Random Intercept Linear Models Research Articles

Digital Self-management, Analgesic Use, and Patient-Reported Outcomes in Knee or Hip Osteoarthritis

ObjectiveTo investigate changes in analgesic use before and after participation in a digital first-line treatment program (exercise, patient education) in persons with knee or hip osteoarthritis (OA) and to explore associations between these changes in medication use and participant-reported pain and function. DesignRetrospective cohort study with pre-post measures. SettingCommunity setting. ParticipantsIndividuals (N=4100; mean age ± SD, 64.5±9.3y; 73.3% women) participating in the digital program. InterventionsA digital supervised education and exercise therapy. Main Outcome MeasuresSelf-reported analgesic use for knee/hip pain during the past month at baseline and 12-week follow-up, knee/hip numeric rating scale pain (0-10, a higher value indicating more pain), and Knee Injury and Osteoarthritis Outcome Score 12 or Hip Disability and Osteoarthritis Outcome Score 12 function subscale (0-100, higher values indicating better function). McNemar test, multivariable logistic regression, and linear random intercept model were used for statistical analyses. ResultsAmong participants, 61.4% and 49.4% were analgesic users at baseline and 12-week follow-up, respectively, (absolute reduction 12.0%; 95% confidence interval, 10.5-13.5). Being female, having hip OA, lower education, higher body mass index, living outside large metropolitan cities, coexisting rheumatoid arthritis, and walking difficulties were associated with higher odds of analgesic use at baseline. At both time points, persons not using analgesics at the time reported better outcomes. All groups but “new users” experienced improvements in their pain and function following participation in digital program with the greatest improvements observed among “quitters.” ConclusionsEngaging in a digital exercise and patient education program as a primary treatment for knee or hip OA was associated with a reduction in the use of analgesics. The greatest improvements were seen for those who stopped analgesic use. These results highlight the importance of providing effective first-line treatment to people with knee or hip OA.

Development of performance and learning rate evaluation models in robot-assisted surgery using electroencephalography and eye-tracking

The existing performance evaluation methods in robot-assisted surgery (RAS) are mainly subjective, costly, and affected by shortcomings such as the inconsistency of results and dependency on the raters’ opinions. The aim of this study was to develop models for an objective evaluation of performance and rate of learning RAS skills while practicing surgical simulator tasks. The electroencephalogram (EEG) and eye-tracking data were recorded from 26 subjects while performing Tubes, Suture Sponge, and Dots and Needles tasks. Performance scores were generated by the simulator program. The functional brain networks were extracted using EEG data and coherence analysis. Then these networks, along with community detection analysis, facilitated the extraction of average search information and average temporal flexibility features at 21 Brodmann areas (BA) and four band frequencies. Twelve eye-tracking features were extracted and used to develop linear random intercept models for performance evaluation and multivariate linear regression models for the evaluation of the learning rate. Results showed that subject-wise standardization of features improved the R2 of the models. Average pupil diameter and rate of saccade were associated with performance in the Tubes task (multivariate analysis; p-value = 0.01 and p-value = 0.04, respectively). Entropy of pupil diameter was associated with performance in Dots and Needles task (multivariate analysis; p-value = 0.01). Average temporal flexibility and search information in several BAs and band frequencies were associated with performance and rate of learning. The models may be used to objectify performance and learning rate evaluation in RAS once validated with a broader sample size and tasks.

