2 years after surgery using the generic EQ-5D-3L survey and heart failure-specific Kansas City Cardiomyopathy Questionnaire (KCCQ-12). Statistical analyses included chi-square tests, analysis of variance, paired t-tests, and general linear random effects models. Between April 1, 2008 and June 30, 2013, 4,422 patients and 1,660 patients (majority males and ≥50 years) who received primary continuous flow LVADs completed baseline EQ-5D-3L and KCCQ-12 questionnaires, respectively, whereas 1,615 and 1,408 patients completed EQ-5D-3L and KCCQ-12 questionnaires at 2 years, respectively. Although paired t-tests and general linear random effects models showed that both heart failure-specific and generic HRQOL improved for all groups across time (p-values <0.05), some differences in HRQOL were found by implant strategy at baseline and 2 years, with a pattern favoring better functioning for patients with BTT. The BTT group reported significantly higher overall HRQOL pre-implant using the KCCQ-12 (BTT = 37.09, BTC = 33.57, and DT = 33.56) and at 2 years using the EQ-5D-3L (BTT = 75.18, BTC = 72.27, and DT = 70.87) (p-values <0.05), although these differences were not clinically important differences. Differences in HRQOL domains were also found. Using generic and heart failure-specific instruments, overall HRQOL generally improved from before to 2 years after mechanical circulatory support implant regardless of implant strategy, although important domain-specific differences by group were identified.

Six-Minute Walk Distance After Critical Illness: A Systematic Review and Meta-Analysis.

Impaired physical functioning is common and long lasting after an intensive care unit (ICU) admission. The 6-minute walk test (6MWT) is a validated and widely used test of functional capacity. This systematic review synthesizes existing data in order to: (1) evaluate 6-minute walk distance (6MWD) in meters over longitudinal follow-up after critical illness, (2) compare 6MWD between acute respiratory distress syndrome (ARDS) versus non-ARDS survivors, and (3) evaluate patient- and ICU-related factors associated with 6MWD. Five databases (PubMed, EMBASE, Cumulative Index of Nursing and Allied Health Literature, PsychINFO, and Cochrane Controlled Trials Registry) were searched to identify studies reporting 6MWT after hospital discharge in survivors from general (ie, nonspeciality) ICUs. The last search was run on February 14, 2018. Databases were accessed via Johns Hopkins University Library. Pooled mean 6MWD were reported, with separate linear random effects models used to evaluate associations of 6MWD with ARDS status, and patient- and ICU-related variables. Twenty-six eligible articles on 16 unique participant groups were included. The pooled mean (95% confidence interval [CI]) 6MWD results at 3- and 12-months post discharge were 361 (321-401) and 436 (391-481) meters, respectively. There was a significant increase in 6MWD at 12 months compared to 3 months (P = .017). In ARDS versus non-ARDS survivors, the mean (95% CI) 6MWD difference over 3-, 6-, and 12-month follow-up was 73 [13-133] meters lower. Female sex and preexisting comorbidity also were significantly associated with lower 6MWD, with ICU-related variables having no consistent associations. Compared to initial assessment at 3 months, significant improvement in 6MWD was reported at 12 months. Female sex, preexisting comorbidity, and ARDS (vs non-ARDS) were associated with lower 6MWT results. Such factors warrant consideration in the design of clinical research studies and in the interpretation of patient status using the 6MWT.

Effects of between-batch variability on the type I error rate in biosimilar development

Biological products are known to have some between-batch variation. However, the traditional method to assess biosimilarity does not consider such between-batch variation. Beta-binomial models and linear random effect models are considered in order to incorporate between-batch variation for the binary endpoints and the continuous endpoints, respectively. In this article, emphasis is on the beta-binomial models for the binary endpoint case. For the linear random effect models of the continuous endpoint case, we cite relevant references along with conducting some simulation studies. Overall, we show that the type I error rates are inflated when biosimilarity is evaluated by the traditional method, which ignores between-batch variation.

Evidence of Between- and Within-Household Child Nutrition Inequality in Malawi: Does the Gender of the Household Head Matter?

