Interpretable principal component analysis for multilevel multivariate functional data.

Many studies collect functional data from multiple subjects that have both multilevel and multivariate structures. An example of such data comes from popular neuroscience experiments where participants' brain activity is recorded using modalities such as electroencephalography and summarized as power within multiple time-varying frequency bands within multiple electrodes, or brain regions. Summarizing the joint variation across multiple frequency bands for both whole-brain variability between subjects, as well as location-variation within subjects, can help to explain neural reactions to stimuli. This article introduces a novel approach to conducting interpretable principal components analysis on multilevel multivariate functional data that decomposes total variation into subject-level and replicate-within-subject-level (i.e., electrode-level) variation and provides interpretable components that can be both sparse among variates (e.g., frequency bands) and have localized support over time within each frequency band. Smoothness is achieved through a roughness penalty, while sparsity and localization of components are achieved by solving an innovative rank-one based convex optimization problem with block Frobenius and matrix $L_1$-norm-based penalties. The method is used to analyze data from a study to better understand reactions to emotional information in individuals with histories of trauma and the symptom of dissociation, revealing new neurophysiological insights into how subject- and electrode-level brain activity are associated with these phenomena. Supplementary materials for this article are available online.

Covariate-guided Bayesian mixture of spline experts for the analysis of multivariate high-density longitudinal data.

With rapid development of techniques to measure brain activity and structure, statistical methods for analyzing modern brain-imaging data play an important role in the advancement of science. Imaging data that measure brain function are usually multivariate high-density longitudinal data and are heterogeneous across both imaging sources and subjects, which lead to various statistical and computational challenges. In this article, we propose a group-based method to cluster a collection of multivariate high-density longitudinal data via a Bayesian mixture of smoothing splines. Our method assumes each multivariate high-density longitudinal trajectory is a mixture of multiple components with different mixing weights. Time-independent covariates are assumed to be associated with the mixture components and are incorporated via logistic weights of a mixture-of-experts model. We formulate this approach under a fully Bayesian framework using Gibbs sampling where the number of components is selected based on a deviance information criterion. The proposed method is compared to existing methods via simulation studies and is applied to a study on functional near-infrared spectroscopy, which aims to understand infant emotional reactivity and recovery from stress. The results reveal distinct patterns of brain activity, as well as associations between these patterns and selected covariates.

Risk for Complicated Grief After the COVID-19 Death of a Marital Partner in Late Life.

OBJECTIVE: To identify 1) complicated grief symptom clusters among acutely-bereaved older adults who have lost a spouse to COVID-19 and 2) if spousal death due to COVID-19 increased risk of developing probable PGD METHODS: Eighty adults participating in a randomized controlled trial for depression prevention (mean age [± SD] = 70.4 [6.6]) completed the Inventory of Complicated Grief, every 3 months over a maximum of 15 months. Twenty-four percent (n = 19) of participants lost a spouse to COVID-19; 76% (n = 61) lost a spouse to other causes of death. Adjusted linear regression examined the associations between COVID-19 bereavement and six symptom clusters: yearning and preoccupation, anger and bitterness, shock and disbelief, estrangement from others, hallucinations, and behavior change. RESULTS: Compared to the non-COVID-19 group, the COVID-19 bereaved group reported greater shock and disbelief, hallucinations of the deceased, and estrangement from others. COVID-19 death was also associated with higher risk for probable prolonged grief disorder (PGD) at 12 months (odds ratio = 4.38, p = 0.027). CONCLUSIONS: Older adults who have lost a spouse to COVID-19 present with specific symptoms of distress and may eventually require clinical care for PGD.

Efficacy of the Transdiagnostic Intervention for Sleep and Circadian Dysfunction for Depression Symptoms and Sleep-Wake Disruption in Older and Younger Adults: Secondary Age-Stratified Analysis of a Randomized Controlled Trial.