EFFECTS OF SOCIAL ENGAGEMENT ON WELL-BEING AND COGNITIVE FUNCTIONING IN OLDER HISPANICS

Abstract Research on engaged lifestyles demonstrates the importance of social engagement for aging well. The present study examined this thesis in the context of minority aging. With the theoretical lens that Hispanic cultures value social relationships, we tested whether social engagement was particularly important for maintaining well-being and cognitive functioning in Foreign-born (FB-H) and U.S.-born Hispanics (USB-H) relative to non-Hispanic Whites (NHW). Participants (N = 2677) was drawn from older adults in the Health and Retirement Study who completed the 2008, 2012, 2016, and 2020 module at least once. Two types of social engagement were assessed. Social interactions were based on the frequency of interactions with close partners (i.e., children, family members, and friends). Social activity participation was measured by the frequency of volunteering, doing charity, joining clubs, and attending meetings of non-religious organizations. Results of random intercept linear models controlling for covariates showed that, consistent with the literature, social interactions predicted better life satisfaction and that both aspects of social engagement predicted better immediate and delayed recall. Unexpected patterns emerged when interaction effects of race/nativity and social engagement were tested. While the general pattern held for NHWs, we found that greater social interactions predicted lower life satisfaction in USB-H. For FB-H, greater participation in social activities correlated with lower performance on immediate recall. Echoing a recent study that reveals mixed evidence for the well-documented paradox that Hispanic immigrants hold health advantages (Tarraf et al., 2020), we call for more research on lifestyle factors and aging trajectories in racially diverse populations.

Do regional labor market conditions matter for temporary work disability duration? A multilevel analysis of workers' compensation claims in five Canadian provinces.

Research has shown how regionally varying labor market conditions are associated with differences in work disability duration. However, the majority of these studies have not used multilevel models to appropriately account for the hierarchical clustering of individuals nested within contextual units (e.g., regions). Studies that have used multilevel models have focused on privately insured workers or on disability not specifically caused by work-related injury or illness. Using claims data from five Canadian provincial workers' compensation systems, linear random-intercept models were used to estimate how much variance in temporary work disability duration ('work disability duration' for brevity) for work-related injuries and musculoskeletal disorders was due to differences between economic regions, what economic region-level labor market characteristics were associated with work disability duration, and what characteristics best explained economic region differences in work disability duration. Economic region characteristics, such as the unemployment rate and proportion of goods-producing employment, were independently associated with individual-level work disability duration. However, economic region variation only accounted for 1.5%-2% of total variation in work disability duration. The majority (71%) of economic region-level variation was explained by the provincial jurisdiction where the worker livedand was injured. Regional variation tended to be greater for female workers than males. The findings suggest that while regional labor market conditions matter for work disability duration, system-level differences in workers' compensation and health care are more important factors influencing work disability duration. Furthermore, while this study includes both temporary and permanent disability claims, the work disability duration measure only captures temporary disability.

Does the female-headed household suffer more than the male-headed from Covid-19 impact on food security? Evidence from Ethiopia

Covid-19 pandemic and accompanying policy measures affected women's economic and social lives differently from men. However, its effect on the food security situation of women and men remains unclear. This study investigated whether female-headed households suffer more from covid 19 impact on food security than male-headed households in Ethiopia using the World Bank High Frequency Phone Survey data. Food Insecurity Experience Scale (FIES) is applied to measure food insecurity and multilevel-mixed effect random intercept linear model is used to analyze the determinants. The survey data show that, during the pandemic in 2020, nearly 27% of households face moderate or severe food insecurity and close to 6% face severe food insecurity. Regression results revealed that the pandemic and accompanying policy measures affected food security of female-headed households differently from male-headed households. The likelihood of female-headed households experiencing moderate or severe food insecurity is significantly higher by 5.7% points than male-headed households. Education level of the household head is the critical determinant of women's food insecurity. Every additional year of education of female heads has a significantly higher impact in reducing the likelihood of experiencing food insecurity than that of male-headed households. It is also found that having access to basic food items reduces the likelihood of facing moderate or severe food insecurity by 7.4% points. Besides, education level of the household head, income change and family size are critical determinants for food insecurity of the sample households. It is recommended that stakeholders and policy makers focus in the gender aspect while designing policy measures to control pandemics and fucus on improving accessibility of basic needs such as food items to all.

