Associations of changes in fat free mass with risk for type 2 diabetes: Hispanic Community Health Study/Study of Latinos

AIMS

To determine whether loss of muscle mass (approximated using fat free mass [FFM]) is associated with risk for type 2 diabetes mellitus (T2DM) in Hispanic/Latino adults in the United States.

METHODS

Participants were Hispanic/Latino adults (18-74-year-olds) who completed Visit 2 of the Hispanic Community Health Study/Study of Latinos (HCHS/SOL; multi-site, prospective cohort study; 6.1-year follow-up) and did not have T2DM at baseline (n = 6264). At baseline and Visit 2, FFM was measured using bioelectrical impedance analysis and fasting glucose, HbA1c, and fasting insulin were measured by examiners. Diabetes was defined according to American Diabetes Association criteria. Survey-weighted Poisson regression models examined the association of percent change in relative FFM (%ΔFFM) with incident prediabetes and T2DM. Survey-weighted multivariable regression models examined associations of %ΔFFM with changes in glucose and insulin measures.

RESULTS

Relative FFM declined by 2.1% between visits. %ΔFFM was inversely associated with incident prediabetes (p-for-trend = 0.001) and with changes in glucose and insulin measures (p-for-trend <0.0001). Findings were null, except for HOMA-IR, after adjustment for changes in adiposity measures. Associations were generally stronger for individuals with baseline overweight/obesity.

CONCLUSIONS

Reducing loss of FFM during adulthood may reduce prediabetes risk (primarily insulin resistance), particularly among individuals with overweight/obesity.

1097 Sleep Disturbances, Sleep Burden, And Depressive Symptoms In US Hispanics/Latinos: Results From The HCHS/SOL Sueño Study

Introduction

While sleep disturbances and depression often co-occur, these associations are understudied among Hispanics/Latinos. We examined the associations of sleep disturbances and sleep burden with depressive symptoms among Hispanic/Latino adults in the United States.

Methods

We used cross-sectional data from the Hispanic Community Health Study/Study of Latinos Sueño Ancillary study (2010-2013). The study enrolled 2072 adults (ages 18-64; 51.5% females) who completed one-week wrist-actigraphy and sleep questionnaires. Sleep burden was operationalized as the total count of sleep disturbances across six domains (duration, efficiency, midpoint, variability, insomnia, sleepiness). Depressive symptoms were assessed using the Center for Epidemiological Studies Depression scale (CESD-10). We used weighted survey linear regressions to evaluate the association of sleep disturbances and sleep burden with elevated depressive symptoms (CESD≥10) in individual models adjusted for age, gender, site, heritage, nativity, education, income, and employment. Sensitivity analyses further adjusted for behavioral health risk factors and apnea-hypopnea index.

Results

An estimated 28.3% had elevated depressive symptoms, 8.0% had short sleep duration (&lt;6 hours of sleep), 10.9% had long sleep duration (&gt;9 hours), 45.2% exhibited a later sleep midpoint (≥4:00AM), 38.4% had high sleep timing variability (upper third tertile for between day sleep midpoint), 15.3% had insomnia (ISI≥10), 17.3% had excessive daytime sleepiness (ESS ≥10), 21.5% had poor sleep efficiency (&lt;85%), and 77.4% had a total sleep burden count of ≥0. Insomnia (ß=0.49,95%CI:.43,.56), later sleep timing (ß=0.10,95%CI:.04,.16), excessive daytime sleepiness (ß=0.19,95%CI:.11,.27), poor sleep efficiency (ß=0.09,95%CI:.02,.17), high variability (ß=0.07, 95%CI:.01,.12), and sleep burden (ß=0.11,95%CI:.09,.13), were each positively associated with elevated depressive symptoms in individual adjusted models and sensitivity analyses. Extreme sleep durations were not associated with elevated depressive symptoms.

Conclusion

Multiple inter-related sleep disturbances, particularly those pertaining to sleep quality and timing, are associated with depression and may be targets for future interventions aimed at improving mood among Hispanics/Latinos.

Support

HL127307, HL098927, HL125748

0609 Sleep Phenotypes in Middle-Aged and Older Hispanics/Latinos. Results from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL)

Introduction

Identifying sleep phenotypes in the diverse and understudied US Hispanic/Latino population is critical to developing interventions and mitigating distal clinical outcomes (e.g. dementias).

Methods

Using latent class analyses (LCA), we identify empirically derived and clinically meaningful sleep phenotypes using data on community dwelling middle-aged/older adults (ages ≥45-years) from the HCHS/SOL (2008-2011) - Investigation of Neurocognitive Aging (n=6,377). Sleep variables used included Apnea/Hypopnea Index (AHI), percent time SpO2&lt;90%, Epworth Sleepiness Scale (ESS), Women’s Health Initiative Insomnia Rating Scale (WHIIRS), self-reported average sleep duration, restless legs symptoms, napping frequency, and sleep quality.

