MIXREGLS: A Program for Mixed-Effects Location Scale Analysis

Relationship between tooth loss and progression of frailty: Findings from the English longitudinal study of aging

OBJECTIVE

The objective of the present study was to investigate the relationship between indicators of oral health status (number of teeth; denture use) and the progression of frailty amongst adults in England.

METHODS

The subjects were participants of the English Longitudinal Study of Aging [ELSA] aged 50 years and older. We used panel data from three waves of the study (Waves 7-9). Indicators of oral health comprised the number of teeth (≥20; 10-19; 1-9; 0) and combination of removable denture usage and the number of teeth. Frailty was assessed by the 32-item Frailty Index (FI). Covariates were age, sex, education, marital status, smoking, alcohol, and physical activity. The longitudinal relationship between oral health indicators and change in FI were investigated using linear mixed-effect models considering frailty as a time-varying variable.

RESULTS

Among the 7,557 participants, compared to those people with 20 or more teeth, change in frailty over time was significantly higher among those with less than 20 teeth: 10-19 teeth (β: 0.249, 95 %CI: 0.116 to 0.382), and 1-9 teeth (β: 0.238, 95 %CI: 0.053 to 0.423) and being edentate (β: 0.286, 95 %CI: 0.106 to 0.465) when adjusting for co-variates. The rise in frailty over time was significantly higher among those with fewer teeth (<20 teeth), including those using dentures.

CONCLUSIONS AND IMPLICATIONS

This longitudinal analysis suggests that tooth loss is associated with accelerated progression of frailty and utilizing dentures did not reduce the trend in frailty. Thus, maintaining a functional natural dentition is important in healthy ageing.

A tutorial for conducting intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (I-MAIHDA) is an innovative approach for investigating inequalities, including intersectional inequalities in health, disease, psychosocial, socioeconomic, and other outcomes. I-MAIHDA and related MAIHDA approaches have conceptual and methodological advantages over conventional single-level regression analysis. By enabling the study of inequalities produced by numerous interlocking systems of marginalization and oppression, and by addressing many of the limitations of studying interactions in conventional analyses, intersectional MAIHDA provides a valuable analytical tool in social epidemiology, health psychology, precision medicine and public health, environmental justice, and beyond. The approach allows for estimation of average differences between intersectional strata (stratum inequalities), in-depth exploration of interaction effects, as well as decomposition of the total individual variation (heterogeneity) in individual outcomes within and between strata. Specific advice for conducting and interpreting MAIHDA models has been scattered across a burgeoning literature. We consolidate this knowledge into an accessible conceptual and applied tutorial for studying both continuous and binary individual outcomes. We emphasize I-MAIHDA in our illustration, however this tutorial is also informative for understanding related approaches, such as multicategorical MAIHDA, which has been proposed for use in clinical research and beyond. The tutorial will support readers who wish to perform their own analyses and those interested in expanding their understanding of the approach. To demonstrate the methodology, we provide step-by-step analytical advice and present an illustrative health application using simulated data. We provide the data and syntax to replicate all our analyses.

Individual variation in life-history timing: synchronous presence, asynchronous events and phenological compensation in a wild mammal

Many animals and plants have species-typical annual cycles, but individuals vary in their timing of life-history events. Individual variation in fur replacement (moult) timing is poorly understood in mammals due to the challenge of repeated observations and longitudinal sampling. We examined factors that influence variation in moult duration and timing among elephant seals (Mirounga angustirostris). We quantified the onset and progression of fur loss in 1178 individuals. We found that an exceptionally rapid visible moult (7 days, the shortest of any mammals or birds), and a wide range of moult start dates (spanning 6-10× the event duration) facilitated high asynchrony across individuals (only 20% of individuals in the population moulting at the same time). Some of the variation was due to reproductive state, as reproductively mature females that skipped a breeding season moulted a week earlier than reproductive females. Moreover, individual variation in timing and duration within age-sex categories far outweighed (76-80%) variation among age-sex categories. Individuals arriving at the end of the moult season spent 50% less time on the beach, which allowed them to catch up in their annual cycles and reduce population-level variance during breeding. These findings underscore the importance of individual variation in annual cycles.

First-interview response patterns of intensive longitudinal psychological and health data

Self-report data are essential in health psychology research where an individual's perception is critical to understanding one's health and psychological status. Intensive data collection over time, including daily diary assessments, is necessary in understanding within- and between-person variability in health and psychological processes over time. An "initial elevation or latent decline" (IELD) effect, inherent of self-report data, is increasingly acknowledged in the social psychology literature, but awareness of this effect in health psychology research is lacking, particularly in studies that emphasize within- and between-person variability in self-reports. The IELD effect is a pattern in which responses tend to be more extreme at the initial interview relative to subsequent responses. This paper illustrates the impact of IELD in applications of mixed-effects models based on observational self-reports and concludes that researchers take such effects into account in data analysis or in the research designing phase to help mitigate such effects.

Alternative covariance structures in mixed-effects models: Addressing intra- and inter-individual heterogeneity

Mixed-effects models for repeated measures and longitudinal data include random coefficients that are unique to the individual, and thus permit subject-specific growth trajectories, as well as direct study of how the coefficients of a growth function vary as a function of covariates. Although applications of these models often assume homogeneity of the within-subject residual variance that characterizes within-person variation after accounting for systematic change and the variances of the random coefficients of a growth model that quantify individual differences in aspects of change, alternative covariance structures can be considered. These include allowing for serial correlations between the within-subject residuals to account for dependencies in data that remain after fitting a particular growth model or specifying the within-subject residual variance to be a function of covariates or a random subject effect to address between-subject heterogeneity due to unmeasured influences. Further, the variances of the random coefficients can be functions of covariates to relax the assumption that these variances are constant across subjects and to allow for the study of determinants of these sources of variation. In this paper, we consider combinations of these structures that permit flexibility in how mixed-effects models are specified to understand within- and between-subject variation in repeated measures and longitudinal data. Data from three learning studies are analyzed using these different specifications of mixed-effects models.

Prenatal Urban Environment and Blood Pressure Trajectories From Childhood to Early Adulthood

Background

Prenatal urban environmental exposures have been associated with blood pressure in children. The dynamic of these associations across childhood and later ages is unknown.

Objectives

The purpose of this study was to assess associations of prenatal urban environmental exposures with blood pressure trajectories from childhood to early adulthood.

Methods

Repeated measures of systolic blood pressure (SBP) and diastolic blood pressure (DBP) were collected in up to 7,454 participants from a UK birth cohort. Prenatal urban exposures (n = 43) covered measures of noise, air pollution, built environment, natural spaces, traffic, meteorology, and food environment. An exposome-wide association study approach was used. Linear spline mixed-effects models were used to model associations of each exposure with trajectories of blood pressure. Replication was sought in 4 independent European cohorts (up to 9,261).

Results

In discovery analyses, higher humidity was associated with a faster increase (mean yearly change in SBP for an interquartile range increase in humidity: 0.29 mm Hg/y, 95% CI: 0.20-0.39) and higher temperature with a slower increase (mean yearly change in SBP per interquartile range increase in temperature: -0.17 mm Hg/y, 95% CI: -0.28 to -0.07) in SBP in childhood. Higher levels of humidity and air pollution were associated with faster increase in DBP in childhood and slower increase in adolescence. There was little evidence of an association of other exposures with change in SBP or DBP. Results for humidity and temperature, but not for air pollution, were replicated in other cohorts.

Conclusions

Replicated findings suggest that higher prenatal humidity and temperature could modulate blood pressure changes across childhood.

