Use and misuse of random forest variable importance metrics in medicine: demonstrations through incident stroke prediction.

BACKGROUND: Machine learning tools such as random forests provide important opportunities for modeling large, complex modern data generated in medicine. Unfortunately, when it comes to understanding why machine learning models are predictive, applied research continues to rely on 'out of bag' (OOB) variable importance metrics (VIMPs) that are known to have considerable shortcomings within the statistics community. After explaining the limitations of OOB VIMPs - including bias towards correlated features and limited interpretability - we describe a modern approach called 'knockoff VIMPs' and explain its advantages. METHODS: We first evaluate current VIMP practices through an in-depth literature review of 50 recent random forest manuscripts. Next, we recommend organized and interpretable strategies for analysis with knockoff VIMPs, including computing them for groups of features and considering multiple model performance metrics. To demonstrate methods, we develop a random forest to predict 5-year incident stroke in the Sleep Heart Health Study and compare results based on OOB and knockoff VIMPs. RESULTS: Nearly all papers in the literature review contained substantial limitations in their use of VIMPs. In our demonstration, using OOB VIMPs for individual variables suggested two highly correlated lung function variables (forced expiratory volume, forced vital capacity) as the best predictors of incident stroke, followed by age and height. Using an organized analytic approach that considered knockoff VIMPs of both groups of features and individual features, the largest contributions to model sensitivity were medications (especially cardiovascular) and measured medical risk factors, while the largest contributions to model specificity were age, diastolic blood pressure, self-reported medical risk factors, polysomnography features, and pack-years of smoking. Thus, we reach very different conclusions about stroke risk factors using OOB VIMPs versus knockoff VIMPs. CONCLUSIONS: The near-ubiquitous reliance on OOB VIMPs may provide misleading results for researchers who use such methods to guide their research. Given the rapid pace of scientific inquiry using machine learning, it is essential to bring modern knockoff VIMPs that are interpretable and unbiased into widespread applied practice to steer researchers using random forest machine learning toward more meaningful results.

Examining the diagnostic validity of the Berlin Questionnaire in a low-income Black American sample.

STUDY OBJECTIVES: Black individuals and individuals of low socioeconomic status are at increased risk for obstructive sleep apnea (OSA). The Berlin Questionnaire is one of the most widely used screening tools for OSA; however, there is limited research on its diagnostic accuracy in low-income Black populations. METHODS: This study analyzed data from an ongoing study taking place among a cohort from 2 predominantly Black neighborhoods in Pittsburgh, Pennsylvania (96.3% Black, 79.6% female). The sample included 269 individuals without a prior diagnosis of OSA who completed the Berlin Questionnaire and also participated in a home sleep apnea test. An apnea-hypopnea index ≥ 15 events/h was used to identify individuals with moderate or severe OSA. RESULTS: 19.3% of individuals met criteria for moderate to severe OSA based on home sleep apnea test, while 31.2% of participants screened as high risk for OSA based on the overall Berlin index. Using apnea-hypopnea index ≥ 15 events/h as the reference standard, the Berlin Questionnaire had a sensitivity of 46.2%, specificity of 72.4%, positive predictive value of 28.6%, and negative predictive value of 84.9% among this sample. Analyses stratified by sex suggested that the Berlin Questionnaire had better diagnostic validity in women than men. CONCLUSIONS: The Berlin Questionnaire has lower sensitivity and positive predictive value in our sample than those observed in general population samples. The measure performed better among women, though a higher proportion of men fell into the moderate or severe OSA range based on the home sleep apnea test. Given the significant downstream consequences of OSA, utilizing screening tools that better detect OSA in Black communities is key. CITATION: Holliday SB, Haas A, Dong L, et al. Examining the diagnostic validity of the Berlin Questionnaire in a low-income Black American sample. J Clin Sleep Med. 2021;17(10):1987-1994.

Demographic, Pregnancy-Related, and Health-Related Factors in Association with Changes in Sleep Among Pregnant Women with Overweight or Obesity.