Results

Mean (M) age was 56.4±8.1 years, and 54.7% were female. Average AHI, ESS, WHIIRS, and sleep duration were 8.7±13.1, 6.0±5.0, 7.6±5.5, and 7.8±1.4, respectively, and 25.8% had zero percent time SpO2 &lt;90%. Fit statistics indicated that a four-class solution provided the best data fit. The derived classes, adjusting for age, sex, income, and acculturation, corresponded with four clinically meaningful groups: (1) 28.8% were asymptomatic [(M) AHI=0.8; (M) ESS=5.6; (M)WHIIRS=7.6; (M) sleep duration=7.8; 0% SpO2&lt;90%=74.1%], (2) 25.7 % were asymptomatic mild sleep apnea [(M) AHI=6.2; (M) ESS=3.8; (M) WHIIRS=2.9; (M) sleep duration=7.8; 0% SpO2&lt;90%=8.8%], (3) 19.4% were symptomatic sleep apnea [(M) AHI=25.6; (M) ESS=8.5; (M) WHIIRS=7.2; (M) sleep duration=7.7; 0% SpO2&lt;90%= 0.5%], and (4) 26.1% were insomnia [(M) AHI=5.7; (M) ESS=6.7; (M) WHIIRS=13.0; (M) sleep duration=7.8; 0% SpO2&lt;90%=10.3%]. Classification into groups 3 and 4 were primarily driven by elevated AHI and WHIIRS scores, respectively. The distribution of scores in the derived groups suggest variations relative to current clinical thresholds.

Conclusion

We identified 4-groups using LCA in a community-based sample of diverse U.S. Hispanic/Latino adults. Better characterization of sleep phenotypes for Hispanics/Latinos can help in developing targeted interventions studies and ameliorate health disparities.

Support

5R01AG048642-05; R21AG056952; R21HL140437.

0607 Clinical Phenotypes of OSA in Diverse Hispanics/Latinos: Results from the Hispanic Community Health Study/Study of Latinos

Introduction

Recent work on US non-Latino Whites and Europeans from clinical samples used obstructive sleep apnea (OSA) symptoms to generate OSA phenotypes for individuals with moderate-severe OSA and proposed between 3-5 clusters. Validating these clusters in a diverse Hispanic/Latino community-based population with different biopsychosocial characteristics is crucial for early OSA identification and more personalized treatment.

Methods

This work is based on baseline data from The Hispanic Community Health Study/Study of Latinos (HCHS/SOL). HCHS/SOL is a prospective cohort study designed using a multisite (Bronx, NY, Chicago, IL, Miami, FL, San Diego, CA) multistage probability sample. The subpopulation of interest included adults 18-74 years (unweighted n=1,623) meeting criteria for moderate-severe OSA symptoms (≥15 Apnea-Hypopnea index (AHI) events per hour). We performed latent class analysis (LCA) using 15 common OSA symptoms to identify phenotype clusters.

Results

Average age was 52.4 ± 13.9 years and 34.1% were female. Mean AHI was 33.8 ± 22.5 events per hour. Fit statistics and clinical significance suggested that a three-class solution provided best fit to the data. The symptom profiles were consistent with (1) a Minimally Symptomatic group (46.8%), (2) a Disturbed Sleep group (38.1%), and (3) a Daytime Sleepiness group (15.1%). Validation analyses using alternative hierarchical and partitioning algorithms also suggested support for a three-class solution.

Conclusion

Sleep apnea phenotypes among diverse Hispanics/Latinos were consistent with recent findings from the Sleep Apnea Global Interdisciplinary Consortium. However, we found notable differences in the prevalence of these clusters relative to Whites. This suggests that other biopsychosocial factors may be contributing to OSA phenotypes among Hispanics/Latinos. Identification of OSA phenotypes in Hispanics/Latinos could inform better sleep interventions and therapeutics and help better align public health resources.

Support

5R01AG048642-05; R21AG056952; R21HL140437.

Educational attainment and coronary and aortic calcification in postmenopausal women

OBJECTIVE

Low socioeconomic status is a risk factor for clinical coronary heart disease, a relatively crude outcome associated with important biases. By avoiding these biases, subclinical assessments could facilitate efforts to understand the association between socioeconomic status and coronary disease. The current study 1) evaluated the nature of the associations between educational attainment and subclinical atherosclerosis and 2) examined if biologic, behavioral, and psychosocial factors mediated these associations.

METHODS

Participants were 308 women from the Healthy Women Study who underwent a clinic examination of risk factors either 5 (N = 32) or 8 (N = 276) years after the menopausal transition. Aortic and coronary calcification were measured using electron beam tomography.