A latent variable mixed-effects location scale model that also considers between-person differences in the autocorrelation

In public health research an increasing number of studies is conducted in which intensive longitudinal data is collected in an experience sampling or a daily diary design. Typically, the resulting data is analyzed with a mixed-effects model or mixed-effects location scale model because they allow one to examine a host of interesting longitudinal research questions. Here, we introduce an extension of the mixed-effects location scale model in which measurement error of the observed variables is considered by a latent factor model and in which-in addition to the mean-or location-related effects-the residual variance of the latent factor and the parameters of the autoregressive process of this latent factor can differ between persons. We show how to estimate the parameters of the model with a maximum likelihood approach, whose performance is also compared with a Bayesian approach in a small simulation study. We illustrate the models using a real data example and end with a discussion in which we suggest questions for future research.

A Mixed-Effects Model in Which the Parameters of the Autocorrelated Error Structure Can Differ between Individuals

Research in psychology has seen a rapid increase in the usage of experience sampling methods and daily diary methods. The data that result from using these methods are typically analyzed with a mixed-effects or a multilevel model because it allows testing hypotheses about the time course of the longitudinally assessed variable or the influence of time-varying predictors in a simple way. Here, we describe an extension of this model that does not only allow to include random effects for the mean structure but also for the residual variance, for the parameter of an autoregressive process of order 1 and/or the parameter of a moving average process of order 1. After we have introduced this extension, we show how to estimate the parameters with maximum likelihood. Because the likelihood function contains complex integrals, we suggest using adaptive Gauss-Hermite quadrature and Quasi-Monte Carlo integration to approximate it. We illustrate the models using a real data example and also report the results of a small simulation study in which the two integral approximation methods are compared.

Uric acid mediated the relationship between obesity and hypertension in children and adolescents: A population‑based cohort study

BACKGROUND AND AIM

Obesity and hyperuricemia (HUA) often coexist and have been widely accepted as risk factors for hypertension, but the role of uric acid (UA) in the relationship between obesity and hypertension remains unknown in children and adolescents.

METHODS AND RESULTS

A total of 7525 subjects aged 6-16 years were from the School-based Cardiovascular and Bone Health Promotion Program (SCVBH) at baseline (2017) and followed up in 2019. Multivariable logistic regression with interaction terms, cross-lagged panel analysis, and causal mediation model were applied to delineate the joint impact of obesity and HUA on hypertension, including the interaction effect, the temporal association, and the mediating effect of UA in the relationship between obesity and hypertension. There were 10.8 % of the participants with normotension at baseline developed hypertension after two years of follow-up. Cross-lagged panel analysis showed that the two-time point association was significant only from baseline BMI to follow-up UA (β1 = 0.302, P < 0.001), but not from baseline UA to follow-up BMI (β2 = 0.002, P = 0.745). Multivariable logistic regression showed that both obesity and HUA increased the risk of hypertension, but no interaction effect between HUA and obesity. The causal mediation analysis found that UA partially mediated the association between BMI and SBP (mediate proportion: 20.3 %, 95 % CI: 17.4-22.9 %) or DBP (mediate proportion: 11.9 %, 95 % CI: 3.9-18.2 %). The results were consistent in the analysis of systolic hypertension rather than diastolic hypertension.

CONCLUSIONS

It is mediating effect that UA played in the progress from obesity to hypertension, particularly systolic hypertension in children and adolescents.

Daily mindfulness, negative affect, and eating behaviors in adolescents at risk for excess weight gain

OBJECTIVE

Loss-of-control and overeating are common in adolescents with high body mass index (BMI). Mindfulness may affect negative affect, and both may relate to loss-of-control and overeating. Yet, there is limited understanding of these associations in adolescents' daily lives.

METHODS

Forty-five adolescents (77% female; Mage  = 14.4 years, SDage  = 1.7 years) with high weight (92% with BMI [kg/m2 ] ≥85th percentile for age/sex) provided daily, repeated measurements of mindfulness, negative affect, loss-of-control, and overeating for ~7 days (M = 5.6 days; range = 1-13). Multilevel mixed modeling was conducted to test within-person (intraindividual) and between-person (interindividual) associations for the same-day (concurrent) and next-day (time-ordered/prospective).

RESULTS

There were within-person and between-person associations of higher mindfulness with lower negative affect on the same-day and next-day. Greater between-person mindfulness related to lower odds of adolescents' loss-of-control occurrence (same-day) and conversely, more perceived control over eating (same-day and next-day). Greater within-person mindfulness related to less odds of next-day overeating.

DISCUSSION

Dynamic relations exist among mindfulness, negative affect, and eating in adolescents at-risk for excess weight gain. Mindfulness may be an important element to consider in loss-of-control and overeating. Future work using momentary-data within an experimental design would help disentangle the intraindividual effects of increasing mindfulness/decreasing negative affect on disordered eating.

PUBLIC SIGNIFICANCE

Loss-of-control and overeating are common in teenagers with high weight. Greater mindfulness-present-moment, non-judgmental attention-and less negative emotions may relate to healthier eating, but we do not know how these processes play out in teenagers' daily lives. Addressing this knowledge gap, the current findings showed that greater daily mindfulness, but not negative affect, related to less loss-of-control/overeating, suggesting the importance of mindfulness for eating patterns in teenagers' daily lives.

Socioeconomic conditions and children's mental health and quality of life during the COVID-19 pandemic: An intersectional analysis

BACKGROUND

Children and adolescents are highly vulnerable to the impact of sustained stressors during developmentally sensitive times. We investigated how demographic characteristics intersect with socioeconomic dimensions to shape the social patterning of quality of life and mental health in children and adolescents, two years into the COVID-19 pandemic.

METHODS

We used data from the prospective SEROCoV-KIDS cohort study of children and adolescents living in Geneva (Switzerland, 2022). We conducted an intersectional Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy by nesting participants within 48 social strata defined by intersecting sex, age, immigrant background, parental education and financial hardship in Bayesian multilevel logistic models for poor health-related quality of life (HRQoL, measured with PedsQL) and mental health difficulties (measured with the Strengths and Difficulties Questionnaire).

RESULTS

Among participants aged 2-17 years, 240/2096 (11.5%, 95%CI 10.1-12.9) had poor HRQoL and 105/2135 (4.9%, 95%CI 4.0-5.9) had mental health difficulties. The predicted proportion of poor HRQoL ranged from 3.4% for 6-11 years old Swiss girls with highly educated parents and no financial hardship to 34.6% for 12-17 years old non-Swiss girls with highly educated parents and financial hardship. Intersectional strata involving adolescents and financial hardship showed substantially worse HRQoL than their counterparts. Between-stratum variations in the predicted frequency of mental health difficulties were limited (range 4.4%-6.5%).

CONCLUSIONS

We found considerable differences in adverse outcomes across social strata. Our results suggest that, post-pandemic, interventions to address social inequities in HRQoL should focus on specific intersectional strata involving adolescents and families experiencing financial hardship, while those aiming to improve mental health should target all children and adolescents.

Studying Movement-Related Behavioral Maintenance and Adoption in Real Time: Protocol for an Intensive Ecological Momentary Assessment Study Among Older Adults

BACKGROUND

Older adults struggle to maintain newly initiated levels of physical activity (PA) or sedentary behavior (SB) and often regress to baseline levels over time. This is partly because health behavior theories that inform interventions rarely address how the changing contexts of daily life influence the processes regulating PA and SB or how those processes differ across the behavior change continuum. Few studies have focused on motivational processes that regulate the dynamic nature of PA and SB adoption and maintenance on microtimescales (ie, across minutes, hours, or days).

OBJECTIVE

The overarching goal of Project Studying Maintenance and Adoption in Real Time (SMART) is to determine the motivational processes that regulate behavioral adoption versus maintenance over microtimescales, using a dual process framework combined with ecological momentary assessment and sensor-based monitoring of behavior. This paper describes the recruitment, enrollment, data collection, and analytics protocols for Project SMART.