BACKGROUND: Women with prepregnancy overweight/obesity are at high risk for obstetric complications and cardiometabolic disease. Poorer sleep quality is associated with obesity in non-pregnant individuals and, during pregnancy, poor sleep predicts negative obstetric and health outcomes. This study examined sleep patterns among women with overweight/obesity and factors associated with different sleep trajectories during pregnancy. METHODS: Women (N = 146, 17-40 years old) with a prepregnancy body mass index ≥ 25 kg/m2 were recruited during early pregnancy. Participants reported demographic information and completed the Pittsburgh Sleep Quality Index (PSQI) at up to six monthly assessments, with the first assessment occurring between 12 and 20 weeks gestation and the final assessment between 35 weeks gestation and delivery. PSQI scores > 5 indicate "poor sleep." RESULTS: On average, women's PSQI scores were 6.66 ± 3.58 in the first half of pregnancy and were significantly higher (worse) at the end of pregnancy (t(644) = 4.55, p < 0.001), with the greatest change occurring in the third trimester (t(636.3) = 3.72, p < 0.001). Women who currently smoked had poorer sleep than women who did not currently smoke (t(1) = 2.29, p = 0.02). Prepregnancy weight status, age, parity, race, education, and income were not significantly associated with sleep changes (t(1) < 1.76, ps > 0.08). The percentage of women with PSQI scores > 5 (the threshold for poor sleep quality) was 37-63% across assessments, with the greatest increase occurring during the third trimester (t(633) = 2.92, p = 0.004). CONCLUSIONS: Sleep quality worsens during the third trimester and is associated with current smoking. Future studies of sleep during pregnancy should examine health outcomes among women with overweight/obesity and early intervention to mitigate sleep disturbances as pregnancy progresses.

Prevalence and correlates of obstructive sleep apnea in urban-dwelling, low-income, predominantly African-American women.

STUDY OBJECTIVES: The current study examined the prevalence and correlates of obstructive sleep apnea in a sample of low-income, predominantly African-American women using two waves of data. METHODS: Participants were adults from two urban neighborhoods who enrolled in the PHRESH Zzz Study (N = 828; Pittsburgh Hill/Homewood Research on Neighborhoods, Sleep, and Health). A subsample who reported never receiving OSA diagnosis completed home sleep apnea testing in 2016 (n = 269, mean age 55.0 years, 79.6% female) and again in 2018 (n = 135). Correlates of OSA tested included demographic and anthropometric variables, health behavior/conditions, psychological distress and general health, smoking status, actigraphy-measured sleep, and neighborhood factors measured at baseline. RESULTS: 18.0% of all 2016 participants reported receiving physician diagnoses of OSA. Among those who completed in-home assessment, 19.3% had AHI ≥15 and 33.8% had AHI ≥5 plus one or more sleep symptoms. Estimates of the prevalence of OSA in all 2016 participants were 33.8%-45.7% based on physician diagnoses and AHI results, depending on the criteria used. Age, gender, BMI, blood pressure, habitual snoring, neighborhood walkability, actigraphy-measured sleep characteristics, and smoking were concurrently associated with OSA in 2016. Changes in AHI categories from 2016 to 2018 were documented. CONCLUSIONS: Low-income African Americans, including women, are a high-risk group for OSA, but remain under-diagnosed and under-treated. The current findings show a high prevalence of OSA in African-American women and are among the first to demonstrate that both individual and neighborhood factors are implicated in OSA prevalence.

Prioritized Research for the Prevention, Treatment, and Reversal of Chronic Disease: Recommendations From the Lifestyle Medicine Research Summit.

Declining life expectancy and increasing all-cause mortality in the United States have been associated with unhealthy behaviors, socioecological factors, and preventable disease. A growing body of basic science, clinical research, and population health evidence points to the benefits of healthy behaviors, environments and policies to maintain health and prevent, treat, and reverse the root causes of common chronic diseases. Similarly, innovations in research methodologies, standards of evidence, emergence of unique study cohorts, and breakthroughs in data analytics and modeling create new possibilities for producing biomedical knowledge and clinical translation. To understand these advances and inform future directions research, The Lifestyle Medicine Research Summit was convened at the University of Pittsburgh on December 4-5, 2019. The Summit's goal was to review current status and define research priorities in the six core areas of lifestyle medicine: plant-predominant nutrition, physical activity, sleep, stress, addictive behaviors, and positive psychology/social connection. Forty invited subject matter experts (1) reviewed existing knowledge and gaps relating lifestyle behaviors to common chronic diseases, such as cardiovascular disease, diabetes, many cancers, inflammatory- and immune-related disorders and other conditions; and (2) discussed the potential for applying cutting-edge molecular, cellular, epigenetic and emerging science knowledge and computational methodologies, research designs, and study cohorts to accelerate clinical applications across all six domains of lifestyle medicine. Notably, federal health agencies, such as the Department of Defense and Veterans Administration have begun to adopt "whole-person health and performance" models that address these lifestyle and environmental root causes of chronic disease and associated morbidity, mortality, and cost. Recommendations strongly support leveraging emerging research methodologies, systems biology, and computational modeling in order to accelerate effective clinical and population solutions to improve health and reduce societal costs. New and alternative hierarchies of evidence are also be needed in order to assess the quality of evidence and develop evidence-based guidelines on lifestyle medicine. Children and underserved populations were identified as prioritized groups to study. The COVID-19 pandemic, which disproportionately impacts people with chronic diseases that are amenable to effective lifestyle medicine interventions, makes the Summit's findings and recommendations for future research particularly timely and relevant.

