Faculty Bibliography

BACKGROUND:Mobile phone and tablet ownership have increased in the United States over the last decade, contributing to the growing use of mobile health (mHealth) interventions to help patients manage chronic health conditions like diabetes. However, few studies have characterized mobile device ownership and the presence of health-related apps on mobile devices in people with a self-reported history of hypertension. OBJECTIVE:This study aimed to describe the prevalence of smartphone, tablet, and basic mobile phone ownership and the presence of health apps by sociodemographic factors and self-reported hypertension status (ie, history) in a nationally representative sample of US adults, and to describe whether mobile devices are associated with health goal achievement, medical decision making, and patient-provider communication. METHODS:Data from 3285 respondents from the 2017 Health Information National Trends Survey were analyzed. Participants were asked if they owned a smartphone, tablet, or basic mobile phone and if they had health apps on a smartphone or tablet. Participants were also asked if their smartphones or tablets helped them achieve a health-related goal like losing weight, make a decision about how to treat an illness, or talk with their health care providers. Chi-square analyses were conducted to test for differences in mobile device ownership, health app presence, and app helpfulness by patient characteristics. RESULTS:Approximately 1460 (37.6% weighted prevalence) participants reported a history of hypertension. Tablet and smartphone ownership were lower in participants with a history of hypertension than in those without a history of hypertension (55% vs 66%, P=.001, and 86% vs 68%, P<.001, respectively). Participants with a history of hypertension were more likely to own a basic mobile phone only as compared to those without a history of hypertension (16% vs 9%, P<.001). Among those with a history of hypertension exclusively, basic mobile phone, smartphone, and tablet ownership were associated with age and education, but not race or sex. Older adults were more likely to report having a basic mobile phone only, whereas those with higher education were more likely to report owning a tablet or smartphone. Compared to those without a history of hypertension, participants with a history of hypertension were less likely to have health-related apps on their smartphones or tablets (45% vs 30%, P<.001) and report that mobile devices helped them achieve a health-related goal (72% vs 63%, P=.01). CONCLUSIONS:Despite the increasing use of smartphones, tablets, and health-related apps, these tools are used less among people with a self-reported history of hypertension. To reach the widest cross-section of patients, a mix of novel mHealth interventions and traditional health communication strategies (eg, print, web based, and in person) are needed to support the diverse needs of people with a history of hypertension.

Introduction: Child fatigue has been associated with behavioral outcomes, including aggression, hyperactivity, and conduct problems, which may affect academic performance. We explored whether fatigue was associated with external behavioral health outcomes in a predominantly Black (Afro-Caribbean and African-American) student population (90%). Ratings of parent and teacher agreement of child fatigue was evaluated. This analysis was part of a larger research program, which included a cluster randomized controlled trial in ten public elementary schools in historically disinvested neighborhoods.
Method(s): A total of 804 first-graders (7+/- 0.6 years old) participated in the study focused on child self-regulation, mental health achievement, parenting and parent involvement. Externalizing behaviors (i.e., conduct problems, aggression, and hyperactivity) were reported by teachers using the Behavior System for Children (BASC-2). A composite score of teacher-perceived child fatigue was created based on ratings of child fatigue, morning alertness, and falling asleep in class. Parent perception of child fatigue was assessed using the Children's Sleep Habits Questionnaire. Regression analysis was conducted to determine the association between teacher's reports of child fatigue and externalizing behavior problems. Cohen's kappa coefficient assessed parent and teacher agreement of child fatigue based on categorical classification of presence of child fatigue.
Result(s): Children who were perceived as fatigued (i.e., tiredness and falling asleep in class) by their teacher were more likely to have a high BASC externalizing composite score (T=60 cut off) (beta = -0.24, p<.001). Cohen's kappa of 0.004 (p<0.05) showed a slight discordance in perception of child fatigue comparing reports from teachers and parents, although results were not significant.
Conclusion(s): Teacher perception of child fatigue was significantly associated with teacher BASC T-score of child externalizing behavior outcomes. Future studies should explore longitudinal relationships between fatigue and mental health