OBJECTIVE: Perform a secondary analysis examining the efficacy of the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C) for depression symptom responses, and explore changes in potential target mechanisms. DESIGN: Secondary analysis of a randomized controlled trial with convenience age subsamples (younger (20-49 year; n = 52) versus and older (50-71 years; n = 35)). SETTING: Community mental health clinics. PARTICIPANTS: Eighty-seven adults with serious mental illness. INTERVENTION: TranS-C versus treatment as usual (TAU). MEASUREMENTS: Outcomes were depression symptoms (Quick Inventory of Depression Symptoms), insomnia symptoms (Insomnia Severity Index), and objective sleep-wake rhythm measures (interdaily stability and relative amplitude). RESULTS: Depression response rates (≥50% symptom reductions) were higher in the TranS-C (35.0%) than the TAU (8.8%) group 6-months postintervention (χ2 = 10.3, p = 0.001). There was a medium effect of TranS-C versus TAU on depression symptoms 6-months postintervention (Cohen's d = -0.40, 95% confidence interval (CI): -0.81, 0.01). In both age groups, there were large treatment effects on insomnia symptoms post-treatment (Cohen's d >0.90). In the older subsample, there were additionally medium treatment effects on post-treatment interdaily stability (Cohen's d = 0.60, 95% CI: -0.11, 1.61). Post-treatment reductions in insomnia symptoms correlated with depression symptom reduction 6-months later in the younger subsample (Spearman rho = 0.59, n = 20, p = 0.008). In older adults, postintervention increases in interdaily stability correlated with depression symptom reductions 6-months later (Spearman rho = -0.52, n = 15, p = 0.049). CONCLUSION: Confirmatory trials are needed, given the low age-specific sample sizes here, to determine if TranS -C's produces durable depression responses by increasing sleep-wake rhythm stability in older adults and improving insomnia symptoms in younger adults. BRIEF ARTICLE SUMMARY: The authors evaluated preliminary efficacy of a behavioral intervention that targets sleep/sleep-wake rhythms, the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C), for depression symptoms in people with serious mental illness. TranS-C was associated with higher depression response rates than treatment as usual 6-months postintervention. The degree of depression symptom response 6-months later was related to the degree of treatment phase improvements in interdaily stability (in older adults) and reduction in insomnia severity (in younger adults). A pragmatic nonpharmacologic intervention, the Transdiagnostic Intervention for Sleep and Circadian Dysfunction, has preliminary efficacy for improving sleep-wake factors and depression symptoms.

Impact of sleep on chronobiology of micturition among healthy older adults.

Nocturia (waking to void) is prevalent among older adults. Disruption of the well-described circadian rhythm in urine production with higher nighttime urine output is its most common cause. In young adults, their circadian rhythm is modulated by the 24-h secretory pattern of hormones that regulate salt and water excretion, including antidiuretic hormone (ADH), renin, angiotensin, aldosterone, and atrial natriuretic peptide (ANP). The pattern of hormone secretion is less clear in older adults. We investigated the effect of sleep on the 24-h secretion of these hormones in healthy older adults. Thirteen participants aged ≥65 yr old underwent two 24-h protocols at a clinical research center 6 wk apart. The first used a habitual wake-sleep protocol, and the second used a constant routine protocol that removed the influence of sleep, posture, and diet. To assess hormonal rhythms, plasma was collected at 8:00 am, 12:00 pm, 4:00 pm, and every 30 min from 7:00 pm to 7:00 am. A mixed-effects regression model was used to compare subject-specific and mean trajectories of hormone secretion under the two conditions. ADH, aldosterone, and ANP showed a diurnal rhythm that peaked during sleep in the wake-sleep protocol. These nighttime elevations were significantly attenuated within subjects during the constant routine. We conclude that sleep has a masking effect on circadian rhythm amplitude of ADH, aldosterone, and ANP: the amplitude of each is increased in the presence of sleep and reduced in the absence of sleep. Disrupted sleep could potentially alter nighttime urine output in healthy older adults via this mechanism.NEW & NOTEWORTHY Nocturia (waking to void) is the most common cause of sleep interruption among older adults, and increased nighttime urine production is its primary etiology. We showed that in healthy older adults sleep affects the 24-h secretory rhythm of hormones that regulate salt-water balance, which potentially alters nighttime urine output. Further studies are needed to elucidate the impact of chronic insomnia on the secretory rhythms of these hormones.