Financial remittances and election turnout in the municipalities of the Republic of Serbia: the economy of voting

This paper investigates possible evidence on whether remittances from abroad have had a significant influence on the outcome of Serbia’s presidential elections. Using the linear random-intercept model and based on presidential- election data for municipalities between 2002 and 2017, the study finds that voter turnout rates were positively related to the share of remittances in the municipalities’ GDP per capita. For a 1% increase in the share of remittances in municipalities’ GDP, the voting turnout increases by 0.31% on average. Also, regarding the within-municipality effect, as the share of remittances in GDP increases over time by 1%, voter turnout decreases by 0.42%. Further, it reveals how the size of the population temporarily abroad had a negative effect on voter participation throughout the entire period examined across both municipalities and time. The study also captures the effects of education and ethnicity. Specifically, as the share of highly educated people increases by 1% over time, so the election participation rates in a particular municipality also increase by 0.83%. After introducing categorical variables for the ethnic composition of the municipality, the study finds that in municipalities with a non-Serb majority, the level of participation was on average 3.31% lower, from 2002 to 2017. Finally, a statistically significant difference in voter turnout was recorded between the economically most developed and the least developed regions: the most developed region had on average 7.71% higher election participation than the region characterized by high emigration and large financial inflows. As far as Serbia is concerned, the final results confirm the findings of previous literature that the economical vacuum, created by the disappearance of state support, is filled by remittances. Possible reasons for the results are discussed. Keywords: Republic of Serbia, temporary migrants, financial remittances, voter turnout, municipality

Impact of duty hours on competency-related knowledge acquisition among community hospital residents.

The effect of duty hour (DH) restrictions on postgraduate residents' acquisition of clinical competencies is unclear. We evaluated the relationship between DHs and competency-related knowledge acquisition using the General Medicine In-training Examination (GM-ITE). We conducted a multicenter, cross-sectional study of community hospital residents among 2019 GM-ITE examinees. Self-reported average DHs per week were classified into five DH categories and the competency domains were classified into four areas: symptomatology and clinical reasoning (CR), physical examination and clinical procedure (PP), medical interview and professionalism (MP), and disease knowledge (DK). The association between these scores and DHs was examined using random-intercept linear models with and without adjustment for confounding factors. We included 4753 participants in the analyses. Of these, 31% were women, and 49.1% were in the postgraduate year (PGY) 2. Mean CR and MP scores were lower among residents in Category 1 (<50 h) than in residents in Category 3 (≥60 and <70 h; reference group). Mean DK scores were lower among residents in Categories 1 and 2 (≥50 and <60 h) than in the reference group. PGY-2 residents in Categories 1 and 2 had lower CR scores than those in Category 3; however, PGY-1 residents in Category 5 showed higher scores. The relationship between DHs and each competency area is not strictly linear. The acquisition of knowledge of physical examination and clinical procedures skills in particular may not be related to DHs.

Impact of the resident duty hours on in-training examination score: A nationwide study in Japan

Purpose The relationship between duty hours (DH) and the performance of postgraduate residents is needed to establish appropriate DH limits. This study explores their relationship using the General Medicine In-training Examination (GM-ITE). Materials and methods In this cross-sectional study, GM-ITE examinees of 2019 had participated. We analyzed data from the examination and questionnaire, including DH per week (eight categories). We examined the association between DH and GM-ITE score, using random-intercept linear models with and without adjustments. Results Five thousand five hundred and ninety-three participants (50.7% PGY-1, 31.6% female, 10.0% university hospitals) were included. Mean GM-ITE scores were lower among residents in Category 2 (45–50 h; mean score difference, −1.05; p < 0.001) and Category 4 (55–60 h; −0.63; p = 0.008) compared with residents in Category 5 (60–65 h; Reference). PGY-2 residents in Categories 2–4 had lower GM-ITE scores compared to those in Category 5. University residents in Category 1 and Category 5 showed a large mean difference (−3.43; p = 0.01). Conclusions DH <60–65 h per week was independently associated with lower resident performance, but more DH did not improve performance. DH of 60–65 h per week may be the optimal balance for a resident's education and well-being.

Cross-sectional associations between number and ages of children and maternal physical activity