RESULTS

Logistic regression analysis with orthogonal polynomials revealed a marginally significant linear trend for coronary calcification, with the more educated groups showing lower calcification than the less educated groups. A significant linear trend was also observed for aortic calcification. In addition, a marginally significant quadratic trend was observed for aortic calcification so that the effect began to reverse at the highest level of education. Measured risk factors were associated with education and with the calcification outcomes, but they explained little of the associations between educational attainment and coronary or aortic calcification. None of the factors tested met the minimum criterion for mediation.

CONCLUSIONS

The findings show that lower education is associated with greater early stage atherosclerosis. Subclinical assessments, such as electron beam tomography, represent useful alternatives for studies of socioeconomic status and coronary artery disease.

Cardiovascular and electrodermal responses to support and provocation: interpersonal methods in the study of psychophysiological reactivity

This study examined the joint and independent effects of experimentally manipulated social contexts and individual differences in hostility and perceived social support on physiological responses to a social stressor, while illustrating the use of the interpersonal circumplex for integrative social psychophysiological research. Undergraduate women completed a speech task in a supportive, neutral, or provoking context and completed measures of hostility and perceived social support. The provoking context evoked the largest blood pressure and heart rate (HR) responses, followed by the neutral and the supportive context. Social context also influenced HR and electrodermal reactivity during task preparation. Hostility elicited higher systolic blood pressure (SBP) reactivity during preparation, speech, and recovery. Perceived social support interacted with context to affect SBP and HR during speech and preparation. The roles of interpersonal characteristics and contexts in the physiological stress response and the utility of interpersonal methods in studying these associations are discussed.

Do negative emotions mediate the association between socioeconomic status and health?

In this chapter, we examine the possibility that negative emotions contribute to the relationship between socioeconomic status (SES) and health. A model of the associations among SES, emotion, and health is presented first. We then review the evidence for this model, showing associations of SES with depression, hopelessness, anxiety, and hostile affect and cognition, and of these negative emotions with disease. Notably, most of the data supporting the model provide only indirect evidence that negative emotions serve as a key contributor to the proposed associations. We, therefore, conclude with recommendations for longitudinal research, especially in children, that will more directly and comprehensively examine negative emotions as possible mediators of the SES and health relationship.

Hostility and cardiovascular reactivity during marital interaction

OBJECTIVE

Prior studies demonstrate that hostile persons respond to social stressors with heightened cardiovascular responses. This study examined the effects of individual differences in hostility and two experimentally manipulated social stressors on cardiovascular reactivity during marital interaction.

METHODS

Sixty couples participated in a discussion task under conditions of high or low evaluative threat and while either agreeing or disagreeing with each other. Individual differences in hostility were assessed with the Buss-Perry Aggression Questionnaire. Participants' appraisal of their spouses' behavior during the interaction task was assessed with a standardized measure. Systolic and diastolic blood pressure and heart rate responses were recorded.

RESULTS

Among husbands, hostility was associated with greater systolic blood pressure reactivity under high, but not low, threat. Appraisals suggested that this might be due to husbands' efforts to assert dominance in the interaction. Wives' hostility scores were unrelated to cardiovascular reactivity, but wives disagreeing with hostile husbands showed greater heart rate reactivity.

CONCLUSIONS

Heightened cardiovascular reactivity to stressful marital interactions among hostile men provides additional evidence of the viability of this psychophysiologic mechanism as a link between hostility and health. The lack of effects among wives suggests sex differences in the social psychophysiology of hostility. Interpersonal concepts and methods are useful in the study of psychosocial risk factors and mechanisms.

Agency, communion, and cardiovascular reactivity during marital interaction

The concepts of agency and communion have been used to describe sex differences in vulnerability to specific stressor domains. This study examined blood pressure and heart rate responses of 60 married couples to experimental manipulations of disagreement (i.e., communion stressor) and achievement challenge (i.e., agency stressor). Consistent with predictions, disagreement elicited heightened cardiovascular reactivity among wives, but not husbands. In contrast, the achievement challenge elicited heightened cardiovascular reactivity among husbands, but not wives. Participants' responses to a circumplex measure of interpersonal appraisal were consistent with the interpretation of differential responses to agency and communion stressors. Results are congruent with a situational approach to sex differences in cardiovascular reactivity and illustrate the utility of interpersonal methods in the explication of psychosocial risk for cardiovascular disease.