METHODS

In Project SMART, older adults engaging in at least 30 minutes of moderate-to-vigorous intensity PA per week complete 3 data collection periods over 1 year, with each data collection period lasting 14 days. Across each data collection period, participants wear an ActiGraph GT3X accelerometer (ActiGraph, LLC) on their nondominant waist and an ActivPAL micro4 accelerometer (PAL Technologies, Ltd) on their anterior thigh to measure PA and SB, respectively. Ecological momentary assessment questionnaires are randomly delivered via smartphone 10 times per day on 4 selected days in each data collection period and assess reflective processes (eg, evaluating one's efficacy and exerting self-control) and reactive processes (eg, contextual cues) within the dual process framework. At the beginning and end of each data collection period, participants complete a computer-based questionnaire to learn more about their typical motivation for PA and SB, physical and mental health, and life events over the course of the study.

RESULTS

Recruitment and enrollment began in January 2021; enrollment in the first data collection period was completed by February 2022; and all participants completed their second and third data collection by July 2022 and December 2022, respectively. Data were collected from 202 older adults during the first data collection period, with approximate retention rates of 90.1% (n=182) during the second data collection period and 88.1% (n=178) during the third data collection period. Multilevel models and mixed-effects location scale modeling will be used to evaluate the study aims.

CONCLUSIONS

Project SMART seeks to predict and model the adoption and maintenance of optimal levels of PA and SB among older adults. In turn, this will inform the future delivery of personalized intervention content under conditions where the content will be most effective to promote sustained behavior change among older adults.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/47320.

Longitudinal Effects of Sex, Aging, and Multiple Sclerosis Diagnosis on Function

BACKGROUND

A gap in research about the trajectories of function among men and women aging with functional limitations because of multiple sclerosis (MS) hinders ability to plan for future needs.

OBJECTIVES

Using a biopsychosocial model, we characterize how men and women with MS report changes over time in their function and test how person-level differences in age, diagnosis duration, and sex influence perceived function.

METHODS

A longitudinal study with multiple waves of surveys was used to collect data on participant perceptions of function, as well as demographic and contextual variables. Self-reported functional limitation was measured over a decade. The study participants were community residing with physician-diagnosed MS.

RESULTS

The people with MS had a diagnosis duration of about 13 years and were around 51 years of age, on average, at the start of the study. They were primarily women and non-Hispanic White. We analyzed the data using mixed-effects models. Subject-specific, functional limitation trajectories were described best with a quadratic growth model. Relative to men, women reported lower functional limitation and greater between-person variation and rates of acceleration in functional limitation scores.

DISCUSSION

Results suggest function progressed through two pathways for over a decade, particularly closer to diagnoses. Variability in trajectories between individuals based on sex and years since diagnosis of disease indicates that men and women with MS may experience perceptions of their function with age differently. This has implications for clinician advice to men and women with MS.

Fast estimation of mixed-effects location-scale regression models

As a result of advances in data collection technology and study design, modern longitudinal datasets can be much larger than they historically have been. Such "intensive" longitudinal datasets are rich enough to allow for detailed modeling of the variance of a response as well as the mean, and a flexible class of models called mixed-effects location-scale (MELS) regression models are commonly used to do so. However, fitting MELS models can pose computational challenges related to the numerical evaluation of multi-dimensional integrals; the slow runtime of current methods is inconvenient for data analysis and makes bootstrap inference impractical. In this paper, we introduce a new fitting technique, called FastRegLS, that is considerably faster than existing techniques while still providing consistent estimators for the model parameters.

WiSER: Robust and scalable estimation and inference of within-subject variances from intensive longitudinal data

The availability of vast amounts of longitudinal data from electronic health records (EHRs) and personal wearable devices opens the door to numerous new research questions. In many studies, individual variability of a longitudinal outcome is as important as the mean. Blood pressure fluctuations, glycemic variations, and mood swings are prime examples where it is critical to identify factors that affect the within-individual variability. We propose a scalable method, within-subject variance estimator by robust regression (WiSER), for the estimation and inference of the effects of both time-varying and time-invariant predictors on within-subject variance. It is robust against the misspecification of the conditional distribution of responses or the distribution of random effects. It shows similar performance as the correctly specified likelihood methods but is 103 ∼ 105 times faster. The estimation algorithm scales linearly in the total number of observations, making it applicable to massive longitudinal data sets. The effectiveness of WiSER is evaluated in extensive simulation studies. Its broad applicability is illustrated using the accelerometry data from the Women's Health Study and a clinical trial for longitudinal diabetes care.

Individual and contextual factors associated with childhood malnutrition: a multilevel analysis of the double burden of childhood malnutrition in 27 countries

BACKGROUND

Malnutrition is a key global health challenge and a major contributor to childhood morbidity and mortality. In recent times, the contrasting coexistence of undernutrition including micronutrient deficiencies and overweight/obesity called double burden of malnutrition has been noted at individual, household or population level and/or at different times in life. The objective of this study was to examine individual, neighborhood and country level factors that are associated with the double burden of childhood malnutrition.

METHODS

We conducted multivariable multilevel logistic regression analyses on the most recent demographic and health datasets from surveys conducted between 2015 and 2020 in low- and middle-income countries. We analyzed data of 138,782 children (level 1) living in 13,788 communities (level 2) from 27 countries (level 3).

RESULTS

The results of our analysis show variation in childhood malnutrition across the 27 countries from as low as 6.5% in Burundi to as high as 29.5% in Timor Leste. After adjusting for all level factors, we found that those who were wasted/overweight tended to have had an episode of diarrhea or fever in the last two weeks preceding the survey, were part of a multiple birth, were being breastfed at the time of the survey and born to mothers with more than one under 5-child resident in neighborhoods with high illiteracy and unemployment rates. The intra-neighbourhood and intra-country correlation coefficients were estimated using the intercept component variance; 44.3% and 21.0% of variance in odds of double burden of childhood malnutrition are consequent upon neighborhood and country level factors respectively.

CONCLUSIONS

Evidence of geographical clustering in childhood malnutrition at community and country levels was found in our study with variability due to neighborhood level factors twice that of country level factors. Therefore, strategies in tackling the double burden of malnutrition must consider these shared drivers, contextual barriers and geographical clustering effects.

Menstrual hygiene practices among adolescent women in rural India: a cross-sectional study

Background

Exclusive use of hygienic methods (sanitary napkins, locally prepared napkins, tampons, and menstrual cups) to prevent the visibility of bloodstains during menstruation is still considerably low among adolescent women in rural India. However, no prior research has explored the prevalence and determinants of exclusive hygienic methods among rural Indian adolescent women. To address this gap, this study examines the factors affecting adolescent women’s exclusive use of hygienic methods in rural India. Additionally, this study explores state- and district-level geographical disparities in the exclusive use of hygienic methods among adolescent women in rural India.

Methods

Information on 95,551 adolescent women from rural India from the latest round of National Family Health Survey (NFHS-5) was analyzed. Bivariate statistics and multilevel logistic regression analysis were used to assess the Individual- and community-level factors associated with exclusive use of hygienic methods among adolescent women in rural India. Choropleth maps were used to discern the geographical disparities in the exclusive use of hygienic methods.

Results

In rural India, only 42% of adolescent women exclusively used hygienic methods, with substantial geographic disparities at the state and district levels. At the state level, the exclusive use of hygienic methods varied from 23% in Uttar Pradesh to 85% in Tamil Nadu. Even greater variation was observed at the district level. There was a clear north-south divide in the exclusive use of hygienic methods among adolescent women in rural India. The results of multilevel logistic regression indicated a considerable amount of variation in the exclusive use of hygienic methods at community level which further reduced when controlled for individual and community-level factors. Rural Indian adolescent women with higher education (AOR:3.20, 95% CI: 2.81–3.64), from general category (AOR: 1.14, 95% CI: 1.07–1.21), with medium mass media exposure (AOR: 1.43, 95% CI: 1.35–1.51), and from richest wealth quintile (AOR: 3.98, 95% CI: 3.69–4.30) were more likely to use hygienic methods exclusively.