Abstract Background Physical activity (PA) has many health benefits, but parenthood is often associated with reduced PA in mothers. Determinants of adult PA are diverse, but ages and number of children may be key factors linked to maternal PA. Few studies have compared PA levels of mothers of school-aged children with those who have younger children, which is vital considering that maternal PA patterns may change at transition to formal schooling. Studies using device-assessment to investigate this association are also sparse. We aimed to investigate the association between number and ages of children and maternal device-measured moderate-to-vigorous PA (MVPA). Methods Cross-sectional analyses were conducted using data from 848 mothers from the Southampton Women's Survey when the index child was 4 or 6-7 years old. Two-level random intercept linear models were run to investigate associations between self-reported number (1, 2, 3+) and ages (4y and under, 5y and over, both age groups) of children, and their potential interaction, and accelerometer-assessed log-transformed minutes of maternal MVPA. Results are presented as % difference in minutes of MVPA. Results 41% of mothers met PA guidelines (150 minutes of MVPA/week). Women who had any children 5y and over engaged in more MVPA than those with only children 4y and under (e.g. % difference in minutes of MVPA: 42.2, 95% CI 18.7 to 70.4 for mothers with only children 5y and over vs those with only children 4y and under). Mothers with 2 or 3+ children did less MVPA than those with 1 child (e.g. 13.9, 1.0 to 25.2 less MVPA for women with 2 children vs those with 1 child). Conclusions Mothers with multiple children or any children 4y and under do less MVPA than those with 1 child or only school-aged children. Interventions and policies are needed to support these mothers to increase their time spent doing MVPA, especially considering the additional health benefits associated with this level of intensity of PA. Key messages Most mothers are insufficiently physically active to gain the associated health benefits. Interventions to increase MVPA particularly need to target mothers of younger children and those with multiple children.

Mathematical Literacy and Collaborative Problem-Solving: Comparison between Korean and U.S. Students in PISA2015

We examined the relationships between mathematics achievement and collaborative problem solving, which has been emphasized as the 21st century skills. Focusing on moderating effects of attitude toward working with others, we analyzed the PISA 2015 data for binational comparison between Korean and the U.S. Random-intercept linear models – where the intercepts vary across schools while slopes are fixed across schools – were applied to the data including 5,545 students in Korea and 5,564 in the U.S. The findings showed the positive linear relationships between mathematics achievement and collaborative problem solving as expected, but the slope was remarkably higher in the Korean model. However, we found no significant moderating effects of attitude toward working with others on the relationship between mathematics achievement and collaborative problem solving. Particularly, students in Korea and the U.S. are likely to have similarly low scores in mathematics achievement when their collaborative problem solving scores are low. However, Korean students would have even higher mathematics achievement scores if students in Korea and the U.S. have similarly high collaborative problem solving scores. The findings suggest the necessity for further scrutinies on attitude toward working with others and research on teaching and learning the 21st century skills in an integrated way.

App-based food Go/No-Go training: User engagement and dietary intake in an opportunistic observational study

Food Go/No-Go training aims to alter implicit food biases by creating associations between perceiving unhealthy foods and withholding a dominant response. Asking participants to repeatedly inhibit an impulse to approach unhealthy foods can decrease unhealthy food intake in laboratory settings. Less is known about how people engage with app-based Go/No-Go training in real-world settings and how this might relate to dietary outcomes. This pragmatic observational study investigated associations between the number of completed app-based food Go/No-Go training trials and changes in food intake (Food Frequency Questionnaire; FFQ) for different healthy and unhealthy food categories from baseline to one-month follow-up. In total, 1234 participants (m(BMI) = 29 kg/m2, m(age) = 43years, 69% female) downloaded the FoodT app and completed food-Go/No-Go training at their own discretion (mean number of completed sessions = 10.7, sd = 10.3, range: 1–122). In pre-registered analyses, random-intercept linear models predicting intake of different foods, and controlled for baseline consumption, BMI, age, sex, smoking, metabolic syndrome, and dieting status, revealed small, significant associations between the number of completed training trials and reductions in unhealthy food intake (b = -0.0005, CI95= [-0.0007;-0.0003]) and increases in healthy food intake (b = 0.0003, CI95 = [0.0000; 0.0006]). These relationships varied by food category, and exploratory analyses suggest that more temporally spaced training was associated with greater changes in dietary intake. Taken together, these results imply a positive association between the amount of training completed and beneficial changes in food intake. However, the results of this pragmatic study should be interpreted cautiously, as self-selection biases, motivation and other engagement-related factors that could underlie these associations were not accounted for. Experimental research is needed to rule out these possible confounds and establish causal dose-response relationships between patterns of engagement with food Go/No-Go training and changes in dietary intake.