A lifespan perspective on depression in the postpartum period in a racially and socioeconomically diverse sample of young mothers.

BACKGROUND: Consistent evidence from retrospective reports and case registry studies indicates that a history of depression is a major risk factor for depression in the peripartum period. However, longitudinal studies with racially and socioeconomically diverse samples of young mothers are lacking, and little is known about developmental patterns of depression across the lifespan that can inform preventive interventions. METHODS: Young primiparous mothers (n = 399, 13-25 years, 81% Black) were recruited from a population-based prospective study that began in childhood. Women reported on depression symptoms for at least 3 years prior to their pregnancy, during pregnancy, and at 4 months postpartum. Linear regression models were used to estimate change in pre-pregnancy depression severity and to evaluate associations between patterns of lifetime history and postpartum depression symptoms. RESULTS: Results revealed high levels of continuity in depression from pregnancy to postpartum, and across multiple years pre-pregnancy to postpartum. Overall, depression severity leading up to pregnancy decreased over time, but patterns of worsening or improving symptoms were not associated with depression severity in the postpartum period. Instead, area under the pre-pregnancy trajectory curve, representing cumulative lifetime depression burden, was uniquely associated with postpartum depression after adjusting for prenatal depression severity. CONCLUSIONS: Depression in the postpartum period should be considered within a lifespan perspective of risk that accumulates before conception. Clinical screening and early interventions are needed in adolescence and young adulthood to prevent the onset and persistence of depressive symptoms that could have long-term implications for peripartum health.

Impact of sedentary behavior and emotional support on prenatal psychological distress and birth outcomes during the COVID-19 pandemic.

Abstract. BACKGROUND: Studies have reported mixed findings regarding the impact of the coronavirus disease 2019 (COVID-19) pandemic on pregnant women and birth outcomes. This study used a quasi-experimental design to account for potential confounding by sociodemographic characteristics. METHODS: Data were drawn from 16 prenatal cohorts participating in the Environmental influences on Child Health Outcomes (ECHO) program. Women exposed to the pandemic (delivered between 12 March 2020 and 30 May 2021) (n = 501) were propensity-score matched on maternal age, race and ethnicity, and child assigned sex at birth with 501 women who delivered before 11 March 2020. Participants reported on perceived stress, depressive symptoms, sedentary behavior, and emotional support during pregnancy. Infant gestational age (GA) at birth and birthweight were gathered from medical record abstraction or maternal report. RESULTS: After adjusting for propensity matching and covariates (maternal education, public assistance, employment status, prepregnancy body mass index), results showed a small effect of pandemic exposure on shorter GA at birth, but no effect on birthweight adjusted for GA. Women who were pregnant during the pandemic reported higher levels of prenatal stress and depressive symptoms, but neither mediated the association between pandemic exposure and GA. Sedentary behavior and emotional support were each associated with prenatal stress and depressive symptoms in opposite directions, but no moderation effects were revealed. CONCLUSIONS: There was no strong evidence for an association between pandemic exposure and adverse birth outcomes. Furthermore, results highlight the importance of reducing maternal sedentary behavior and encouraging emotional support for optimizing maternal health regardless of pandemic conditions.

Spectra in low-rank localized layers (SpeLLL) for interpretable time-frequency analysis.