Conclusion

Wide differential across biodemographic and socioeconomic groups, and substantial geographic disparities at state- and district-level in the exclusive use of hygienic methods suggests a need to adopt context-specific interventions for adolescent women in rural India. Distribution of subsidized or free menstrual hygiene methods to disadvantaged adolescent women, and in the low-prevalence districts may increase the level of exclusive use of hygienic methods remarkably.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-022-14622-7.

Examining whether affectively-charged motivations predict subsequent affective response during physical activity: An ecological momentary assessment study

BACKGROUND

Evidence suggests positive affective response during physical activity increases the likelihood of engaging in and maintaining regular activity exercise in the future. Elucidating antecedents for a positive affective response may help identify intervention strategies to increase activity. Affectively-charged motivations (e.g., desires, urges, dreading) have been posited as proximal antecedents to physical activity but have yet to be examined in terms of their influence on affective response in real-world settings. The current study used ecological momentary assessment (EMA) to examine within-subject effects of pre-physical activity affectively-charged motivation on subsequent affective response during physical activity.

METHODS

Participants included 56 adults (M = 39.18 years, SD = 11.98; 67.86% female) who completed a 14-day smartphone-based EMA study. Prior to starting physical activity (time t), participants self-initiated an event-contingent EMA survey that assessed affectively-charged motivation for physical activity (i.e., rating scale from "dreading it" to "excited to do it"). EMA surveys prompted during subsequent physical activity (time t + 15 min) assessed affective response (i.e., feeling good-bad, energized-exhausted, thrilled-miserable, interested-bored, and relaxed-nervous). Multi-level linear regression models examined within-subject effects of pre- physical activity affectively-charged motivations on subsequent affective response during physical activity controlling for between-subjects effects of affectively-charged motivation, age, biological sex, time of day, and day of the week.

RESULTS

Overall, there were N = 304 physical activity occasions in the analysis (M = 5.43, SD = 3.97). When individuals reported more positive affectively-charged motivation for physical activity than usual before physical activity occasions, they reported feeling more energized (Estimate = 0.22, p < 0.001), good (Estimate = 0.25, p < 0.001), thrilled (Estimate = 0.12, p = 0.02), and interested (Estimate = 0.24, p < 0.001) during subsequent physical activity. Affectively-charged motivation was not associated with feeling more relaxed (Estimate = 0.11, p = 0.13) during subsequent physical activity.

CONCLUSION

Momentary affectively-charged motivations predicted more positive affective response during subsequent physical activity among active adults. Future research can explore potential sources of intraindividual differences in affectively-charged motivations and further examine these associations with future physical activity behavior. To improve positive affective responses, interventions may boost affectively-charged motivations through real-time mobile prompting in naturalistic settings.

Within-Person variations in mindfulness mediate effects of daily stressors on psychological distress in adolescence

OBJECTIVE

We tested two competing models linking daily stress, mindfulness, and psychological distress in adolescence: 1) whether daily mindfulness moderates the impact of daily stressors on psychological distress or 2) whether mindfulness mediates the relationship between greater daily stressors and psychological distress.

METHODS

Every evening for a week, 138 adolescents completed ecological momentary assessments (EMAs). Daily diaries assessed negative events, work-school conflict, mindfulness, and perceived stress. Multilevel mediation and moderation were tested.

RESULTS

Results indicated that there were meaningful variations in adolescent mindfulness from day-to-day, and supported mediation rather than moderation; the within-person association between stressors and psychological distress was mediated by mindfulness at the daily level.

CONCLUSION

It may be too challenging for adolescents to remain in a mindful state during stress to effectively use mindfulness as a buffer. Instead, daily stressors may indirectly impact psychological distress through decreasing mindfulness.

Health and labor force participation among older workers in Switzerland: a growth curve analysis

This study investigated how individual trajectories of self-rated health (SRH) and working hours among older workers in Switzerland are interrelated and how this relationship varies based on occupation. We used data from the Swiss Household Panel to analyze the long-term trajectories of older workers measured in terms of working hours and SRH. The sample included more than 4000 workers aged 50 to 65(men)/64(women). We ran a bivariate response multilevel model for growth that allowed the examination of between- and within-individual changes over time. On a between-individual level, we found that the upper non-manual workers were the most heterogeneous occupational group in terms of working hours and the lower non-manual workers were the most heterogeneous occupational group in terms of health. Within all occupational groups, we found a significant relationship between the level of working hours and level of SRH. The individual-level statistics showed consistently strongest effects for manual workers. This result confirms our hypothesis that labor force participation in individuals of the manual occupational group is more sensitive to their health status. Our findings contribute to the debate regarding the importance of older workers' health in the context of the extension of working life.

A Lasso and a Regression Tree Mixed-Effect Model with Random Effects for the Level, the Residual Variance, and the Autocorrelation

Research in psychology is experiencing a rapid increase in the availability of intensive longitudinal data. To use such data for predicting feelings, beliefs, and behavior, recent methodological work suggested combinations of the longitudinal mixed-effect model with Lasso regression or with regression trees. The present article adds to this literature by suggesting an extension of these models that-in addition to a random effect for the mean level-also includes a random effect for the within-subject variance and a random effect for the autocorrelation. After introducing the extended mixed-effect location scale (E-MELS), the extended mixed-effect location-scale Lasso model (Lasso E-MELS), and the extended mixed-effect location-scale tree model (E-MELS trees), we show how its parameters can be estimated using a marginal maximum likelihood approach. Using real and simulated example data, we illustrate how to use E-MELS, Lasso E-MELS, and E-MELS trees for building prediction models to forecast individuals' daily nervousness. The article is accompanied by an R package (called mels) and functions that support users in the application of the suggested models.

Intensive Longitudinal Data Collection Using Microinteraction Ecological Momentary Assessment: Pilot and Preliminary Results

Background

Ecological momentary assessment (EMA) uses mobile technology to enable in situ self-report data collection on behaviors and states. In a typical EMA study, participants are prompted several times a day to answer sets of multiple-choice questions. Although the repeated nature of EMA reduces recall bias, it may induce participation burden. There is a need to explore complementary approaches to collecting in situ self-report data that are less burdensome yet provide comprehensive information on an individual’s behaviors and states. A new approach, microinteraction EMA (μEMA), restricts EMA items to single, cognitively simple questions answered on a smartwatch with single-tap assessments using a quick, glanceable microinteraction. However, the viability of using μEMA to capture behaviors and states in a large-scale longitudinal study has not yet been demonstrated.

Objective

This paper describes the μEMA protocol currently used in the Temporal Influences on Movement & Exercise (TIME) Study conducted with young adults, the interface of the μEMA app used to gather self-report responses on a smartwatch, qualitative feedback from participants after a pilot study of the μEMA app, changes made to the main TIME Study μEMA protocol and app based on the pilot feedback, and preliminary μEMA results from a subset of active participants in the TIME Study.

Methods

The TIME Study involves data collection on behaviors and states from 246 individuals; measurements include passive sensing from a smartwatch and smartphone and intensive smartphone-based hourly EMA, with 4-day EMA bursts every 2 weeks. Every day, participants also answer a nightly EMA survey. On non–EMA burst days, participants answer μEMA questions on the smartwatch, assessing momentary states such as physical activity, sedentary behavior, and affect. At the end of the study, participants describe their experience with EMA and μEMA in a semistructured interview. A pilot study was used to test and refine the μEMA protocol before the main study.

Results

Changes made to the μEMA study protocol based on pilot feedback included adjusting the single-question selection method and smartwatch vibrotactile prompting. We also added sensor-triggered questions for physical activity and sedentary behavior. As of June 2021, a total of 81 participants had completed at least 6 months of data collection in the main study. For 662,397 μEMA questions delivered, the compliance rate was 67.6% (SD 24.4%) and the completion rate was 79% (SD 22.2%).