This paper uses a variance decomposition approach to investigate inter- and intra-household inequalities in under-five children’s nutritional status in Malawi and distinguish between explained and unexplained variance. The adopted linear random effect model is based on a version of the conceptual framework by UNICEF, and the adopted dataset is from the Malawian Endline Survey 2014. The selected impact variable is child stunting. The explanatory variables are representative of the underlying and basic causes of this nutritional status. This study adds to the literature by providing an estimate by the gender of the household head and up-to-date evidence on the case of Malawi, where the prevalence of child malnutrition is rampant. The results reveal that between-household inequalities, especially those unexplained by the adopted model, mostly drive the long-term child nutritional inequalities. Policy suggestions for the design of nutritional and redistributive policies aimed at achieving the Sustainable Development Goals (SDGs) and the country’s growth and development strategy are provided.

Histopathology of the right ventricular outflow tract and the relation to hemodynamics in patients with repaired tetralogy of Fallot

ObjectiveTo evaluate the relationship between myocardial histopathology and tissue Doppler imaging (TDI) variables of the right ventricle and postoperative peak systolic right-to-left ventricular pressure ratio (Prv/Plv) in patients undergoing intracardiac repair for tetralogy of Fallot (TOF). MethodsOperatively resected crista supraventricularis muscle specimens from 93 patients undergoing intracardiac repair for TOF, aged 18 months to 26 years (mean, 7.02 ± 5.35 years) were subjected to light microscopy. TDI-derived parameters between the normal and abnormal categories of myocardium, the evolution of Prv/Plv, and its relationship to TDI-derived variables were tested using generalized linear random effects model using xtreg command. ResultsThe incidence of myocyte hypertrophy, myocytolysis, and perivascular fibrosis was 89.2%, 83.8%, and 77.4%, respectively. Although tricuspid annular peak systolic excursion, peak myocardial velocity during systole (s′), and early diastolic basal lengthening of right ventricle (e′) continued to improve among patients with myocardial hypertrophy, myocytolysis, and perivascular fibrosis, there was an absence of improvement of the late diastolic relaxation of right ventricular free wall (a′) in patients with perivascular fibrosis. Although there was improvement of postoperative Prv/Plv in patients with myocardial fibrosis as compared with normal histology, the values were not statistically significant (β [standard error] −0.07 [0.08], P = .3). ConclusionsThe great majority of myocardial tissues in cyanotic TOF indicate pre-existing hypertrophic, degenerative, and fibrotic changes. Perivascular fibrosis affects the diastolic compliance of the right ventricle and may account for the absence of improvement of late diastolic relaxation (a′) and greater postoperative Prv/Plv in the absence of a residual surgical lesion.

Computation and comparison of estimators under different linear random-effects models

Estimations of unknown parameters in linear regression models have been main objects of study in statistical data analysis and inference. This work concerns the comparison problem of different estimations of unknown parameters in two linear random-effects models (LREMs) with different structures. We shall derive analytical formulas for calculating best linear unbiased predictors/best linear unbiased estimators (BLUPs/BLUEs) of all unknown parameter spaces in the two competing LREMs. We then discuss relationships among BLUPs/BLUEs under the two LREMs by using some matrix analysis tools, including various consequences and conclusions on some special cases of LREMs.

Mixedsde: A Package to Fit Mixed Stochastic Differential Equations

Stochastic differential equations (SDEs) are useful to model continuous stochastic processes. When (independent) repeated temporal data are available, variability between the trajectories can be modeled by introducing random effects in the drift of the SDEs. These models are useful to analyse neuronal data, crack length data, pharmacokinetics, financial data, to cite some applications among other. The R package focuses on the estimation of SDEs with linear random effects in the drift. The goal is to estimate the common density of the random effects from repeated discrete observations of the SDE. The package mixedsde proposes three estimation methods: a Bayesian parametric, a frequentist parametric and a frequentist nonparametric method. The three procedures are described as well as the main functions of the package. Illustrations are presented on simulated and real data.

Analysis of learning deficits in aged rats on the W-track continuous spatial alternation task.