The time-varying frequency characteristics of many biomedical time series contain important scientific information. However, the high-dimensional nature of the time-varying power spectrum as a surface in time and frequency limits its direct use by applied researchers and clinicians for elucidating complex mechanisms. In this article, we introduce a new approach to time-frequency analysis that decomposes the time-varying power spectrum in to orthogonal rank-one layers in time and frequency to provide a parsimonious representation that illustrates relationships between power at different times and frequencies. The approach can be used in fully nonparametric analyses or in semiparametric analyses that account for exogenous information and time-varying covariates. An estimation procedure is formulated within a penalized reduced-rank regression framework that provides estimates of layers that are interpretable as power localized within time blocks and frequency bands. Empirical properties of the procedure are illustrated in simulation studies and its practical use is demonstrated through an analysis of heart rate variability during sleep.

Activity patterns related to depression symptoms in stressed dementia caregivers.

OBJECTIVES: Self-reported activity restriction is an established correlate of depression in dementia caregivers (dCGs). It is plausible that the daily distribution of objectively measured activity is also altered in dCGs with depression symptoms; if so, such activity characteristics could provide a passively measurable marker of depression or specific times to target preventive interventions. We therefore investigated how levels of activity throughout the day differed in dCGs with and without depression symptoms, then tested whether any such differences predicted changes in symptoms 6 months later. DESIGN, SETTING, PARTICIPANTS, AND MEASUREMENTS: We examined 56 dCGs (mean age = 71, standard deviation (SD) = 6.7; 68% female) and used clustering to identify subgroups which had distinct depression symptom levels, leveraging baseline Center for Epidemiologic Studies of Depression Scale-Revised Edition and Patient Health Questionnaire-9 (PHQ-9) measures, as well as a PHQ-9 score from 6 months later. Using wrist activity (mean recording length = 12.9 days, minimum = 6 days), we calculated average hourly activity levels and then assessed when activity levels relate to depression symptoms and changes in symptoms 6 months later. RESULTS: Clustering identified subgroups characterized by: (1) no/minimal symptoms (36%) and (2) depression symptoms (64%). After multiple comparison correction, the group of dCGs with depression symptoms was less active from 8 to 10 AM (Cohen's d ≤ -0.9). These morning activity levels predicted the degree of symptom change on the PHQ-9 6 months later (per SD unit ß = -0.8, 95% confidence interval: -1.6, -0.1, p = 0.03) independent of self-reported activity restriction and other key factors. CONCLUSIONS: These novel findings suggest that morning activity may protect dCGs from depression symptoms. Future studies should test whether helping dCGs get active in the morning influences the other features of depression in this population (i.e. insomnia, intrusive thoughts, and perceived activity restriction).

Blood Pressure Cuff Inflation Briefly Increases Female Adolescents' Restlessness During Sleep on the First But Not Second Night of Ambulatory Blood Pressure Monitoring.

OBJECTIVE: Ambulatory blood pressure monitoring (ABPM) increases restlessness during adults' sleep in laboratory settings, but there is little evidence of an association among adolescents or in naturalistic environments. This study examined activity levels before and after blood pressure cuff inflation during sleep to determine whether and for how long ABPM increased restlessness during sleep in healthy adolescents. METHODS: Two hundred thirty-four healthy adolescents (mean age = 15.72 [1.30] years; 54% female; 57% Black) completed two consecutive nights of hourly ABPM and wrist-worn actigraphy. Activity counts during sleep, averaged across 5-minute bins, were compared in the 20 minutes before and after blood pressure cuff inflation using a four-level mixed model (bins within hours within nights within participants). Interactions of bin with night, sex, and race were examined. Covariates included age, sex, and race. RESULTS: Activity counts in the 5-minute bin immediately after cuff inflation were 10% to 14% higher than all other bins before ( p < .001) and after ( p < .001) cuff inflation. This effect differed by night and sex, as activity levels during 5-minute post-cuff inflation were elevated only on night 1 ( p values < .001) and only in female participants ( p values < .001). Effects did not differ by race. CONCLUSIONS: Cuff inflation during ABPM briefly increased adolescent female participants' restlessness during sleep. Habituation occurred after one night, so two nights of ABPM may minimize impact on sleep. If only one night of ABPM is feasible, excluding 5 minutes of actigraphy data after each cuff inflation may accommodate the impact of ABPM on restlessness during sleep.