Conclusions

The TIME Study provides opportunities to explore a novel approach for collecting temporally dense intensive longitudinal self-report data in a sustainable manner. Data suggest that μEMA may be valuable for understanding behaviors and states at the individual level, thus possibly supporting future longitudinal interventions that require within-day, temporally dense self-report data as people go about their lives.

Catching Up on Multilevel Modeling

This review focuses on the use of multilevel models in psychology and other social sciences. We target readers who are catching up on current best practices and sources of controversy in the specification of multilevel models. We first describe common use cases for clustered, longitudinal, and cross-classified designs, as well as their combinations. Using examples from both clustered and longitudinal designs, we then address issues of centering for observed predictor variables: its use in creating interpretable fixed and random effects of predictors, its relationship to endogeneity problems (correlations between predictors and model error terms), and its translation into multivariate multilevel models (using latent-centering within multilevel structural equation models). Finally, we describe novel extensions—mixed-effects location–scale models—designed for predicting differential amounts of variability.

An extension of the mixed-effects growth model that considers between-person differences in the within-subject variance and the autocorrelation

Experience sampling methods have led to a significant increase in the availability of intensive longitudinal data. Typically, this type of data is analyzed with a mixed-effects model that allows to examine hypotheses concerning between-person differences in the mean structure by including multiple random effects per individual (eg, random intercept and random slopes). Here, we describe an extension of this model that-in addition to the random effects for the mean structure-also includes a random effect for the within-subject variance and a random effect for the autocorrelation. After the description of the model, we show how its parameters can be efficiently estimated using a marginal maximum likelihood (ML) approach. We then illustrate the model using a real data example. We also present the results of a small simulation study in which we compare the ML approach with a Bayesian estimation approach.

Affective variability and childhood abuse increase the risk for nonsuicidal self-injury following psychiatric hospitalization

Childhood abuse is associated with myriad negative behavioral health outcomes, including nonsuicidal self-injury (NSSI). Models aimed at understanding the mechanism whereby childhood abuse may exert negative effects frequently highlight alterations or dysregulations in experienced affect, particularly during times of stress. The period immediately following discharge from psychiatric hospitalization represents a time of risk for individuals experiencing NSSI. We aimed to investigate the extent to which childhood abuse predicts affective variability and examine whether affective variability predicts who will experience NSSI thoughts or behaviors in the 6 months following hospital discharge. Participants were adults (N = 133) engaged in a 3-week ecological momentary assessment study immediately following psychiatric hospitalization who returned for a 6-month follow-up. Location scale modeling (LSM) was used to model the impact of childhood abuse on affect variability; estimates of affect variability were then resampled and examined as risk factors for NSSI at follow-up. Overall, individuals who experienced childhood abuse reported lower positive affect intensity. Participants exposed to emotional and sexual abuse reported less between-person variability in positive affect, and those who experienced sexual abuse evidenced higher between-person variability in negative affect. Models indicated that higher overall negative affect was associated with a higher NSSI risk, ORs ∼ 1.82-2.10 even after accounting for childhood abuse and lifetime NSSI. These findings indicate the importance of in vivo affect as a critical influence on behavior during high-risk periods, reinforcing the need to move beyond traditional cross-sectional approaches of assessment, analysis, and treatment.

Digitally characterizing the dynamics of multiple health behavior change

OBJECTIVE

We applied the ORBIT model to digitally define dynamic treatment pathways whereby intervention improves multiple risk behaviors. We hypothesized that effective intervention improves the frequency and consistency of targeted health behaviors and that both correlate with automaticity (habit) and self-efficacy (self-regulation).

METHOD

Study 1: Via location scale mixed modeling we compared effects when hybrid mobile intervention did versus did not target each behavior in the Make Better Choices 1 (MBC1) trial (n = 204). Participants had all of four risk behaviors: low moderate-vigorous physical activity (MVPA) and fruit and vegetable consumption (FV), and high saturated fat (FAT) and sedentary leisure screen time (SED). Models estimated the mean (location), between-subjects variance, and within-subject variance (scale).

RESULTS

Treatment by time interactions showed that location increased for MVPA and FV (Bs = 1.68, .61; ps < .001) and decreased for SED and FAT (Bs = -2.01, -.07; ps < .05) more when treatments targeted the behavior. Within-subject variance modeling revealed group by time interactions for scale (taus = -.19, -.75, -.17, -.11; ps < .001), indicating that all behaviors grew more consistent when targeted.

METHOD

Study 2: In the MBC2 trial (n = 212) we examined correlations between location, scale, self-efficacy, and automaticity for the three targeted behaviors.

RESULTS

For SED, higher scale (less consistency) but not location correlated with lower self-efficacy (r = -.22, p = .014) and automaticity (r = -.23, p = .013). For FV and MVPA, higher location, but not scale, correlated with higher self-efficacy (rs = .38, .34, ps < .001) and greater automaticity (rs = .46, .42, ps < .001).

CONCLUSIONS

Location scale mixed modeling suggests that both habit and self-regulation changes probably accompany acquisition of complex diet and activity behaviors. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Variability in Hourly Activity Levels: Statistical Noise or Insight Into Older Adult Frailty?

BACKGROUND

Frailty is associated with lower mean activity; however, hourly activity is highly variable among older individuals. We aimed to relate frailty to hourly activity variance beyond frailty's association with mean activity.

METHOD

Using the 2010-2011 National Social Life, Health and Aging Project wrist accelerometry data (n = 647), we employed a mixed-effects location scale model to simultaneously determine whether an adapted phenotypic frailty scale (0-4) was associated with the log10-mean hourly counts per minute (cpm) and between-and within-subject hourly activity variability, adjusting for demographics, health characteristics, season, day-of-week, and time-of-day. We tested the significance of a Frailty × Time-of-day interaction and whether adjusting for sleep time altered relationships.

RESULTS

Each additional frailty point was associated with a 7.6% (10-0.0343, β = -0.0343; 95% CI: -0.05, -0.02) lower mean hourly cpm in the morning, mid-day, and late afternoon but not evening. Each frailty point was also associated with a 24.5% (e0.219, β = 0.219; 95% CI: 0.09, 0.34) greater between-subject hourly activity variance across the day; a 7% (e0.07, β = 0.07; 95% CI: 0.01¸ 0.13), 6% (e0.06, β = 0.06; 95% CI: 0, 0.12), and 10% (e0.091, β = 0.091; 95% CI: 0.03, 0.15) greater within-subject hourly activity variance in the morning, mid-day, and late afternoon, respectively; and a 6% (e-0.06, β = -0.06; 95% CI: -0.12, -0.003) lower within-subject hourly activity variance in the evening. Adjusting for sleep time did not alter results.

CONCLUSIONS

Frail adults have more variable hourly activity levels than robust adults, a potential novel marker of vulnerability. These findings suggest a need for more precise activity assessment in older adults.

Intraindividual variability in sleep schedule: effects of an internet-based cognitive-behavioral therapy for insomnia program and its relation with symptom remission

STUDY OBJECTIVES

Sleep schedule consistency is fundamental to cognitive-behavioral therapy for insomnia (CBT-I), although there is limited evidence suggesting whether it predicts treatment response. This analysis tested whether: (1) an Internet-based CBT-I program affects intraindividual variability (IIV) in sleep schedule and (2) sleep schedule IIV predicts insomnia symptom remission.

METHODS

This secondary analysis compares participants (N = 303) randomized to an Internet-based CBT-I program (SHUTi-Sleep Healthy Using the Internet) or Internet-based patient education (PE). Participants reported daily bedtimes and rising times on 10 online sleep diaries collected over 2 weeks at baseline and 9-week post-intervention assessment. Participants completed the Insomnia Severity Index (ISI) at post-assessment and 6-month follow-up; symptom remission was defined by ISI < 8. Mixed effects location scale modeling was used to examine the effect of SHUTi on bedtime and rising time IIV; a novel two-staged analysis examined the effect of bedtime and rising time IIV on insomnia symptom remission.