Introduction: Sleep health is associated with many domains of functioning. Yet, changing behaviors linked to improved sleep health is difficult. Beliefs about the health impact of sleep may motivate behavior change. This analysis examined which beliefs about sleep might motivate sleep behavior change.
Method(s): Data were from the Sleep and Healthy Activity, Diet, Environment, and Socialization (SHADES) study, consisting of N=1007 community-dwelling adults age 22-60. Participants were asked, regarding "the single most important thing you personally could do to improve your sleep," whether participants were in the stage of precontemplation (not considered change), contemplation (considered but not decided), preparation (decided but not acting), and action stages of change from the transtheoretical model. They were also asked items from the Sleep Practices and Attitudes Questionnaire (SPAQ) regarding the degree to which they agree with whether "not getting enough sleep" can cause sleepiness, drowsy driving, weight gain, heart disease, high cholesterol, hypertension, moodiness, lower energy, decreased sex drive, missed days at work, decreased performance, memory/concentration problems, diabetes, and/or tiredness. Ordinal logistic regressions evaluated increased likelihood of stage of change, based on degree of agreement with those statements, adjusted for age, sex, race/ethnicity, and education. Post-hoc analyses also examined sleep duration as an additional covariate.
Result(s): In adjusted analyses, stage of change was associated with degree of agreement that insufficient sleep can cause sleepiness (OR=1.17, p=0.035), weight gain (OR=1.20, p<0.0005), heart disease (OR=1.21, p=0.001), cholesterol (OR=1.13, p=0.047), hypertension (OR=1.16, p=0.014), moodiness (OR=1.42, p<0.0005), decreased energy (OR=1.30, p=0.002), absenteeism (OR=1.13, p=0.007), decreased performance (OR=1.20, p=0.003), concentration/ memory problems (OR=1.23, p=0.004), diabetes (OR=1.14, p=0.042), and feeling tired (OR=1.39, p<0.0005). When sleep duration was added to the model, significant relationships remained for weight, heart, hypertension, moodiness, energy, absenteeism, performance, memory, diabetes, and tiredness.
Conclusion(s): Degree of belief that insufficient sleep can cause outcomes such as moodiness, occupational problems, and health problems may impact whether an individual is contemplating/ attempting to improve their sleep. This may guide education/outreach efforts

Introduction: Sleep is related to socioeconomic status and impacts health. This study evaluated whether foregoing medical care due to cost impacts sleep and plays a role in sleep disparities and the sleep-obesity relationship.
Method(s): Data from the 2017 Behavioral Risk Factor Surveillance System (N=39,267 from 7 states). Sleep duration was assessed as hours/day. Participants were asked, "Was there a time in the past 12 months when you needed to see a doctor but could not because of cost?" They were also asked for information about age, sex, race/ethnicity, education, income, employment, overall health, and access to health insurance. They were also asked for height/weight, which was used to compute body mass index (BMI).
Result(s): Access to health insurance was not associated with habitual sleep duration. However, foregoing medical care was associated with less sleep (B=-0.26, 95%CI[-0.35,-0.17], p<0.0005). There was an interaction with race/ethnicity; compared to non-Hispanic Whites, the effect was 115% larger among Blacks/African-Americans, 13% larger in Hispanics/Latinos, 101% larger and in the opposite direction for Asians, and non-significant for Multiracial. Race/ethnicity relationships to sleep duration were stratified by foregoing care. Among those who did not (90%), both short and long sleep duration were more likely among Blacks/African-Americans and other minority groups. Among those who did forego care (10%), these effects were dramatically reduced. Further, when sleep duration was evaluated as a predictor of obesity, this relationship was only seen among those who did not forego care.
Conclusion(s): Foregoing medical care due to cost is an independent risk factor for insufficient sleep, irrespective of income, employment, and access to insurance. It disproportionately affects Blacks/ African-Americans and may represent part of the reason why sleep disparities exist even after adjustment for most socioeconomic indices. Further, foregoing medical care may present such health risks that this subsumes the relationship between sleep and obesity