Urban social environment and low birth weight in 360 Latin American cities

ObjectiveUsing data compiled by the SALURBAL project (Urban Health in Latin America; ‘Salud Urbana en América Latina’) we quantified variability in low birth weight (LBW) across cities in Latin America, and evaluated the associations of socio-economic characteristics at various levels (maternal, sub-city and city) with the prevalence of LBW.MethodsThe sample included 8 countries, 360 cities, 1321 administrative areas within cities (sub-city units) and birth registers of more than 4.5 million births for the year 2014. We linked maternal education from birth registers to data on socioeconomic characteristics of sub-cities and cities using the closest available national population census in each country. We applied linear and Poisson random-intercept multilevel models for aggregated data.ResultsThe median prevalence of city LBW by country ranged from a high of 13% in Guatemala to a low of 5% in Peru (median across all cities was 7.8%). Most of the LBW variability across sub-cities was between countries, but there were also significant proportions between cities within a country, and within cities. Low maternal education was associated with higher prevalence of LBW (Prevalence rate ratios (PRR) for less than primary vs. completed secondary or more 1.12 95% CI 1.10, 1.13) in the fully adjusted model. In contrast, higher sub-city education and a better city social environment index were independently associated with higher LBW prevalence after adjustment for maternal education and age, city population size and city gross domestic product (PRR 1.04 95% CI 1.03, 1.04 per SD higher sub-city education and PRR 1.02 95% CI 1.00, 1.04 per SD higher SEI). Larger city size was associated with a higher prevalence of LBW (PRR 1.06; 95% CI 1.01, 1.12).ConclusionOur findings highlight the presence of heterogeneity in the distribution of LBW and the importance of maternal education, local and broader social environments in shaping LBW in urban settings of Latin America. Implementing context-sensitive interventions guided to improve women’s education is recommended to tackle LBW in the region.

Running barefoot leads to lower running stability compared to shod running - results from a randomized controlled study

Local dynamic running stability is the ability of a dynamic system to compensate for small perturbations during running. While the immediate effects of footwear on running biomechanics are frequently investigated, no research has studied the long-term effects of barefoot vs. shod running on local dynamic running stability. In this randomized single-blinded controlled trial, young adults novice to barefoot running were randomly allocated to a barefoot or a cushioned footwear running group. Over an 8-week-period, both groups performed a weekly 15-min treadmill running intervention in the allocated condition at 70% of their VO2 max velocity. During each session, an inertial measurement unit on the tibia recorded kinematic data (angular velocity) which was used to determine the short-time largest Lyapunov exponents as a measure of local dynamic running stability. One hundred running gait cycles at the beginning, middle, and end of each running session were analysed using one mixed linear multilevel random intercept model. Of the 41 included participants (48.8% females), 37 completed the study (drop-out = 9.7%). Participants in the barefoot running group exhibited lower running stability than in the shod running group (p = 0.037) with no changes during the intervention period (p = 0.997). Within a single session, running stability decreased over the course of the 15-min run (p = 0.012) without differences between both groups (p = 0.060). Changing from shod to barefoot running reduces running stability not only in the acute phase but also in the longer term. While running stability is a relatively new concept, it enables further insight into the biomechanical influence of footwear.

Generalised M-quantile random-effects model for discrete response: An application to the number of visits to physicians.

In this paper, we extend the linear M-quantile random intercept model (MQRE) to discrete data and use the proposed model to evaluate the effect of selected covariates on two count responses: the number of generic medical examinations and the number of specialised examinations for health districts in three regions of central Italy. The new approach represents an outlier-robust alternative to the generalised linear mixed model with Gaussian random effects and it allows estimating the effect of the covariates at various quantiles of the conditional distribution of the target variable. Results from a simulation experiment, as well as from real data, confirm that the method proposed here presents good robustness properties and can be in certain cases more efficient than otherapproaches.