Agency, communion, and cardiovascular reactivity during marital interaction

The concepts of agency and communion have been used to describe sex differences in vulnerability to specific stressor domains. This study examined blood pressure and heart rate responses of 60 married couples to experimental manipulations of disagreement (i.e., communion stressor) and achievement challenge (i.e., agency stressor). Consistent with predictions, disagreement elicited heightened cardiovascular reactivity among wives, but not husbands. In contrast, the achievement challenge elicited heightened cardiovascular reactivity among husbands, but not wives. Participants' responses to a circumplex measure of interpersonal appraisal were consistent with the interpretation of differential responses to agency and communion stressors. Results are congruent with a situational approach to sex differences in cardiovascular reactivity and illustrate the utility of interpersonal methods in the explication of psychosocial risk for cardiovascular disease.

Construct validation of health-relevant personality traits: Interpersonal circumplex and five-factor model analyses of the aggression questionnaire

The general literature on personality traits as risk factors for physical illness--as well as the specific literature on health consequences of anger, hostility, and aggressive behavior--often suffers from incomplete or inconsistent construct validation of personality measures. This study illustrates the utility of two conceptual tools in this regard--the five-factor model and the interpersonal circumplex. The similarities and differences among anger, hostility, verbal aggressiveness, and physical aggressiveness as measured by the Buss and Perry (1992) Aggression Questionnaire were identified. Results support the interpretation of anger and hostility as primarily reflecting neurotic hostility and antagonistic hostility to a lesser extent. In contrast, verbal and physical aggressiveness can he seen as primarily reflecting antagonistic hostility, and to a lesser extent neurotic hostility. Further, verbal aggressiveness was associated with hostile dominance, whereas hostility was associated with hostile submissiveness. These findings identify potentially important distinctions among these related constructs and illustrate the potential integrative value of standard validation procedures.

Interpersonal control and cardiovascular reactivity: goals, behavioral expression, and the moderating effects of sex

Cardiovascular responses to stressful stimuli have been implicated in the development of cardiovascular disease. However, the effects of social stressors on short-term changes in blood pressure and heart rate (i.e., cardiovascular reactivity [CVR]) are not well understood. The independent effects of an incentive to exert interpersonal influence and the expression of socially controlling behavior on CVR were examined in 96 undergraduates. For men, both the incentive to exert influence and the enactment of a controlling interpersonal style produced larger increases in systolic blood pressure. By contrast, although the incentive to be influential increased women's CVR, the enactment of a cooperative role produced the largest increases in blood pressure among women. The effects of social dominance on CVR, sex differences in CVR, and interpersonal approaches to the study of these psychophysiological mechanisms are discussed.

Interpersonal control and cardiovascular reactivity: goals, behavioral expression, and the moderating effects of sex

Cardiovascular responses to stressful stimuli have been implicated in the development of cardiovascular disease. However, the effects of social stressors on short-term changes in blood pressure and heart rate (i.e., cardiovascular reactivity [CVR]) are not well understood. The independent effects of an incentive to exert interpersonal influence and the expression of socially controlling behavior on CVR were examined in 96 undergraduates. For men, both the incentive to exert influence and the enactment of a controlling interpersonal style produced larger increases in systolic blood pressure. By contrast, although the incentive to be influential increased women's CVR, the enactment of a cooperative role produced the largest increases in blood pressure among women. The effects of social dominance on CVR, sex differences in CVR, and interpersonal approaches to the study of these psychophysiological mechanisms are discussed.

The circadian rhythm of temperature during light treatment for winter depression

Circadian rhythm abnormalities have been implicated in winter seasonal affective disorder. We examined the circadian temperature rhythm of 22 patients with winter depression and 10 normal controls who had participated in various high-intensity light treatment experiments. We did not find abnormalities in the baseline phase or amplitude of the temperature rhythm in patients compared to controls. Nor did we find abnormalities in the phase-shifting response to morning light. There was some evidence that the "phase-delayed" half of the patients responded poorly to phase advances produced by morning light, whereas the "phase-advanced" half of the patients responded poorly when their rhythms delayed. However, the antidepressant responses during the best week (week of lowest depression score) were unrelated to temperature rhythm phase shifts. In general, there was not strong support for a relationship between circadian rhythms changes and antidepressant response.

Circadian rhythms during gradually delaying and advancing sleep and light schedules

The circadian rhythms of the night shift worker show very little phase shift in response to the daytime sleep and night work schedule. One strategy for producing circadian adaptation may be to use appropriately timed exposure to high-intensity light. We attempted to shift the circadian temperature rhythms of seven normal subjects while they followed a sleep schedule that gradually delayed (2 h per day) until sleep occurred during the daytime, as is customary for workers during the night shift. After 5 days, the sleep schedule was gradually advanced back to baseline. High illuminance light (2 h per day) and the attenuation or avoidance of sunlight were timed to facilitate temperature rhythm phase shifts. In general, the temperature rhythm did not shift along with the sleep-wake schedule, but appeared either to free run or remain entrained to the natural 24-h zeitgebers. This study showed how difficult it can be to shift human circadian rhythms in the field, when subjects are exposed to competing 24-hr zeitgebers.