Using Simultaneous Confidence Bands to Calculate the Margin of Error in Estimating Typical Biomechanical Waveforms.

Studies of human movement usually collect data from multiple repetitions of a task and use the average of all movement trials to approximate the typical kinematics or kinetics pattern for each individual. Few studies report the expected accuracy of these estimated mean kinematics or kinetics waveforms for each individual. The purpose of this study is to demonstrate how simultaneous confidence bands, which is an approach to quantify uncertainty across an entire waveform based on limited data, can be used to calculate margin of error (MOE) for waveforms. Bilateral plantar pressure data were collected from 70 participants as they walked over 4 surfaces for an average of at least 300 steps per surface. The relationship between MOE and the number of steps included in the analysis was calculated using simultaneous confidence bands, and 3 methods commonly used for pointwise estimates: intraclass correlation, sequential averaging, and T-based MOE. The conventional pointwise approaches underestimated the number of trials required to estimate biomechanical waveforms within a desired MOE. Simultaneous confidence bands are an objective approach to more accurately estimate the relationship between the number of trials collected and the MOE in estimating typical biomechanical waveforms.

Longitudinal Association Between Depressive Symptoms and Multidimensional Sleep Health: The SWAN Sleep Study.

BACKGROUND: Depressive symptoms and sleep disturbances disproportionately affect midlife women. While there may be a bidirectional association, few studies have examined whether depressive symptoms are longitudinally associated with subsequent sleep. Sleep is typically considered unidimensional, despite emerging evidence that multidimensional sleep health provides novel information on the sleep-health link. PURPOSE: The current study examined whether higher depressive symptoms were longitudinally associated with poorer multidimensional sleep health. METHOD: Depressive symptoms were assessed with the Center for Epidemiologic Studies Depression Scale across six to nine annual assessments in 302 midlife women from the Study of Women's Health Across the Nation. Six months after their last assessment, actigraphy (mean ± standard deviation = 29.3 ± 6.9 days) and self-report were used to assess sleep health components: efficiency, duration, mid-sleep timing, regularity, alertness, and satisfaction, which were dichotomized and summed to create a composite multidimensional sleep health score. Mixed-effects models were used to evaluate the longitudinal associations between depressive symptoms and multidimensional sleep health, as well as individual sleep health components, adjusting for covariates. Exploratory analyses stratified models by race/ethnicity. RESULTS: Higher depressive symptoms were associated with subsequent poorer multidimensional sleep health (p < .0.001) and lower alertness (p < .0001) and satisfaction with sleep (p < .0001). CONCLUSIONS: Our finding that higher average depressive symptoms were associated longitudinally with actigraphy-measured poorer sleep health in midlife women is novel and converges with the larger body of evidence that these two common symptoms are strongly associated. The bidirectional relationship between these two prevalent symptoms needs to be studied in prospective longitudinal studies.

Management and Outcomes of Spontaneous Pneumomediastinum in Children.