RESULTS

At post-assessment, SHUTi participants reported about 30% less bedtime and 32% less rising time variability compared to PE (ps < 0.03). Bedtime and rising time IIV was not independently associated with likelihood of insomnia symptom remission at the subsequent time point (ps > 0.18), nor did sleep schedule IIV moderate treatment response (ps > 0.12).

CONCLUSIONS

Findings demonstrate that an Internet-delivered CBT-I program can effectively increase users' sleep schedule consistency relative to an educational control. This consistency, however, was not related to treatment outcome when defined by insomnia symptom remission, suggesting that enforcing rigid sleep schedules for patients may not be necessary for treatment success.

CLINICAL TRIAL REGISTRATION

NCT00328250.

Feasibility of ecological momentary assessment to study depressive symptoms among cancer caregivers

OBJECTIVE

Ecological momentary assessment (EMA) may help with the development of more targeted interventions for caregivers' depression, yet the use of this method has been limited among cancer caregivers. This study aimed to demonstrate the feasibility of EMA among cancer caregivers and the use of EMA data to understand affective correlates of caregiver depressive symptoms.

METHODS

Caregivers (N = 25) completed a depressive symptom assessment (Patient Health Questionnaire-8) and then received eight EMA survey prompts per day for 7 days. EMA surveys assessed affect on the orthogonal dimensions of valence and arousal. Participants completed feedback surveys regarding the EMA protocol at the conclusion of the week-long study.

RESULTS

Of 32 caregivers approached, 25 enrolled and participated (78%), which exceeded the a priori feasibility cutoff of 55%. The prompt completion rate (59%, or 762 of 1,286 issued) did not exceed the a priori cutoff of 65%, although completion was not related to caregivers' age, employment status, physical health quality of life, caregiving stress, or depressive symptoms or the patients' care needs (ps > 0.22). Caregivers' feedback about their study experience was generally positive. Mixed-effects location scale modeling showed caregivers' higher depressive symptoms were related to overall higher reported negative affect and lower positive affect, but not to affective variability.

CONCLUSIONS

Findings from this feasibility study refute potential concerns that an EMA design is too burdensome for distressed caregivers. Clinically, findings suggest the potential importance of not only strategies to reduce overall levels of negative affect, but also to increase opportunities for positive affect.

MixWILD: A program for examining the effects of variance and slope of time-varying variables in intensive longitudinal data

The use of intensive sampling methods, such as ecological momentary assessment (EMA), is increasingly prominent in medical research. However, inferences from such data are often limited to the subject-specific mean of the outcome and between-subject variance (i.e., random intercept), despite the capability to examine within-subject variance (i.e., random scale) and associations between covariates and subject-specific mean (i.e., random slope). MixWILD (Mixed model analysis With Intensive Longitudinal Data) is statistical software that tests the effects of subject-level parameters (variance and slope) of time-varying variables, specifically in the context of studies using intensive sampling methods, such as ecological momentary assessment. MixWILD combines estimation of a stage 1 mixed-effects location-scale (MELS) model, including estimation of the subject-specific random effects, with a subsequent stage 2 linear or binary/ordinal logistic regression in which values sampled from each subject's random effect distributions can be used as regressors (and then the results are aggregated across replications). Computations within MixWILD were written in FORTRAN and use maximum likelihood estimation, utilizing both the expectation-maximization (EM) algorithm and a Newton-Raphson solution. The mean and variance of each individual's random effects used in the sampling are estimated using empirical Bayes equations. This manuscript details the underlying procedures and provides examples illustrating standalone usage and features of MixWILD and its GUI. MixWILD is generalizable to a variety of data collection strategies (i.e., EMA, sensors) as a robust and reproducible method to test predictors of variability in level 1 outcomes and the associations between subject-level parameters (variances and slopes) and level 2 outcomes.

Modelling inter-individual differences in latent within-person variation: The confirmatory factor level variability model

Psychological theories often produce hypotheses that pertain to individual differences in within-person variability. To empirically test the predictions entailed by such hypotheses with longitudinal data, researchers often use multilevel approaches that allow them to model between-person differences in the mean level of a certain variable and the residual within-person variance. Currently, these approaches can be applied only when the data stem from a single variable. However, it is common practice in psychology to assess not just a single measure but rather several measures of a construct. In this paper we describe a model in which we combine the single-indicator model with confirmatory factor analysis. The new model allows individual differences in latent mean-level factors and latent within-person variability factors to be estimated. Furthermore, we show how the model's parameters can be estimated with a maximum likelihood estimator, and we illustrate the approach using an example that involves intensive longitudinal data.

Bayesian Multivariate Mixed-Effects Location Scale Modeling of Longitudinal Relations Among Affective Traits, States, and Physical Activity

Intensive longitudinal studies and experience sampling methods are becoming more common in psychology. While they provide a unique opportunity to ask novel questions about within-person processes relating to personality, there is a lack of methods specifically built to characterize the interplay between traits and states. We thus introduce a Bayesian multivariate mixed-effects location scale model (M-MELSM). The formulation can simultaneously model both personality traits (the location) and states (the scale) for multivariate data common to personality research. Variables can be included to predict either (or both) the traits and states, in addition to estimating random effects therein. This provides correlations between location and scale random effects, both across and within each outcome, which allows for characterizing relations between any number of personality traits and the corresponding states. We take a fully Bayesian approach, not only to make estimation possible, but also because it provides the necessary information for use in psychological applications such as hypothesis testing. To illustrate the model we use data from 194 individuals that provided daily ratings of negative and positive affect, as well as their physical activity in the form of step counts over 100 consecutive days. We describe the fitted model, where we emphasize, with visualization, the richness of information provided by the M-MELSM. We demonstrate Bayesian hypothesis testing for the correlations between the random effects. We conclude by discussing limitations of the MELSM in general and extensions to the M-MELSM specifically for personality research.

Using Ecological Momentary Assessment to Examine the Effects of Duty Status on Acute Stress and Tiredness in Firefighters: A Pilot Study

OBJECTIVE

Characterize firefighters' acute stress and tiredness by duty status (ie, "off night/day," "on night/day").

METHODS

Thirty nine career firefighters completed three, eight-day smartphone-based ecological momentary assessment periods with seven surveys per day assessing stress and tiredness. Mixed-effects location scale models examined duty status effects on stress and tiredness.

RESULTS

Firefighters' lowest stress and tiredness levels were when off-duty (β = 16.27 and β = 24.71, respectively) and their highest levels were when on-duty (β = 24.47 and β = 32.18, respectively). Within-subject effects of duty status accounted for a larger proportion of variability in stress and tiredness for all duty types, except for stress when "on-duty night/off-duty day."

CONCLUSIONS

Firefighters had more similar stress and tiredness outcomes when they were on-duty and less similar outcomes when off-duty. This could be due to firefighters having more similar experiences when they are on- versus off-duty.

Geographical and sociodemographic differences in discontinuation of medication for Chronic Obstructive Pulmonary Disease - A Cross-Classified Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA)

BACKGROUND

While discontinuation of COPD maintenance medication is a known problem, the proportion of patients with discontinuation and its geographical and sociodemographic distribution are so far unknown in Sweden. Therefore, we analyse this question by applying an innovative approach called multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA).