Sleep problems in advanced cancer patients and their caregivers: Who is disturbing whom?

Background The aims of the study were to understand sleep problems and their effects in advanced cancer patients and spousal and intimate partner caregivers and to examine the directionality of the link between patients' and caregivers' sleep problems. Methods Fifty-four advanced cancer patients and their spousal and intimate partners were administered a battery of questionnaires that included the Pittsburgh Sleep Quality Index and the Center for Epidemiological Studies at the patients' cancer diagnosis and at 2, 4, and 6 months after diagnosis. Results Patients' and caregivers' sleep duration was significantly related. Using cross-lagged panel analyses, caregivers' sleep quality significantly predicted patients' sleep quality and patients' sleep quality subsequently predicted caregivers' sleep quality. Patients' sleep latency significantly was found to significantly predict caregivers' sleep latency. Conclusion Patients diagnosed with cancer and their intimate partners have poor sleep quality and sleep patterns are related.

Responsiveness of Patient-Reported Outcomes to Treatment Among Patients With Type 2 Diabetes Mellitus and OSA.

BACKGROUND: The Patient-Reported Outcomes Measurement Information System (PROMIS) includes two instruments to quantify sleep symptoms (sleep disturbance [SDA] and sleep-related impairment [SRI]) in diverse populations across a wide symptom spectrum. However, the responsiveness of PROMIS measures to treatment of sleep disorders is unknown. We examined the responsiveness of the PROMIS sleep scales to the treatment of OSA. METHODS: We collected SDA, SRI, and Epworth Sleepiness Scale (ESS) before and after initiation of positive airway pressure (PAP) in patients with type 2 diabetes newly diagnosed with OSA. To compare responsiveness, we compared effect sizes and classifications of symptom improvement using both the reliable change method and thresholds of minimum important difference (MID). RESULTS: A total of 103 patients completed assessments pre- and post-PAP. SDA, SRI, and ESS scores all declined significantly with PAP therapy. We observed the largest effect size for SDA (-0.64; 95% CI, -0.86 to -0.42), followed by SRI (-0.43; 95% CI, -0.63 to -0.23), and ESS (-0.28; 95% CI, -0.42 to -0.15). More patients experienced the reliable change category of symptom remission categorized by the PROMIS measures (SDA: 23.3%; SRI: 31.1%) relative to the ESS (5.8%) (P < .001 for both). Using the MID, SDA and SRI also classified more patients as improved (SDA: 54.4%; SRI: 49.5%) relative to the ESS (35.0%) (P < .001 for both pairwise comparisons). CONCLUSIONS: PROMIS sleep measures were more likely than the ESS to detect an improvement with PAP therapy. Incorporating PROMIS measures into research and clinical care may provide a more sensitive assessment of symptomatic response to OSA treatment.

Impact of sleep health on self-perceived health status.

Although sleep habits have long been recognized as a promoter of health, the World Health Organization 2014 report on non-communicable diseases (NCDs) only listed smoking, alcohol intake, diet and physical activity (PA) as key modifiable risk factors that could enhance health and prevent NCDs. Cross-sectional data on 4385 surveys from the 2015 Catalan Health Survey, representative of the 2015 non-institutionalized Catalan population over age 14, were used to assess and compare the independent associations of low PA (International Physical Activity Questionnaire (IPAQ): low activity); poor diet (PREvención con DIeta MEDiterránea questionnaire (PREDIMED): low-adherent); poor sleep health (Satisfaction, Alertness, Timing, Efficiency and Duration scale (SATED): <8); smoking status; and, alcohol intake (high-risk drinker based on standard drink units) with having a poor self-perceived health status. Logistic regression models adjusted by age, gender, education level and number of comorbidities showed that poor sleep health had the strongest independent association with poor self-perceived health status (OR = 1.70; 95%CI: 1.37-2.12), followed by poor diet (OR = 1.37; 95%CI: 1.10-1.72) and low PA (OR = 1.31; 95%CI: 1.01-1.69). This suggests that sleep habits should be included among the important modifiable health risk factors and be considered a key component of a healthy lifestyle.

A Couples-Oriented Intervention for Positive Airway Pressure Therapy Adherence: A Pilot Study of Obstructive Sleep Apnea Patients and Their Partners.