Introduction: Previous studies have shown that racial/ethnic minorities are more likely to be short and/or long sleepers, which may increase risk for morbidity/mortality. This analysis provides a more recent update from a very large national dataset, including representation of additional groups and examination of the role of poverty.
Method(s): Data from the 2016 Behavioral Risk Factor Surveillance System (BRFSS, collected by the CDC) were used. N=464,671 adults >18yrs from all US states/territories provided data on sleep, demographics, and socioeconomics. Sleep duration was categorized as very short (<=4h), short (5-6h), normal (7-8h as reference), and long (>=9h). Race/ethnicity was self-reported as Non-Hispanic White, Black/ African-American, Hispanic/Latino, American-Indian/Alaskan-Native (AIAN), Native Hawaiian/Pacific-Islander (NHPI), or Multiracial/Other. Covariates included age, sex, relationship status, education, employment, and home ownership. Interactions were explored with poverty (income<$20,000) were explored. Multinomial logistic regressions were weighted using BRFSS-specific weights.
Result(s): A significant race-by-poverty interaction was seen (p<0.0005). Compared to non-poor Non-Hispanic White, increased very short sleep was seen among those who were non-poor Black/African-American (RRR=2.1, p<0.0005), Asian (RRR=1.6, p=0.001), AIAN (RRR=1.4, p=0.001), NHPI (RRR=2.0, p=0.002), and Multiracial/Other (RRR=2.2, p<0.0005), and poor Non-Hispanic White (RRR=1.8, p<0.0005), Black/African-American (RRR=1.8, p<0.0005), AIAN (RRR=1.5, p=0.007), NHPI (RRR=2.4, p=0.005), and Multiracial/Other (RRR=3.4, p<0.0005). Compared to non-poor White, increased short sleep was seen among non-poor Black/African-American (RRR=1.7, p<0.0005), Asian (RRR=1.3, p<0.0005), AIAN (RRR=1.2, p=0.02), NHPI (RRR=1.3, p=0.02), Multiracial/Other (RRR=1.3, p<0.0005), and poor Non-Hispanic White (RRR=1.3, p<0.0005), Black/African-American (RRR=1.4, p<0.0005), Asian (RRR=1.3, p=0.04), and Multiracial/Other (RRR=2.2, p<0.0005). Compared to non-poor Non-Hispanic White, increased long sleep was seen for Non-Poor Black/African-American (RRR=1.4, p<0.0005), Poor Non-Hispanic White (RRR=1.3, p<0.0005), Black/African-American (RRR=1.4, p<0.0005), and AIAN (RRR=1.3, p<0.05).
Conclusion(s): Established racial/ethnic sleep disparities are supported in this large national sample, with additional information on understudied vulnerable groups including AI/AN and NH/PI. Further, the this study as the contribution of poverty status

Introduction: We determined whether Obstructive Sleep Apnea (OSA) and beta-Amyloid Burden (Abeta) act additively or synergistically to promote conversion from cognitive normal (CN) to mild cognitive impairment (MCI) and from MCI to AD.
Method(s): In this longitudinal observational study, we examined CN (n=298) and MCI (n=418) older adults from the ADNI database (adni.loni.usc.edu). OSA was self-reported during a clinical interview. Brain Abeta was assessed using Florbetapir-PET imaging. The primary outcome of the analysis was conversion from CN to MCI (CN participants) and from MCI to AD (MCI participants). Participants were required to have a baseline and at least one follow-up clinical visit that identified their cognitive status. Logistic mixed-effects models with random intercept and slope were used to assess associations between OSA, Abeta, and risk of conversion from CN to MCI, and MCI to AD. All models included age at baseline, sex, APOE4 status, years of education, and their interactions with time.
Result(s): Of the 716 participants, 329 (46%) were women. The overall mean (SD) age was 74.7 (5.0) years, and the overall mean (SD) follow-up time was 5.5 (1.7) years (Range: 2.7 - 10.9 years). In CN participants at baseline, conversion to MCI was associated with both OSA (beta = 0.418; 95% CI, 0.133 to 0.703; P < .001) and higher Abeta-burden (beta = 0.554; 95% CI, 0.215 to 0.892; P < .001). The interaction of OSA and Abeta burden with time was significant (beta = 1.169, 95% CI, 0.776 to 1.562; P < .001), suggesting a synergistic effect. In MCI participants at baseline, conversion to AD was associated with both OSA (beta = 0.637; 95% CI, 0.291 to 0.982; P < .001) and higher Abeta-burden (beta = 1.061; 95% CI, 0.625 to 1.497; P < .001). The interaction of OSA and Abeta burden with time was significant (beta = 1.312, 95% CI, 0.952 to 1.671; P < .001), suggesting a synergistic effect.
Conclusion(s): In both CN and MCI elderly, Abeta modified the risk of progression to AD in OSA participants. OSA patients maybe more physiologically susceptible as Abeta load becomes increasingly abnormal

Introduction: Population estimates for sleep duration and quality are inconsistent because they rely on self-report and smaller samples using objective data. Tracking and wearable devices may provide more accurate estimates of sleep duration and quality. In this study, we investigated estimates of sleep duration and quality among 2 million users of a mobile sleep application in an urban city in the United States (U.S.).
Method(s): We examined sleep duration and quality from 2,194,897 users of SleepCycle, a popular sleep tracking app. over a four-year period (2015 to 2018). In this analysis, we specifically explored differences in sleep duration and quality by demographic factors, including age and sex. We utilized graphical matrix representations of data (heat maps) and geospatial analyses to compare sleep duration (in hours) and sleep quality (based on a composite score amalgamating time in bed, deep-sleep time, sleep consistency, and amount of times fully awake), considering potential effects of day of the week and seasonality.
Result(s): Among SleepCycle users, 45.6% (n=1,001,335) were female. The average age of the sample was 31.0 years. The mean sleep duration of the total sample was 7.11 hours; women slept longer than did men (M=7.27 hours vs. M=7.00 hours, p<.001). Increasing age tended to be associated with longer sleep duration and better sleep quality. Results also showed sleep duration was longer on weekends (M=7.19 hours), compared to weeknights (M=7.09 hours). Sleep duration was longest (M=7.18 hours) during the winter, but shortest during the summer (M=7.11 hours). Sleep quality was highest (M=72.75) during the winter, but lowest during the summer (M=71.99).
Conclusion(s): Our findings from big data are consistent with previously reported estimates of sleep duration and quality. Sleep duration varied by age, sex, day of the week, and season. Future studies should determine whether estimates of sleep duration and quality are affected by environmental factors such as geographic location