PATIENTS AND METHODS

We analysed 49,019 patients categorized into 18 sociodemographic contexts and 21 counties of residence. All patients had a hospital COPD diagnosis and had been on inhaled maintenance medication during the 5 years before the study baseline in 2010. We defined "discontinuation" as the absolute lack of retrieval from a pharmacy of any inhaled maintenance medication during 2011. We performed a cross-classified MAIHDA and obtained the average proportion of discontinuation, as well as county and sociodemographic absolute risks, and compared them with a proposed benchmark value of 10%. We calculated the variance partition coefficient (VPC) and the area under the receiver operating characteristics curve (AUC) to quantify county and sociodemographic differences. To summarize the results, we used a framework with 15 scenarios defined by the size of the differences and the level of achievement in relation to the benchmark value.

RESULTS

Around 18% of COPD patients in Sweden discontinued maintenance medication, so the benchmark value was not achieved. There were very small county differences (VPC=0.35%, AUC=0.54). The sociodemographic differences were small (VPC=4.98%, AUC=0.57).

CONCLUSION

Continuity of maintenance medication among COPD patients in Sweden could be improved by reducing the unjustifiably high prevalence of discontinuation. The very small county and small sociodemographic differences should motivate universal interventions across all counties and sociodemographic groups. Geographical analyses should be combined with sociodemographic analyses, and the cross-classified MAIHDA is an appropriate tool to assess health-care quality.

Specifying Location-Scale Models for Heterogeneous Variances as Multilevel SEMs

Standard multilevel models focus on variables that predict the mean while the within-group variability is largely treated as a nuisance. Recent work has shown the advantage of including predictors for both the mean (the location submodel) and the variability (the scale submodel) within a single model. Constrained versions of the model can be fit in standard mixed effect model software, but the most general version with random effects in each of the location and scale submodels has been noted for being difficult to fit and estimate in software. However, the latest release of Mplus includes new capabilities that facilitate fitting the general version of the model as a multilevel structural equation model (SEM). This article introduces the general form of the model that includes location and scale random effects (called the location-scale model) and notes how it can be envisioned as a multilevel SEM. We provide a tutorial with example analyses and Mplus code for the model with two-level cross-sectional data and three-level repeated measures data and discuss how such a model has potential to extend recent developments in organizational science.

Two-part mixed-effects location scale models

Longitudinal time use data afford the opportunity to study within- and between-individual differences, but can present challenges in data analysis. Often the response set includes a large number of zeros representing those who did not engage in the target behavior. Coupled with this is a continuous measure of time use for those who did engage. The latter is strictly positive and skewed to the right if relatively few individuals engage in the behavior to a greater extent. Data analysis is further complicated for repeated measures, because within-individual responses are typically correlated, and some respondents may have missing data. This combination of zeros and positive responses is characteristic of a type of semicontinuous data in which the response is equal to a discrete value and is otherwise continuous. Two-part models have been successfully applied to cross-sectional time use data when the research goals distinguish between a respondent's likelihood to engage in a behavior and the time spent conditional on any time being spent, as these models allow different covariates to relate to each distinct aspect of a behavior. Two-part mixed-effects models extend two-part models for analysis of longitudinal semicontinuous data to simultaneously address longitudinal decisions to engage in a behavior and time spent conditional on any time spent. Heterogeneity between and within individuals can be studied in unique ways. This paper presents applications of these models to daily diary data to study individual differences in time spent relaxing or engaged in leisure activities for an adult sample.

Extending the mixed‐effects model to consider within‐subject variance for Ecological Momentary Assessment data

Ecological Momentary Assessment data present some new modeling opportunities. Typically, there are sufficient data to explicitly model the within-subject (WS) variance, and in many applications, it is of interest to allow the WS variance to depend on covariates as well as random subject effects. We describe a model that allows multiple random effects per subject in the mean model (eg, random location intercept and slopes), as well as random scale in the error variance model. We present an example of the use of this model on a real dataset and a simulation study that shows the benefit of this model, relative to simpler approaches.

From Nuisance to Novel Research Questions: Using Multilevel Models to Predict Heterogeneous Variances

Constructs that reflect differences in variability are of interest to many researchers studying workplace phenomena. The aggregation methods typically used to investigate “variability-based” constructs suffer from several limitations, including the inability to include Level 1 predictors and a failure to account for uncertainty in the variability estimates. We demonstrate how mixed-effects location-scale (MELS) and heterogeneous variance models, which are direct extensions of traditional mixed-effects (or multilevel) models, can be used to test mean (location)- and variability (scale)-related hypotheses simultaneously. The aims of this article are to demonstrate (a) how the MELS and heterogeneous variance models can be estimated with both nested cross-sectional and longitudinal data to answer novel research questions about constructs of interest to organizational researchers, (b) how a Bayesian approach allows for the inclusion of random intercepts and slopes when predicting both variability and mean levels, and finally (c) how researchers can use a multilevel approach to predict between-group heterogeneous variances. In doing so, this article highlights the added value of viewing variability as more than a statistical nuisance in organizational research.

A mixed-effects location scale model for time-to-event data: A smoking behavior application

In general, mixed-effects location scale models (MELS) allow assessment of within-person and between-person variability with time-to-event data for outcomes that follow a normal or ordinal distribution. In this article, we extend the mixed-effects location scale model to time-to-event data in relation to smoking data. Better understanding of the time-graded within-person variability of factors involved in nicotine dependence can be helpful to researchers in their efforts to fine-tune smoking cessation programs. We illustrate the MELS model with data on time to first cigarette measured every day for 7 days in smokers randomized to two groups: a) those asked to keep smoking, or b) those asked to stop. Our results show that some individuals remain very stable in their time to first cigarette over the week, while others show variable patterns. The stable individuals smoked every day, did not smoke immediately upon waking, and were all in the group asked to keep smoking. Conversely, the variable individuals had at least one day during which they did not smoke, other days during which they smoked within the first 5 min of waking, and they were almost all in the group asked to quit smoking. These findings suggested that MELS have the potential to provide insights on how people try to stop smoking. More importantly, this model can be applied to other clinically important outcomes such as time to relapse in a range of cessation programs.

Personality and Sleep: Neuroticism and Conscientiousness Predict Behaviourally Recorded Sleep Years Later

Sleep is one key feature of people's lives that defines their daily routine and reflects overall health and well–being. To test the relevance of personality for core aspects of sleep, we examined if personality traits across the five broad personality domains predicted behaviourally recorded, week–long sleep characteristics up to five years later (alongside subjective sleep quality). Data from 382 participants (63% female, aged 34–82 years) were drawn from the longitudinal study on Midlife in the United States Study—Biomarker project. In terms of mean tendencies, both neuroticism and conscientiousness signalled more sleep continuity (fewer interruptions) alongside better subjective quality. In terms of intra–individual sleep variability, neuroticism predicted more variability in sleep duration, continuity, and subjective sleep quality, while conscientiousness predicted less variability in sleep duration and sleep continuity. Extraversion, agreeableness, and openness traits did not generally foreshadow behaviourally recoded sleep, only higher ratings of subjective quality. These links were robust to the impact of demographic factors and were not moderated by the duration of time between personality and sleep assessments. The findings distinguish which personality traits foreshadow core aspects of sleep and also implicate multiple traits as predictors of variability, not just mean tendencies, in behaviourally recorded sleep. © 2019 European Association of Personality Psychology

Do fluctuations in positive affective and physical feeling states predict physical activity and sedentary time?

OBJECTIVES

Higher levels of positive affect and feelings of energy and vitality are associated with greater physical activity (PA) and lower sedentary time (ST). However, whether fluctuations in these feelings contribute to the regulation of these behaviors is unclear. This study examined the extent to which within-person variability in positive affect and feeling energetic predicted participants' overall levels of PA and ST.

DESIGN

This analysis combined data from four ecological momentary assessment (EMA) studies (age_range: 8-73 years) with ambulatory monitoring via waist-worn accelerometry (N=661).