Bisphenol A exposure and reduced male sex hormone levels in a Norwegian population

Background: Bisphenol A (BPA) is used in production of polycarbonate plastics and epoxy resins. Despite endocrine disrupting properties and suspected impact on male reproductive health BPA is still frequently used in a wide variety of consumer products. Few studies have addressed BPA exposure and steroid hormones in a general population sample. We aimed to investigate whether BPA exposure is associated with steroid hormone concentrations (17β-estradiol, estrone, estrone 3-sulfate, progesterone, progesterone 17α-OH-progesterone, testosterone, androstenedione, dehydroepiandrosterone sulfate (DHEAS), aldosterone, cortisone and cortisol) in a healthy Norwegian population. Methods: The study included 205 women and 232 men aged 18-48 years, participating in the RHINESSA study (2014-15) in Norway. Urine concentration of BPA and steroid hormones in serum were measured using liquid chromatography - mass spectrometry. We applied linear mixed effects random intercept models with sex hormone levels (one at a time) as outcome, BPA concentration as exposure and adjusted for age, sex, BMI, smoking and time of day for urine and blood sampling. Results: Bisphenol A was detected in 96% of the urine samples; and median concentration was 1.32 ng/mL (5th-95th percentile: 0.50-5.15, max: 21.9 ng/mL). Increasing exposure to BPA (per ng/mL) was associated with decreasing levels of testosterone (-11.72 pmol/cL, 95% Confidence Interval (CI): -22.93, -0.51) and DHEAS (-6.83 nmol/cL, 95% CI: -12.72, -0.94) for men and women combined. Restricting the analyses to men, increasing exposure to BPA was associated with decreasing levels of 17β-estradiol (-0.88 pmol/L, 95% CI: -1.72, -0.04) and with elevated endogenous aldosterone levels (4.32 pmol/L, 95% CI: 0.91, 7.73). Conclusions: The urinary biomarker concentration of BPA is ubiquitous, but relatively low. BPA was associated with an altered hormone profile, in particular with reduced levels of androgens as well as with 17β-estradiol which is synthesized from androgen precursors. Whether the association is of clinical relevance needs to be elucidated.

An alternative to confidence intervals constructed after a Hausman pretest in panel data

SummaryConsider panel data modelled by a linear random intercept model that includes a time‐varying covariate. Suppose that our aim is to construct a confidence interval for the slope parameter. Commonly, a Hausman pretest is used to decide whether this confidence interval is constructed using the random effects model or the fixed effects model. This post‐model‐selection confidence interval has the attractive features that it (a) is relatively short when the random effects model is correct and (b) reduces to the confidence interval based on the fixed effects model when the data and the random effects model are highly discordant. However, this confidence interval has the drawbacks that (i) its endpoints are discontinuous functions of the data and (ii) its minimum coverage can be far below its nominal coverage probability. We construct a new confidence interval that possesses these attractive features, but does not suffer from these drawbacks. This new confidence interval provides an intermediate between the post‐model‐selection confidence interval and the confidence interval obtained by always using the fixed effects model. The endpoints of the new confidence interval are smooth functions of the Hausman test statistic, whereas the endpoints of the post‐model‐selection confidence interval are discontinuous functions of this statistic.

PMS67 EXPERIENCE AND PREFERENCE-BASED EQ-5D-3L VALUE SET DERIVED FROM LONGITUDINAL DATA: RESULTS FROM THE INTERNATIONAL COSTS AND UTILITIES OF OSTEOPOROTIC FRACTURES STUDY (ICUROS)