Background: Partner involvement can influence positive airway pressure (PAP) therapy use among patients with obstructive sleep apnea (OSA). This study assessed the feasibility, acceptability, and preliminary efficacy of a couples-oriented education and support (CES) intervention for PAP adherence. Participants: Thirty newly diagnosed OSA patients and their partners were randomly assigned to one of three groups: an education and support intervention directed at both patient and partner (CES), an education and support intervention directed only at the patient (PES), or usual care (UC). Methods: Feasibility and acceptability were assessed through enrollment and posttreatment program evaluations, respectively. Assessments of sleep quality, daytime sleepiness, and daytime function were obtained from both patients and partners at baseline and 3 months after PAP initiation. Objective PAP adherence was assessed at 1 week, 1 month, and 3 months. Results: Recruitment and attrition data suggest adequate feasibility. All patients and partners in the CES group reported that the intervention was helpful. Patients in the CES and PES groups increased their PAP adherence over the first month of treatment, whereas PAP adherence decreased over this period in the UC group. For patients, large to medium effects for sleep quality (d = -1.01), daytime sleepiness (d = -0.51), and daytime function (d = 0.51) were found for the CES group. The PES and UC groups effect sizes were large to small for sleep quality (d = -0.94; d = -0.40), daytime sleepiness (d = -0.42; d = -0.82), and daytime function (d = 0.41; d = 0.57), respectively. For partners, large effects for daytime sleepiness (d = -1.31) and daytime function (d = 1.54) and small to medium effect for sleep quality (d = -0.31) were found for the CES group. Worsening of sleep quality (d = 0.65) and no change in daytime sleepiness or daytime function were found for the PES group. For the UC group, medium to large effects were found for sleep quality (d = -0.77), daytime sleepiness (d = -0.77), and daytime function (d = 0.65). Conclusions: The findings of this pilot study provide support for taking a couples intervention approach to improve PAP adherence.

Prospective Analyses of Cytokine Mediation of Sleep and Survival in the Context of Advanced Cancer.

OBJECTIVE: The aims of this study were to examine the potential association between sleep problems, symptom burden, and survival in patients with advanced cancer. METHODS: A prospective study of 294 patients with gastrointestinal cancer administered questionnaires assessing sleep, depression, anxiety, stress, pain, fatigue, and health-related quality of life. Serum levels of cytokines including interleukin (IL)-1α, IL-1ß, tumor necrosis factor α, IL-10, IL-2, and interferon-γ were measured to assess biological mediation between sleep and survival. Survival was measured as time from diagnosis to death. RESULTS: Fifty-nine percent of patients reported poor sleep quality, 53% reported poor sleep efficiency, 39% reported sleep latency greater than 30 minutes, and 45% reported sleeping less than 6 hours or greater than 10 hours. We found a significant association between sleep duration and symptom burden. Shorter sleep duration was significantly associated with higher levels of fatigue (r = -0.169, p = .01), pain (r = -0.302, p = .01), anxiety (r = -0.182, p = .01), depression (r = -0.172, p = .003), and lower levels of quality of life (r = 0.240, p = .01). After adjustment for demographic, psychological, and disease-specific factors, short sleep duration was associated with reduced survival (hazard ratio [HR] linear = 0.485, 95% confidence interval = 0.275-0.857) and there was also evidence for a quadratic pattern (HR quadrati = 1.064, 95% confidence interval = 1.015-1.115) suggesting a curvilinear relationship between sleep duration and survival. Interleukin 2 was the only cytokine significantly related to survival (HR = 1.01, p = .003) and sleep duration (ß = -30.11, p = .027). When of IL-2 was added to the multivariable model, short and long sleep (ß = -0.557, p = .097; ß = 0.046, p = .114) were no longer significantly related to survival, suggesting mediation by IL-2. CONCLUSION: Sleep duration was associated with symptom burden and poorer survival and IL-2 was found to mediate the association between sleep and survival. Screening and treatment of sleep problems in patients diagnosed with cancer are warranted.

Deployment Length, Inflammatory Markers, and Ambulatory Blood Pressure in Military Couples.

INTRODUCTION: Changes in the frequency, duration, and nature of military deployments over the past 14 years have spurred efforts to understand the effects of deployment on the health of military service members and their spouses. However, few studies have examined the impact of deployments on health outcomes in both veterans and their partners. This study aims to examine the association between deployment length and health, including ambulatory blood pressure (BP) and stress-related markers of inflammation, in military veterans and their spouses. MATERIALS AND METHODS: This study includes 32 male veterans and 29 female civilian partners. Veterans reported about their deployment and military experiences, including deployment length, combat exposure, and post-traumatic stress disorder (PTSD) symptoms. Plasma measures of inflammatory markers, C-reactive protein (CRP) and interleukin 6 (IL-6), were collected from veterans and spouses. Participants also completed 48 hours of BP monitoring for calculation of mean arterial pressure (MAP) during wakefulness and sleep, and sleep/wake MAP ratio, as an indicator BP nondipping. Regression models examined the association between deployment length and each outcome in the combined sample of veterans and their spouses, including tests of interactions between gender and deployment length, controlling for age, gender, waist circumference, current PTSD, and combat exposure. RESULTS: Longer deployment length was associated with higher CRP levels in veterans and their spouses, although this effect became nonsignificant when limiting analyses to individuals with CRP ≤10 mg/L. There was a significant gender by deployment length interaction effect on MAP ratio, such that longer deployments were associated with higher MAP ratios in female spouses. There was no significant effect of combat exposure in these models. CONCLUSION: Longer deployments are associated with health-related markers in military veterans as well as their spouses. These results suggest the importance of monitoring health during and postdeployment, and of finding ways to mitigate the adverse impact of deployment on health in both members of military couples.