OBJECTIVES: Management of spontaneous pneumomediastinum in the pediatric population is highly variable. There are limited data on the use of diagnostic tests and the need for admission. Our objectives were to characterize the management of pediatric spontaneous pneumomediastinum, determine the diagnostic yield of advanced imaging, and describe the patients' outcomes. METHODS: This is a retrospective cohort study of all patients presenting to a single tertiary pediatric emergency department between January 2008 and February 2015 diagnosed with pneumomediastinum. Patients were identified using 2 complementary strategies: International Classification of Diseases, Ninth Revision billing codes and a keyword search of the hospital radiology database. RESULTS: We identified 183 patients with spontaneous pneumomediastinum. The mean age was 12.8 ± 4.8 years. Diagnosis was established by chest radiograph (CXR) in 165 (90%) patients, chest computed tomography in 15 (8%), neck imaging in 2 (1%), and abdominal imaging in 1. After diagnosis, many patients underwent additional studies: repeat CXR (99, 54%), chest computed tomography (53, 29%), esophagram (45, 25%), and laryngoscopy (15, 8%). Seventy-eight percent of patients (n = 142) were admitted with a median length of stay of 27 hours (18.4-45.6 hours). Six patients returned to the emergency department within 96 hours for persistent chest pain; 2 were admitted, and 1 was found to have worsening pneumomediastinum on CXR. We performed a secondary analysis on 3 key subgroups: primary spontaneous pneumomediastinum (64, 35%), secondary gastrointestinal-associated pneumomediastinum (31, 17%), and secondary respiratory-associated pneumomediastinum (88, 48%). No patients in the study received an invasive intervention for pneumomediastinum. In all patients, further studies did not yield additional diagnostic information. CONCLUSIONS: Our data suggest that patients with spontaneous pneumomediastinum who are clinically well appearing can be managed conservatively with clinical observation, avoiding exposure to radiation and invasive procedures.

Profiles of Accelerometry-Derived Physical Activity Are Related to Perceived Physical Fatigability in Older Adults.

Physical activity (PA) is associated with greater fatigability in older adults; little is known about magnitude, shape, timing and variability of the entire 24-h rest-activity rhythm (RAR) associated with fatigability. We identified which features of the 24-h RAR pattern were independently and jointly associated with greater perceived physical fatigability (Pittsburgh Fatigability Scale, PFS, 0-50) in older adults (n = 181, 71.3 ± 6.7 years). RARs were characterized using anti-logistic extended cosine models and 4-h intervals of PA means and standard deviations across days. A K-means clustering algorithm approach identified four profiles of RAR features: "Less Active/Robust", "Earlier Risers", "More Active/Robust" and "Later RAR". Quantile regression tested associations of each RAR feature/profile on median PFS adjusted for age, sex, race, body mass index and depression symptomatology. Later rise times (up mesor; ß = 1.38, p = 0.01) and timing of midpoint of activity (acrophase; ß = 1.29, p = 0.01) were associated with higher PFS scores. Lower PA between 4 a.m. and 8 a.m. was associated with higher PFS scores (ß = -4.50, p = 0.03). "Less Active/Robust" (ß = 6.14, p = 0.01) and "Later RAR" (ß = 3.53, p = 0.01) patterns were associated with higher PFS scores compared to "Earlier Risers". Greater physical fatigability in older adults was associated with dampened, more variable, and later RARs. This work can guide development of interventions aimed at modifying RARs to reduce fatigability in older adults.

Alcohol Use Trajectories Before and After Pregnancy Among Adolescent and Young Adult Mothers.

BACKGROUND: Moderate to heavy alcohol use during pregnancy (≥3 drinks/occasion) is linked to a range of adverse offspring outcomes. Prior studies suggest that adolescent and young adult mothers may be particularly vulnerable to these risky drinking behaviors during and after pregnancy. This study used latent class growth analysis (LCGA) to examine changes in risky drinking from prepregnancy to postpregnancy and identified prepregnancy predictors representing cognitive and social factors to inform prevention. METHODS: Participants included 432 adolescent and young adult mothers (aged 13 to 24; 78% Black) assessed annually since childhood in the Pittsburgh Girls Study. Participants prospectively reported frequency of risky drinking (≥3 drinks/occasion) in the year of pregnancy and 2 years before and after pregnancy, as well as social-cognitive risk factors for drinking (alcohol expectancies, drinking motives, caregiver, and peer alcohol use). RESULTS: Approximately 11% of young mothers in this community-based sample engaged in risky drinking at least once a month during the year of pregnancy, with greater frequency observed for young adult (vs. adolescent) mothers. LCGA revealed 4 risky drinking trajectories across the perinatal period, including 3 relatively stable trajectories (low, moderate, and high frequency) and 1 distinct trajectory ("postpartum initiators") characterized by increased risky drinking post-childbirth. Peer alcohol use during and after pregnancy distinguished higher risk from lower risk trajectories, controlling for age, race, and age of drinking onset. Exposure to caregiver alcohol use and prepregnancy coping motives for drinking also elevated risk, but only for mothers who became pregnant in early adulthood (vs. adolescence). CONCLUSIONS: Adolescent and young adult mothers show heterogeneous risky drinking trajectories from prepregnancy to postpregnancy. Pregnant women in the transitional age of early adulthood who report exposure to caregiver alcohol use, more friends who drink, and coping motives for drinking may have elevated risk for moderate-heavy drinking during the perinatal period and may benefit from targeted intervention.