Own health state utility values (HSUV, “experience-based utility”) is important for economic valuation and other purposes. There is only one preference-based EQ-5D value set derived from direct patient experience. Furthermore, the impact of interactions between EQ-5D dimensions on HSUV is poorly understood. Therefore we constructed a value set using direct patient experience and explored the impact of interactions between health states. The ICUROS was a multinational prospective observational study on the consequences of fragility fracture. In the study, patients who sustained a fragility fracture completed EQ-5D-3L and Time-trade-off (TTO) questionnaires for five time points: before fracture (recall), within two weeks after fracture, and at 4, 12 and 18 months after fracture. We derived a value set by fitting a random intercept linear model with TTO as the dependent variable and ‘moderate’ (Level 2, “L2”) or ‘severe’ impairment (Level 3, “L3”) in each EQ-5D dimension as dummy variables. We also explored the potential impact of interactions between impairment in dimensions using rigorous Lasso regression with cluster robust standard errors. In total 4,873 patients reported 17,779 EQ-5D and TTO pairs. Mean (SD) age at study enrolment was 71 (11) years and 81% were women. In the random intercept model, all coefficients had the expected sign and were statistically significant (p<0.001). The resulting value set was: 0.92 - 0.03xMobillityL2 - 0.08xMobillityL3 - 0.12xSelf-careL2 - 0.14xSelf-careL3 - 0.02xUsual ActivitiesL2 - 0.03xActivitiesL3 - 0.02xPain/DiscomfortL2 - 0.10xPain/DiscomfortL3 - 0.05xAnxiety/DepressionL2 ´- 0.11xAnxiety/DepressionL3. The Lasso regression indicated that interactions between mobility and self-care, anxiety/depression and self-care, and anxiety/depression and usual activities may be important. This experience-based EQ-5D-3L value set had consistently ordered and statistically significant coefficients, demonstrating face validity. Furthermore, the results show that explicit modelling of interactions may improve the predictive power of EQ-5D value sets. Compared to value sets based on hypothetical health, the decrements associated with impaired health were generally smaller.

Beneficial effects of acute high-intensity exercise on electrophysiological indices of attention processes in young adult men

BackgroundEmerging research suggests that a single bout of aerobic exercise can improve cognition, brain function and psychological health. Our aim was to examine the effects of high-intensity exercise on cognitive-performance and brain measures of attention, inhibition and performance-monitoring across a test-battery of three cognitive tasks. MethodUsing a randomised cross-over design, 29 young men completed three successive cognitive tasks (Cued Continuous Performance Task [CPT-OX]; Eriksen Flanker Task; four-choice reaction-time task [Fast Task]) with simultaneous electroencephalogram (EEG) recording before and after a 20-min high-intensity cycling exercise and resting control session. Cognitive-performance measures, EEG power and event-related potential measures, were obtained during the tasks. Random-intercept linear models were used to investigate the effects of exercise, compared to rest, on outcomes. ResultsA single bout of exercise significantly (p < 0.05) increased the amplitude of the event-related potential Go P3, but had no effect on the contingent negative variation (CNV), Cue P3 or NoGo P3, during the CPT-OX. Delta power, recorded during the CPT-OX, also significantly increased after exercise, whereas there was no effect on cognitive-performance in this task. Exercise did not influence any cognitive-performance or brain measures in the subsequent Flanker or Fast Tasks. ConclusionAcute high-intensity exercise improves brain-indices reflecting executive and sustained attention during task performance (Go P3 and delta activity), in the CPT-OX, but not anticipatory attention (Cue P3 and CNV) or response inhibition (NoGo P3) in young-adult men. Exercise had no effect on cognitive-performance or brain measures in the subsequent Flanker and Fast tasks, which may potentially be explained by the time delay after exercise.

More Precise Estimation of Lower-Level Interaction Effects in Multilevel Models

ABSTRACTIn hierarchical data, the effect of a lower-level predictor on a lower-level outcome may often be confounded by an (un)measured upper-level factor. When such confounding is left unaddressed, the effect of the lower-level predictor is estimated with bias. Separating this effect into a within- and between-component removes such bias in a linear random intercept model under a specific set of assumptions for the confounder. When the effect of the lower-level predictor is additionally moderated by another lower-level predictor, an interaction between both lower-level predictors is included into the model. To address unmeasured upper-level confounding, this interaction term ought to be decomposed into a within- and between-component as well. This can be achieved by first multiplying both predictors and centering that product term next, or vice versa. We show that while both approaches, on average, yield the same estimates of the interaction effect in linear models, the former decomposition is much more precise and robust against misspecification of the effects of cross-level and upper-level terms, compared to the latter.

Robust integer-valued designs for linear random intercept models

ABSTRACTThis article considers the robust design problem for linear random intercept models with both departures from fixed effects and correlated errors on a finite design space. Two strategies are proposed. One is a worst-case method minimizing the maximum value of the MSE of estimates for the fixed effects over the departure. The other is an average-case method minimizing the average value of the MSE with respect to some priors for the class of departure functions and correlation structures of random errors. Two examples are given to show robust designs for two polynomial models.