Sleep duration is associated with survival in advanced cancer patients.

OBJECTIVE: Sleep problems have been linked to increased risk of mortality in the general population. Limited evidence suggests similar relationships among people diagnosed with cancer. The aims of the present study were to investigate the type and rates of sleep problems in advanced cancer patients and examine whether sleep problems are associated with survival. METHODS: A prospective study of 292 patients with advanced cancers affecting the hepatobiliary and pancreatic systems were administered a battery of questionnaires measuring sociodemographic information, sleep, and depression. Descriptive statistics, ANOVA, Chi-square, Kaplan-Meier survival, and Cox regression analyses were performed to test the aims. RESULTS: The majority of patients were male (64%) and the mean age was 62 years (SD = 11). Fifty-nine percent of patients reported poor sleep quality; 43% reported sleeping ≤6 h and 2% ≥10 h; 40% reported sleep latency of 30 min or greater; average sleep efficiency was 80%. Of the 292 patients, 58% reported clinically levels of depression and depressive symptoms were related to shorter sleep duration (p = 0.02). After adjusting for factors known to contribute to survival, a curvilinear relationship was observed between sleep duration and mortality: short and long sleep duration were associated with increased mortality [linear term: hazard ratio (HR) = 0.485, 95% confidence interval (CI) = 0.275-0.857; quadratic term: HR = 1.064, 95% CI = 1.015-1.115]. CONCLUSIONS: Consistent with findings in the general population, a curvilinear relationship between sleep duration and mortality was observed in advanced cancer patients. The high prevalence of sleep problems and link with mortality warrants routine screening and development of evidence-based treatments for sleep problems in the oncology setting.

Objective and subjective sleep disorders in automated peritoneal dialysis.

BACKGROUND: Automated peritoneal dialysis (APD) is one of the fastest growing dialysis modalities. It is unknown whether sleep and mood are disturbed while performing repeated overnight exchanges. OBJECTIVES: In this report, we aim to describe and compare the prevalence of sleep-disordered breathing (SDB), periodic limb movements (PLMS), poor sleep quality (SQ), and depression among APD patients compared with stages 3b-5 (estimated glomerular filtration rate ≤44 ml/min/1.73 m2) chronic kidney disease (CKD) and hemodialysis (HD) patients. DESIGN: This is a cross-sectional, descriptive study. SETTING: Study participants were recruited from outpatient nephrology clinics, local dialysis centers, and the Thomas E. Starzl Transplant Institute in Western Pennsylvania between April 2004 and July 2009. PATIENTS: There were 186 participants in this study including 22 APD patients, 89 CKD patients, and 75 HD patients. MEASUREMENTS: In-home polysomnography was performed and two questionnaires were completed, the Pittsburgh Sleep Quality Index (PSQI) and the Patient Health Questionnaire-9 (PHQ-9). METHODS: SDB and PLMS were quantified by in-home unattended polysomnography; poor SQ was defined by a score >5 on the PSQI, and the presence of moderate to severe depression was defined by a score >5 on the PHQ-9. RESULTS: The APD patients had a median age of 37.5 years, were predominantly female (72.7 %), and had a median body mass index (BMI) of 23.8 kg/m2. In univariate analyses, APD patients had significantly lower apnea-hypopnea index compared to HD patients by 12.2 points (likelihood ratio test p = 0.008) and revealed the least percent of TST with nocturnal hypoxemia compared to CKD patients by 2.7 points, respectively (likelihood ratio test p = 0.01). The APD group had also significantly greater stages 3 to 4 sleep compared to the CKD patients by 8.6 points (likelihood ratio test p = 0.009). In multivariate analyses and after adjustment for age, gender, race, and BMI, both APD and HD patients had higher average PSQI scores than CKD patients by 2.54 and 2.22 points, respectively (likelihood ratio test p = 0.005). No other comparisons of sleep parameters among groups reached statistical significance. LIMITATIONS: The limitations of this study are the small sample size of the APD population and the demographic and clinical differences among the three study groups. CONCLUSIONS: Despite differences in univariate analyses, after multivariate adjustment, APD patients had similar sleep parameters and sleep architecture and as poor SQ and symptoms of depression as HD patients. Future studies with larger APD cohorts are needed.