White Matter Integrity Underlying Depressive Symptoms in Dementia Caregivers.

OBJECTIVE: We sought to determine whether the aspects of white matter connectivity implicated in major depression also relate to mild depressive symptoms in family dementia caregivers (dCGs). METHODS: Forty-one dCGs (average age=69 years, standard deviation=6.4) underwent a 7 Tesla 64-direction (12-minute) diffusion-weighted imaging sequence. We compared the fractional anisotropy (FA) of 11 white matter features between dCGs with (n=20) and without (n=21) depressive symptoms (Patient Health Questionnaire-9 scores ≥5). RESULTS: Caregivers reporting depression symptoms had lower FA in tracts connecting to the posterior cingulate cortex (Cohen's d = -0.9) and connecting dorsolateral prefrontal with rostral cingulate regions (Cohen's d = -1.2). CONCLUSIONS: Posterior cingulate and dorsolateral prefrontal-to-rostral cingulate white matter, implicated in prior studies of major depression, appear relevant to mild depression in dCGs.

The association between physical activity and a composite measure of sleep health.

PURPOSE: Physical activity has been associated with several individual dimensions of sleep. However, the association between physical activity and sleep health, a construct that emphasizes the multidimensional nature of sleep, has not been explored. This analysis examined the relationship between physical activity and a composite measure of sleep health. METHODS: A total of 114 adults (66% female, 60.3 ± 9.2 years) were included in the analyses. Participants reported daily light-intensity physical activity (LPA) and moderate- and vigorous-intensity physical activity (MVPA) via diary, while wearing a pedometer (Omron HJ-720ITC) to measure daily steps. Sleep health was measured using the RU_SATED questionnaire, which addresses regularity of sleep patterns, satisfaction with sleep, daytime alertness, and sleep timing, efficiency, and duration. Multiple linear regression, binary logistic regression, and analysis of covariance (ANCOVA) were utilized for analyses. RESULTS: Mean sleep health score was 9.6 ± 2.4 (0 [poor]-12 [good]). Participants reported 62.9 ± 66.0 and 51.2 ± 51.2 min/day of LPA and MVPA, respectively, and took 5585.5 ± 2806.7 steps/day. Greater MVPA was associated with better sleep health (ß = 0.27, P = 0.005) and sleep health scores differed between those reporting < 30 min/day and ≥ 60 min/day of MVPA (P = 0.004). Greater MVPA was associated with higher odds of having good sleep satisfaction (OR = 1.58 [1.14-2.20], P < 0.01), timing (OR = 2.07 [1.24-3.46], P < 0.01), and duration (OR = 1.48 [1.02-2.18], P = 0.04). Pedometer-based physical activity and LPA were not related to sleep health or its individual dimensions. CONCLUSIONS: In middle- to older-aged adults, higher-intensity activity, but not lower-intensity or volume of activity, was associated with greater sleep health. These data suggest that physical activity intensity may be important for sleep health.

Incorporating prior information with fused sparse group lasso: Application to prediction of clinical measures from neuroimages.