Young and aged animals were tested on a spatial alternation task that consisted of two interleaved components: (1) an "outbound" or alternation component (working memory) and (2) an "inbound" component, requiring the animal to remember to return to a central location in space (spatial memory). In the present study, aged rats made more outbound errors throughout testing, resulting in significantly more days to reach learning criterion, as compared to young rats. Furthermore, while all animals were able to learn the hippocampus-dependent inbound component of the task, most aged animals remained just above chance on the outbound component, even after extended testing days. Aged rats may be more impaired on the outbound part of the task because it requires cooperation of both the hippocampus and mPFC, each of which is compromised with age. In addition to presenting these results, we compare one commonly used analysis (repeated measures ANOVA) and two less common hierarchical modeling techniques (hierarchical generalized linear model and state-space random effects model) to determine the best method for comparing population learning over time. We found that hierarchical modeling is the most appropriate for this task and that a state-space model better captures the behavioral responses. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

RANDOM FOREST ALGORITHM OPTIMIZATION OF ENTERPRISE FINANCIAL INFORMATION MANAGEMENT SYSTEM

The optimization of random forest algorithms for enterprise financial information management systems is studied in this paper. A random forest algorithm was proposed to improve the data processing capabilities of the financial system. This paper proposes a random forest model on the premise of referring to the latest results of machine learning. The algorithm was introduced into the real estate business financial management system in this paper. First, the samples are divided into training samples and test samples, and the direct prediction method and the two-step prediction method are applied. Mean SR and MAPE were used to compare the prediction accuracy of different algorithms and it was found that the direct prediction method is better. In the algorithm used in this paper, the random forest effect is the best. Then the linear regression, decision tree, neural network and random forest model fitting effects were compared and the best fitting degree of random forest was found.

Depression, anxiety and delinquency: Results from the Pittsburgh Youth Study

PurposeThe main aim of this research is to investigate to what extent within-individual changes in anxiety and depression are related to within-individual changes in theft and violence. MethodsThe youngest sample of the Pittsburgh Youth Study (PYS), a prospective longitudinal survey of 503 boys followed up from age 7 onwards, was analyzed. Depression and anxiety were measured for boys from ages 11 to 16 as were moderate and serious forms of self-reported theft and violence. A hierarchical linear random effects model was used to investigate anxiety and depression as potential causes or outcomes of these forms of delinquency. ResultsThe results showed that the between-individual correlations were consistently higher than the corresponding within-individual correlations, and provided little evidence to discern the directionality of the potential relationships between depression, anxiety and delinquency. Using a random effects approach, there was limited evidence that prior depression or anxiety was related to later offending, but there was evidence that offending (particularly theft and serious violence) was associated with later increases in anxiety, and to a lesser extent depression. ConclusionsThis study indicates that depression and anxiety were outcomes of offending. If replicated, this would suggest that evidence-based interventions which reduced offending would have a desirable influence in reducing depression and anxiety. However, interventions for depression should still form part of responsive interventions. More research which explores within-individual changes in longitudinal studies with repeated measures is needed.

Diagnostic Prevalence of Common Psychiatric Comorbidities of Alcohol Use Disorders in India: a Systematic Review

The diagnostic prevalence of mood and anxiety disorders in Indian individuals with alcohol use disorders (AUDs) has important implications for clinical decision making and for cross-cultural psychiatry research. Here, we present a systematic review of comorbid mood and anxiety disorders in Indian patients with AUDs. We find rates of comorbid mood disorders ranging from 6 to 65% (estimated overall effect from a linear random effects model of 18.0%, 95% CI [5.6, 45.1]) and rates of comorbid anxiety disorders ranging from 0 to 30% (estimated overall effect of 2.4%, 95% CI [0.9, 5.8]) in Indian populations. These findings are placed in the context of estimated population prevalence of comorbid mood and anxiety disorders in US adults with AUDs who have recently sought treatment of 40.7 (95% CI [32.6, 48.7]) and 33.4% (95% CI [25.2, 41.6]), respectively. Potential causes of these differences include differences in diagnosis, in cultural norms around alcohol use and the acceptance of mental health care, and in the types of studies being compared. The findings highlight the importance of systematic assessments of treatment outcomes in local clinical populations as a complement to an understanding of the medical literature from other cultural contexts.