MISE EN CONTEXTE: La dialyse péritonéale automatisée (DPA) est l'une des méthodes de traitement pour l'insuffisance rénale terminale connaissant la plus forte croissance. Toutefois, l'influence de la pratique répétée de cette procédure sur la qualité du sommeil et sur l'humeur du patient demeure peu documentée. OBJECTIFS DE L'ÉTUDE: L'étude visait à caractériser et à définir la prévalence de troubles respiratoires du sommeil, de mouvements involontaires des membres, d'effets sur la qualité du sommeil et de symptômes dépressifs chez les patients sous DPA en comparaison avec la prévalence de ces mêmes paramètres chez les patients souffrant d'insuffisance rénale chronique (eDFG à ≤44 ml/min/1.73 m2) et chez les patients sous hémodialyse. CADRE ET TYPE D'ÉTUDE: Il s'agit d'une étude transversale descriptive qui s'est tenue dans l'ouest de la Pennsylvanie (États-Unis) entre avril 2004 et juillet 2009. Le recrutement des participants s'est effectué au sein des patients en consultation externe des cliniques de néphrologie, dans les centres de dialyses régionaux et au Thomas E. Starzl Transplant Institute. PARTICIPANTS: Un total de 186 patients a pris part à cette étude, parmi lesquels on comptait 22 patients sous traitement par DPA, 89 patients en situation d'insuffisance rénale chronique et 75 patients sous hémodialyse. MESURES: La cueillette des données incluait des mesures de polysomnographie effectuées à domicile par les patients ainsi que les réponses fournies à deux questionnaires : le Pittsburgh Sleep Quality Index (PSQI) et le Patient Health Questionnaire-9 (PHQ-9). RÉSULTATS: La cohorte de patients sous DPA se constituait en majorité de femmes (72,3 %) de 37,5 ans d'âge médian et dont l'indice de masse corporelle (IMC) se situait en moyenne à 23,8 kg/m2. Les analyses univariées ont démontré un index apnées-hypopnées inférieur de 12,2 points (p = 0,0008) et un plus faible pourcentage de troubles du sommeil avec hypoxémie nocturne par 2,7 points (p = 0,01) chez les patients sous DPA en comparaison avec les patients sous hémodialyse. Les analyses multivariées, après correction en fonction de l'âge, du sexe, de l'origine ethnique et de l'IMC des participants, ont quant à elles démontré que les patients sous DPA et sous hémodialyse ont obtenu des résultats plus élevés au questionnaire PSQI que les patients souffrant d'insuffisance rénale chronique, avec des moyennes supérieures de 2,54 et 2,22 points respectivement (p = 0,005). La comparaison entre les différents groupes en ce qui concerne les autres paramètres mesurés n'a pas atteint le degré de signification statistique. LIMITES DE L'ÉTUDE: Les données recueillies au cours de cette étude concernent un faible échantillon de patients sous DPA. De plus, en raison des grandes différences notées dans les données cliniques et démographiques des trois groupes de patients qui la constituaient, la cohorte étudiée manquait d'uniformité. CONCLUSIONS: En dépit des différences notées dans les analyses univariées et de l'obligation d'effectuer des corrections dans les analyses multivariées, les résultats montrent que les patients sous DPA ne présentent pas de différences significatives pour les paramètres de sommeil mesurés. Les patients sous DPA maintiennent des paramètres et une architecture de sommeil semblables à ceux mesurés chez les patients sous traitement par hémodialyse : la qualité du sommeil demeure faible, mais ils ne présentent pas plus de symptômes de dépression. Des études de plus grande envergure sont requises afin de mieux définir l'influence de la DPA sur la qualité du sommeil et sur l'humeur des patients qui la pratiquent.

Actigraphy-defined measures of sleep and movement across the menstrual cycle in midlife menstruating women: Study of Women's Health Across the Nation Sleep Study.