Introduction: We determined whether the co-occurrence of OSA and hypertension interact synergistically to promote beta-Amyloid burden and cognitive decline in clinically normal older adults Methods: Prospective longitudinal study utilizing NYU cohort of community-dwelling cognitively-normal elderly, with baseline and at least one follow-up of CSF-Abeta42 (measured using ELISA), and neuropsychological visits. OSA was defined using AHI4%. Hypertension diagnosis was according to AHA-guidelines. Cognitive variables assessed included Logic-2, Animal-Fluency [AF], Vegetable-Fluency [VF]), Boston-Naming-Test [BNT], Digit-Symbol-Substitution-Test [DSST], Trails Making Test-A and B [TMT-A and B]). Linear mixed-effects models with random intercept and slope were used to assess associations between OSA, hypertension, and longitudinal changes in CSF-Abeta and cognition, controlling for age-at-baseline, sex, APOE4-status, years-of-education, and their interactions with time.
Result(s): Of the 98 participants, 63 (64.3%) were women. The mean (SD) age was 69.6 (7.3) years and follow-up time was 2.46 (0.64) years. OSA and hypertension were each associated with faster rate-of-change in CSF-Abeta42 (beta = -3.11; 95%CI, -3.71, -2.51; and beta= -2.82, 95% CI -3.29, -2.35, P < .01 for both respectively). The interaction of OSA and hypertension with time was significant (beta= -1.28, 95% CI -1.78 to -0.78, P < .01) suggesting a synergistic effect. No significant associations were seen between annual-changes in CSF-Abeta42 and cognitive-decline. However, faster decline in VF, and DSST were associated with OSA (beta = -0.054; 95%CI, -0.094, -0.013; P = .02; beta = -0.058; 95%CI, -0.084, -0.033; P < .05 for both respectively), and with hypertension (beta = -0.048; 95%CI, -0.079, -0.017; P = .04; beta = -0.078; 95%CI, -0.098, -0.057; P = .002; respectively). The interaction of OSA and hypertension with time was significant for both VF and DSST (beta = -0.033, 95%CI, -0.048, -0.018; P < .001 and beta = -0.040, 95%CI, -0.064, -0.016; P < .001, respectively), suggesting a synergistic effect.
Conclusion(s): In cognitive-normal elderly OSA individuals, vascular risk may complement AD-biomarkers in assessing risk of prospective cognitive-decline in preclinical AD

Introduction: Based on principles of community-based participatory research methods (CBPR), a community-oriented framework was applied in three studies that focused on African- Americans/ Blacks (herein referred to as Blacks): The Metabolic Syndrome Outcome Study (MetSO), Tailored Approach to Sleep Health Education (TASHE), and Peer-Based Sleep Health Education and Social Support (PEERS-ED). We describe results of our application of this framework to enroll and study Blacks in these NIHfunded studies of obstructive sleep apnea (OSA).
Method(s): Our community-oriented framework includes strategic guidelines for effective intervention to engage communities in research and ensure cultural and linguistic appropriateness of sleep messages in behavioral interventions. Strategies included: 1) focus groups and in-depth interviews with key stakeholders; 2) establishing a community advisory board; 3) conducting Delphi surveys to identify high-priority diseases and conditions. Community barriers were identified through an iterative process using surveys and focus groups. Stakeholder groups were integral during the development, implementation and dissemination, reflecting a patient-oriented decision-making process with respect to key intervention components.
Result(s): MetSO, TASHE, and PEERS-ED reached nearly 3,000 Blacks at risk of OSA in New York City. Of those, 2,000 were screened for OSA. Sleep brochures were distributed to over 10,000 individuals. The mean age of community participants was 62+/-14 years; 69% were female; 43% had an annual income <$10,000; and 37% had 10); 10% reported an insomnia diagnosis and 12% used sleep medications. Based on WatchPAT data, 24% had moderate OSA and 18%, severe OSA. Compared to blacks receiving standard sleep messages, those exposed to tailored sleep messages in our interventions were nearly 4 times as likely to adhere to OSA care.
Conclusion(s): Community outreach may be an effective strategy in the reach and spread of sleep messages among low-income Blacks at-risk for OSA