The concentration-response between long-term PM2.5 exposure and mortality; A meta-regression approach

BackgroundLong-term exposure to ambient fine particulate matter (≤ 2.5 μg/m3 in aerodynamic diameter; PM2.5) is significantly associated with increased risk of premature mortality. Our goal was to provide an updated meta-analysis of all-cause and cause-specific mortality associated with exposure to PM2.5 and to better estimate the risk of death as a function of air pollution levels. MethodsWe systematically searched all published cohort studies examining the association between long term exposure to PM2.5 and mortality. We applied multivariate linear random effects meta-analysis with random effects for cohort, and study within cohort. Meta-regression techniques were used to test whether study population or analytic characteristics modify the PM2.5 -mortality association and to estimate the shape of the concentration-response curve. ResultsA total of 53 studies that provided 135 estimates of the quantitative association between the risk of mortality and exposure to PM2.5 were included in the meta-analysis. There were 39 studies from North America, 8 from Europe, and 6 from Asia. Since 2015, 17 studies of long-term air pollution exposure have been published, covering, wider geographic areas with a wider range of mean exposures (e.g. <12 or > 20 μg/m3). A penalized spline showed the slope decreased at higher concentrations but appeared to level off. We found that the inverse transform of average PM2.5 well approximated that spline and provided a parametric estimate that fit better than a linear or logarithmic term for average PM2.5. In addition, we found that studies using space time exposure models or fixed monitors at Zip-code scale (as compared to land use regression method), or additionally controlling for area level socio-economic status, or with mean exposure less than 10 μg/m3 were associated with higher mortality effect estimates. ConclusionsThis meta-analysis provides strong evidence for the adverse effect of PM2.5 on mortality, that studies with poorer exposure have lower effect size estimates, that more control for SES increases effect size estimates, and that significant effects are seen below 10 µg/m3. The concentration -response function produced here can be further applied in the global health risk assessment of air particulate matter.

Linear scalar-on-surface random effects regression models

Many research fields increasingly involve analyzing data of a complex structure. Models investigating the dependence of a response on a predictor have moved beyond the ordinary scalar-on-vector regression. We propose a regression model for a scalar response and a surface (or a bivariate function) predictor. The predictor has a random component and the regression model falls in the framework of linear random effects models. We estimate the model parameters via maximizing the log-likelihood with the ECME (Expectation/Conditional Maximization Either) algorithm. We use the approach to analyze a data set where the response is the neuroticism score and the predictor is the resting-state brain function image. In the simulations we tried, the approach has better performance than two other approaches, a functional principal component regression approach and a smooth scalar-on-image regression approach.

Approximate maximum likelihood estimation for stochastic differential equations with random effects in the drift and the diffusion

Consider N independent stochastic processes \((X_i(t), t\in [0,T])\), \(i=1,\ldots , N\), defined by a stochastic differential equation with random effects where the drift term depends linearly on a random vector \(\Phi _i\) and the diffusion coefficient depends on another linear random effect \(\Psi _i\). For these effects, we consider a joint parametric distribution. We propose and study two approximate likelihoods for estimating the parameters of this joint distribution based on discrete observations of the processes on a fixed time interval. Consistent and \(\sqrt{N}\)-asymptotically Gaussian estimators are obtained when both the number of individuals and the number of observations per individual tend to infinity. The estimation methods are investigated on simulated data and show good performances.

Abstract 380: The association of sleep with metabolic pathways and metabolites: Evidence from the Dietary Approaches to Stop Hypertension (DASH)-Sodium Feeding Study