OBJECTIVE: This study aims to evaluate patterns in actigraphy-defined sleep measures across the menstrual cycle by testing the hypothesis that sleep would be more disrupted in the premenstrual period (ie, within the 14 d before menses). METHODS: A community-based longitudinal study of wrist actigraphy-derived sleep measures was conducted in 163 (58 African American, 78 white, and 27 Chinese) late-reproductive-age (mean [SD], 51.5 [2.0] y) women from the Study of Women's Health Across the Nation Sleep Study. Daily measures of sleep (sleep efficiency [%] and total sleep time [minutes]) and movement during sleep (mean activity score [counts]) were characterized using wrist actigraphy across a menstrual cycle or 35 days, whichever was shorter. Data were standardized to 28 days to account for unequal cycle lengths and divided into four weekly segments for analyses. RESULTS: Sleep efficiency declined gradually across the menstrual cycle, but the decline became pronounced on the fourth week (the premenstrual period). Compared with the third week, sleep efficiency declined by 5% (P < 0.0001) and mean total sleep time was 25 minutes less (P = 0.0002) on the fourth week. We found no significant differences between the mean for the second week and the mean for the third week. The association of weekly segments with sleep efficiency or total sleep time was modified by sociodemographic and lifestyle factors, including body mass index, race, study site, financial strain, marital status, and smoking. CONCLUSIONS: Among late-reproductive-age women, sleep varies systematically across the menstrual cycle, including a gradual decline in sleep efficiency across all weeks, with a more marked change premenstrually during the last week of the menstrual cycle. These sleep changes may be modified by altering lifestyle factors.

Sleep hygiene behaviors among midlife women with insomnia or sleep-disordered breathing: the SWAN sleep study.

BACKGROUND: Insomnia and sleep-disordered breathing (SDB) are the most common sleep disorders among midlife women. Although promoting sleep hygiene behaviors may be a useful behavioral approach for the management of insomnia or SDB, the frequency with which women engage in these behaviors is unclear. METHODS: Participants were from the Study of Women's Health Across the Nation (SWAN) Sleep Study (N=321; age range=48-58 years). Out of the full sample, 10.3% (n=33) met Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition diagnostic criteria for insomnia, 15.3% (n=49) had clinically significant SDB (apnea-hypopnea index ≥15), and 4.7% (n=15) met criteria for both insomnia and SDB, resulting in an overall prevalence of 15.0% (n=48) for insomnia and 19.9% (n=64) for SDB. Participants provided diary-based assessments of sleep hygiene behaviors for 14-35 days. Two positive behaviors (sufficient exercise, regular morning out-of-bed time) and four negative behaviors (taking long daytime naps, caffeine consumption near bedtime, alcohol consumption near bedtime, smoking) were examined. These behaviors were compared between women with and without insomnia or SDB following adjustment for sociodemographic factors and mental and physical health indices. RESULTS: Women with insomnia engaged in significantly fewer negative sleep hygiene behaviors than women without insomnia (1.61±0.15 vs. 2.09±0.09 behaviors; p<0.01); specifically, women with insomnia were less likely to take long naps (odds ratio [OR]=0.30, 95% confidence interval [CI]: 0.12-0.74) or consume caffeine near bedtime (OR=0.44, 95% CI: 0.20-0.98). In contrast, women with SDB were less likely to be physically active than women without SDB (OR=0.52, 95% CI: 0.27-0.98), but no other differences in sleep hygiene behaviors were observed. CONCLUSIONS: These data suggest that insomnia in midlife women is not associated with poor sleep hygiene. Increasing physical activity may be a valuable recommendation for midlife women with SDB.

Sleep disturbances and nocturnal symptoms: relationships with quality of life in a population-based sample of women with interstitial cystitis/bladder pain syndrome.

STUDY OBJECTIVES: To characterize the nature and impact of sleep disturbances on quality of life (QOL) in women with interstitial cystitis/bladder pain syndrome (IC/BPS). METHODS: Participants were 3,397 women from a telephone probability survey who met IC/BPS symptom criteria. Sleep quality, duration, and IC/BPS nocturnal symptoms (i.e., trouble sleeping due to bladder pain, urgency, or needing to use the bathroom), general QOL (mental and physical health and sexual functioning), and IC/BPS QOL impairment were assessed via self-report during telephone interview. RESULTS: Over half of the sample reported poor sleep quality, sleep duration ≤ 6 hours, or trouble sleeping due to IC/BPS symptoms. After covariate adjustment, short sleep duration was significantly associated with greater IC/BPS QOL impairment (ß = -0.04; p < 0.001) and poorer self-reported physical health (ß = 1.86; p < 0.001). Poor sleep quality was significantly associated with greater IC/BPS QOL impairment (ß = 0.06; p < 0.001), poorer self-reported physical health (ß = -2.86; p < 0.001), and greater sexual dysfunction (ß = -0.04; p < 0.05). IC/BPS nocturnal symptoms were significantly associated with greater IC/BPS impairment (ß = 0.14; p < 0.001), poorer physical health (ß = -2.76; p < 0.001) and mental health (ß = 0.52; p < 0.01), and greater sexual dysfunction (ß = -0.10; p < 0.001), after covariate adjustment. After further adjustment for IC/BPS nocturnal symptoms, we found that poor sleep quality and short sleep duration were independent correlates of poor self-reported physical health. CONCLUSIONS: Poor sleep quality and short sleep duration, as well as disorder-specific sleep disturbances, are highly prevalent in women with IC/BPS and are associated with poorer disease-specific and general QOL.