METHODS

Positive affect and energy were assessed through EMA several times per day across 4-7 days. Accelerometer data was used to create the following behavioral outcomes: (1) meeting MVPA guidelines (children: 60 minutes/day, adults: 30 minutes/day) and (2) minutes of ST per hour of accelerometer wear. A two-stage analytic approach was used to test the study aim. In the first stage, Mixed-Effects Location Scale Modeling decomposed mean levels and variability in positive affect and energy. In the second stage, a linear or logistic regression (depending on whether the outcome was continuous or dichotomous, respectively) was tested to investigate associations between subject-level mean and variability in EMA ratings and the behavioral outcome.

RESULTS

Greater subject-level variability but not subject-level mean of feeling energetic was associated with lower odds of meeting MVPA guidelines (β=-0.43, p<0.05).

CONCLUSIONS

Fluctuations in physical feeling states may deplete self-regulatory resources involved in planning and implementing PA behavior. Alternatively, being more physically active may stabilize one's perceived energy levels.

Family contexts and sleep during adolescence

Sleep deprivation among adolescents has received much attention from health researchers and policymakers. Recent research indicates that variation in sleep duration from night to night is associated with multiple health outcomes. While there is evidence that sleep deprivation is socially patterned, we know little about how social contexts are associated with nightly sleep variation during adolescence (a life course stage when nightly sleep variation is particularly high). Given the importance of family environments for influencing adolescents’ sleep patterns, we hypothesized that disadvantaged family contexts would be associated with higher intra-individual variation (IIV) in nightly sleep duration, in addition to lower average nightly sleep duration. We tested these hypotheses in a diverse, population-based sample of 11–17 year-olds (N = 1095) from the Adolescent Health and Development in Context Study. Using survey and ecological momentary assessment data and a novel form of multi-level regression modeling (location-scale mixed modeling), we found that adolescents living in unmarried-parent, low SES, economically insecure, and high caregiver stress families had higher IIV in sleep than adolescents in families with more resources and less caregiver stress. There were fewer family effects on average sleep duration. This suggests family social and economic contexts are associated with an under-researched aspect of adolescent sleep, nightly variation, and may contribute to adolescent sleep problems with implications for their health and health disparities.

•

Consistency in nightly sleep duration is important for adolescent health.

•

SES and family stress associated with variability in sleep duration in adolescents.

•

Caregiver distress also associated with lower mean sleep duration among adolescents.

Mean level of positive affect moderates associations between volatility in positive affect, mental health, and alcohol consumption among mothers

Affective volatility (i.e., variability) is typically conceptualized as a marker of poor mental and behavioral health. The current research challenges this notion. Among individuals who typically experience low positive affect (PA), vacillation between bouts of joy and marked anhedonia may be associated with better behavioral health relative to a restricted range of PA experience. We therefore tested the hypothesis that mean levels of PA captured via an 8-day intensive repeated Ecological Momentary Assessment (EMA) would moderate the association of EMA-measured PA variability with behavioral health measured at intake among working mothers (N = 202) with young children-a population at risk for experiencing repeated stress- and reward-induced PA changes. Mixed-effects location scale modeling parsed variance in PA mean and variability, which were used in subject-level regression models of associations with depressive symptoms, anxiety, and alcohol use. PA mean moderated the associations of PA variability with depressive symptoms (Interaction: β = .20, p = .02) and alcohol consumption frequency (Interaction: β = .22, p = .02). PA variability and depressive symptom and alcohol consumption levels were positively associated at higher PA mean levels, whereas PA variability was inversely associated with depressive symptom and alcohol consumption at low mean PA. PA Mean × Variability Interactions were not significant for anxiety and binge drinking. We conclude that (a) intensive longitudinal modeling of Affect Mean × Variability Interaction effects may provide incremental information in psychopathology research; (b) PA volatility does not unilaterally indicate poor behavioral health; and (c) a nuanced perspective on the role of PA volatility may benefit clinical services for working mothers. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

A Mixed-Effects Location Scale Model for Dyadic Interactions

We present a mixed-effects location scale model (MELSM) for examining the daily dynamics of affect in dyads. The MELSM includes person and time-varying variables to predict the location, or individual means, and the scale, or within-person variances. It also incorporates a submodel to account for between-person variances. The dyadic specification can accommodate individual and partner effects in both the location and the scale components, and allows random effects for all location and scale parameters. All covariances among the random effects, within and across the location and the scale are also estimated. These covariances offer new insights into the interplay of individual mean structures, intra-individual variability, and the influence of partner effects on such factors. To illustrate the model, we use data from 274 couples who provided daily ratings on their positive and negative emotions toward their relationship - up to 90 consecutive days. The model is fit using Hamiltonian Monte Carlo methods, and includes subsets of predictors in order to demonstrate the flexibility of this approach. We conclude with a discussion on the usefulness and the limitations of the MELSM for dyadic research.

Modeling Intraindividual Variability in Three-Level Multilevel Models

Abstract. The mixed-effects location scale model is an extension of a multilevel model for longitudinal data. It allows covariates to affect both the within-subject variance and the between-subject variance (i.e., the intercept variance) beyond their influence on the means. Typically, the model is applied to two-level data (e.g., the repeated measurements of persons), although researchers are often faced with three-level data (e.g., the repeated measurements of persons within specific situations). Here, we describe an extension of the two-level mixed-effects location scale model to such three-level data. Furthermore, we show how the suggested model can be estimated with Bayesian software, and we present the results of a small simulation study that was conducted to investigate the statistical properties of the suggested approach. Finally, we illustrate the approach by presenting an example from a psychological study that employed ecological momentary assessment.

The Power to Detect and Predict Individual Differences in Intra-Individual Variability Using the Mixed-Effects Location-Scale Model

Our goal is to provide empirical scientists with practical tools and advice with which to test hypotheses related to individual differences in intra-individual variability using the mixed-effects location-scale model. To that end, we evaluate Type I error rates and power to detect and predict individual differences in intra-individual variability using this model and provide empirically-based guidelines for building scale models that include random and/or systematically-varying fixed effects. We also provide two power simulation programs that allow researchers to conduct a priori empirical power analyses. Our results aligned with statistical power theory, in that, greater power was observed for designs with more individuals, more repeated occasions, greater proportions of variance available to be explained, and larger effect sizes. In addition, our results indicated that Type I error rates were acceptable in situations when individual differences in intra-individual variability were not initially detectable as well as when the scale-model individual-level predictor explained all initially detectable individual differences in intra-individual variability. We conclude our paper by providing study design and model building advice for those interested in using the mixed-effects location-scale model in practice.

The role of public and private natural space in children's social, emotional and behavioural development in Scotland: A longitudinal study

INTRODUCTION

Poor mental health in childhood has implications for health and wellbeing in later life. Natural space may benefit children's social, emotional and behavioural development. We investigated whether neighbourhood natural space and private garden access were related to children's developmental change over time. We asked whether relationships differed between boys and girls, or by household educational status.

METHODS

We analysed longitudinal data for 2909 urban-dwelling children (aged 4 at 2008/9 baseline) from the Growing Up in Scotland (GUS) survey. The survey provided social, emotional and behavioural difficulty scores (Strengths and Difficulties Questionnaire (SDQ)), and private garden access. Area (%) of total natural space and parks within 500m of the child's home was quantified using Scotland's Greenspace Map. Interactions for park area, total natural space area, and private garden access with age and age² were modelled to quantify their independent contributions to SDQ score change over time.

RESULTS

Private garden access was strongly related to most SDQ domains, while neighbourhood natural space was related to better social outcomes. We found little evidence that neighbourhood natural space or garden access influenced the trajectory of developmental change between 4 and 6 years, suggesting that any beneficial influences had occurred at younger ages. Stratified models showed the importance of parks for boys, and private gardens for the early development of children from low-education households.

CONCLUSION

We conclude that neighbourhood natural space may reduce social, emotional and behavioural difficulties for 4-6 year olds, although private garden access may be most beneficial.