Predicting clinical variables from whole-brain neuroimages is a high-dimensional problem that can potentially benefit from feature selection or extraction. Penalized regression is a popular embedded feature selection method for high-dimensional data. For neuroimaging applications, spatial regularization using the ℓ1 or ℓ2 norm of the image gradient has shown good performance, yielding smooth solutions in spatially contiguous brain regions. Enormous resources have been devoted to establishing structural and functional brain connectivity networks that can be used to define spatially distributed yet related groups of voxels. We propose using the fused sparse group lasso (FSGL) penalty to encourage structured, sparse, and interpretable solutions by incorporating prior information about spatial and group structure among voxels. We present optimization steps for FSGL penalized regression using the alternating direction method of multipliers algorithm. With simulation studies and in application to real functional magnetic resonance imaging data from the Autism Brain Imaging Data Exchange, we demonstrate conditions under which fusion and group penalty terms together outperform either of them alone.

Cardiovascular Stress Reactivity and Carotid Intima-Media Thickness: The Buffering Role of Slow-Wave Sleep.

OBJECTIVE: Exaggerated cardiovascular reactivity to acute psychological stress has been associated with increased carotid intima-media thickness (IMT). However, interstudy variability in this relationship suggests the presence of moderating factors. The current study aimed to test the hypothesis that poor nocturnal sleep, defined as short total sleep time or low slow-wave sleep, would moderate the relationship between cardiovascular reactivity and IMT. METHODS: Participants (N = 99, 65.7% female, age = 59.3 ± 9.3 years) completed a two-night laboratory sleep study and cardiovascular examination where sleep and IMT were measured. The multisource interference task was used to induce acute psychological stress, while systolic and diastolic blood pressure and heart rate were monitored. Moderation was tested using the PROCESS framework in SPSS. RESULTS: Slow-wave sleep significantly moderated the relationship between all cardiovascular stress reactivity variables and IMT (all pinteraction ≤ .048, all ΔRinteraction ≥ .027). Greater stress reactivity was associated with higher IMT values in the low slow-wave sleep group and lower IMT values in the high slow-wave sleep group. No moderating effects of total sleep time were observed. CONCLUSIONS: The results provide evidence that nocturnal slow-wave sleep moderates the relationship between cardiovascular stress reactivity and IMT and may buffer the effect of daytime stress-related disease processes.

The Relationship Between Childhood Trauma and Poor Sleep Health in Adulthood.

OBJECTIVE: Childhood trauma has been related to adverse behavioral, mental, and health outcomes later in life. Sleep may be a potential mechanism through which childhood trauma is related to adverse health. The current retrospective study aimed to characterize the relationship between childhood trauma exposure and sleep health, a novel multidimensional measure of sleep. METHODS: Participants (N = 161; mean [standard deviation] age = 59.85 [9.06] years; 67.7% female) retrospectively reported trauma exposure using the Trauma History Questionnaire. Childhood trauma was defined as the number of reported traumatic events before 18 years of age. Trauma exposure after 18 years of age and across the life-span was also recorded. Sleep health was derived both from diary- and actigraphy-assessed measures of sleep regularity, timing, efficiency, and duration, subjective sleep satisfaction, and daytime sleepiness from the Epworth Sleepiness Scale. The relationships between childhood trauma exposure and sleep health were examined using hierarchical linear regression, controlling for relevant covariates. RESULTS: In unadjusted models, a greater number of childhood trauma exposures were associated with poorer diary- and actigraphy-measured sleep health in adulthood. After adjustment for current stress, depression history, and other sociodemographic covariates, greater childhood trauma remained significantly associated with poorer sleep health (diary: ß = -0.20, ΔR = 0.032; actigraphy: ß = -0.19, ΔR = 0.027). Trauma exposure after 18 years of age and across the life-span did not relate to diary- or actigraphy-based sleep health. CONCLUSIONS: Childhood trauma may affect sleep health in adulthood. These findings align with the growing body of evidence linking childhood trauma to adverse health outcomes later in life.