Traditional and nontraditional cardiovascular risk factors in comorbid insomnia and sleep apnea.

OBJECTIVES: Insomnia and sleep apnea frequently co-occur and are independently associated with an increased risk of cardiovascular disease, but little is known about cardiovascular disease risk among individuals with comorbid insomnia and sleep apnea. The current study examined traditional risk factors and a physiologic biomarker of cardiovascular risk in comorbid insomnia and sleep apnea. DESIGN: Community-based participatory research study. PARTICIPANTS: The sample comprised 795 participants without preexisting cardiovascular disease from the Heart Strategies Concentrating On Risk Evaluation (Heart SCORE) study. MEASUREMENTS AND RESULTS: Participants were assessed for symptoms of insomnia and sleep apnea risk, as well as for presence of obesity, smoking, a sedentary lifestyle, hypertension, dyslipidemia, and diabetes. Baseline resting brachial artery diameter was measured by B-mode ultrasonography. A total of 138 participants (17.4%) met criteria for insomnia syndrome alone, 179 (22.5%) were at high risk for sleep apnea alone, 95 (11.9%) reported both insomnia syndrome and high sleep apnea risk, and 383 (48.2%) reported having neither insomnia nor sleep apnea symptoms Both high sleep apnea risk alone and comorbid insomnia and high sleep apnea risk groups had greater frequencies of obesity, sedentary lifestyle, hypertension, and three or more traditional cardiovascular risk factors and significantly larger brachial artery diameters than the insomnia alone group and those without insomnia or sleep apnea symptoms. No differences in traditional cardiovascular risk factors or brachial artery diameter were found between the high sleep apnea risk and comorbid groups. CONCLUSIONS: These findings suggest that sleep apnea is a major contributor to cardiovascular risk and co-occurring insomnia does not appear to add to this risk.

A 24-hour approach to the study of health behaviors: temporal relationships between waking health behaviors and sleep.

BACKGROUND: Although sleep is often associated with waking health behaviors (WHB) such as alcohol consumption, caffeine use, smoking, and exercise, the causal direction of these relationships is unclear. PURPOSE: The present study used time series data to examine the temporal dynamics of WHB and sleep characteristics in participants of the Study of Women's Health Across the Nation Sleep Study. METHODS: Three hundred three women completed daily assessments of WHB and wore wrist actigraphs to measure sleep characteristics for the duration of the study (mean = 29.42 days, SD = 6.71). RESULTS: Vector autoregressive modeling revealed that weekly patterns of sleep and WHB best predicted subsequent sleep and WHB suggesting that the associations between WHB and sleep persist beyond their immediate influence. Some WHB predicted some subsequent sleep characteristics, but sleep did not predict subsequent WHB. CONCLUSIONS: These novel findings provide insight into the temporal dynamics of 24-h behaviors and encourage consideration of both sleep and WHB in health promotion and behavior change efforts.

Sleep and risk for high blood pressure and hypertension in midlife women: the SWAN (Study of Women's Health Across the Nation) Sleep Study.

OBJECTIVE: Inadequate self-reported sleep is related to high blood pressure (BP). Our study investigated cross-sectional and longitudinal relationships between poor sleep measured by in-home polysomnography (PSG) and BP. METHODS: Midlife participants (132 black, 164 white, and 59 Chinese) were from the SWAN (Study of Women's Health Across the Nation) ancillary sleep study. In-home PSG measured sleep apnea, duration, efficiency, and electroencephalogram (EEG) total delta and beta power during nonrapid eye movement (NREM) sleep. Women subsequently were followed annually for 4.5 (1-7)years for BP and hypertensive status (>140/90 mmHg or use of antihypertensive medication). Covariates were age, race, site, and educational attainment, with time-covariates of BP medications, body mass index, diabetes mellitus (DM), cigarette smoking, and menopausal status. RESULTS: Sleep duration and efficiency were unrelated to BP cross-sectionally or longitudinally in multivariate models. Women with higher total beta power were more likely to be hypertensive at the time of the sleep study; women with lower total delta power were more likely to show increases in diastolic BP (DBP) and to be at risk for incident hypertension across follow-up. CONCLUSIONS: Low NREM delta power may be a risk factor for future hypertension. Quantitative EEG measures are worthy of future investigations of hypertension risk.