Abstract Background Sleep is increasingly being viewed as an important determinant of health, and appears to be associated with cardiovascular disease, diabetes and several site-specific cancers. Few epidemiologic studies have used data from a feeding study to explore the impact of sleep habits up on metabolomic profile. Methods The Dietary Approaches to Stop Hypertension (DASH)-Sodium feeding trial randomized individuals to either the DASH (low fat, high protein, low-fat dairy and high fruits and vegetables) or control diet (12 weeks) and three levels of sodium intake (30 days each). In a subgroup of 97 participants, we measured the levels of 531 metabolites in serial fasting plasma samples at two time points using liquid chromatography and gas chromatography mass spectrometry. We then assessed the association between each metabolite and sleep using a linear random effects model adjusted for age, sex, race, dietary pattern and sodium level. The resulting p-values were combined using Fisher's method to estimate the association with 38 metabolic pathways. Data about sleep were collected at the end of two dietary intervention phases. Participants recorded times for going-to-sleep and waking while wearing 24-hour ambulatory blood pressure monitors. From this we produced two sleep variables - sleep mid-point (the median time between going to sleep and waking) and sleep duration. We undertook further analysis to explore the associations between bedtime, wake-time and metabolic pathways and metabolites. Results Nineteen pathways were significantly associated (p &amp;lt;0.05) with sleep mid-point, but only the γ-glutamyl amino acid metabolism pathway was associated with sleep mid-point and wake-time at Bonferroni-corrected threshold of 0.0013 (7 individual peptide metabolites contributed and showed positive associations). Fifty-six metabolites were associated with sleep mid-point (FDR&amp;lt;0.20), although none of them reached Bonferroni-corrected significance (p-value &amp;lt;10-5). Notable top metabolites (pathways given in brackets) associated with sleep mid-point and wake-time were CMPF (fatty acid, decarboxylate), erythrulose (an advanced glycation end-product), isovalerate (fatty acid metabolism) and HWESASXX (polypeptide). Neither sleep duration nor bedtime were associated with metabolite pathways or metabolites and there was no significant interaction between sex, race, BMI or dietary intervention and sleep. Conclusion Within our study we found multiple metabolites associated with sleep timing. In particular, the γ-glutamyl pathway metabolites were associated with sleep mid-point and wake-time. These metabolites are formed using the enzyme γ-glutamyltransferase (GGT), which is expressed in response to oxidative stress and glutathione synthesis - a process that may play a role in carcinogenesis, diabetes, and cardiovascular disease. Citation Format: Vanessa L. Gordon-Dseagu, Andriy Derkach, Qian Xiao, Ishmael Williams, Joshua Sampson, Rachael Stolzenberg-Solomon. The association of sleep with metabolic pathways and metabolites: Evidence from the Dietary Approaches to Stop Hypertension (DASH)-Sodium Feeding Study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 380.

Variability and reliability study of overall physical activity and activity intensity levels using 24\xa0h-accelerometry-assessed data

Background24 h-accelerometry is now used to objectively assess physical activity (PA) in many observational studies like the German National Cohort; however, PA variability, observational time needed to estimate habitual PA, and reliability are unclear.MethodsWe assessed 24 h-PA of 50 participants using triaxial accelerometers (ActiGraph GT3X+) over 2 weeks. Variability of overall PA and different PA intensities (time in inactivity and in low intensity, moderate, vigorous, and very vigorous PA) between days of assessment or days of the week was quantified using linear mixed-effects and random effects models. We calculated the required number of days to estimate PA, and calculated PA reliability using intraclass correlation coefficients.ResultsBetween- and within-person variance accounted for 34.4–45.5% and 54.5–65.6%, respectively, of total variance in overall PA and PA intensities over the 2 weeks. Overall PA and times in low intensity, moderate, and vigorous PA decreased slightly over the first 3 days of assessment. Overall PA (p = 0.03), time in inactivity (p = 0.003), in low intensity PA (p = 0.001), in moderate PA (p = 0.02), and in vigorous PA (p = 0.04) slightly differed between days of the week, being highest on Wednesday and Friday and lowest on Sunday and Monday, with apparent differences between Saturday and Sunday. In nested random models, the day of the week accounted for < 19% of total variance in the PA parameters. On average, the required number of days to estimate habitual PA was around 1 week, being 7 for overall PA and ranging from 6 to 9 for the PA intensities. Week-to-week reliability was good (intraclass correlation coefficients, range, 0.68–0.82).ConclusionsIndividual PA, as assessed using 24 h-accelerometry, is highly variable between days, but the day of assessment or the day of the week explain only small parts of this variance. Our data indicate that 1 week of assessment is necessary for reliable estimation of habitual PA.