Fully-adjusted Models Research Articles

PMH2 PREDICTORS OF TREATMENT RESISTANT DEPRESSION AMONG ADULTS WITH CHRONIC NON-CANCER PAIN CONDITIONS: A MACHINE LEARNING APPROACH

Presence of chronic non-cancer pain conditions (CNPC) among adults with major depressive disorder (MDD) often reduces benefits of antidepressant therapy, thereby increasing the possibility of treatment resistance. This study sought to identify leading predictors of treatment resistant depression (TRD) among adults with MDD and CNPC using machine learning approaches. This retrospective dynamic cohort study included adult patients (≥18 years) with CNPC and newly diagnosed MDD on antidepressant therapy from a de-identified 10% sample of Optum® Clinformatics® Data Mart (2007-2017). TRD was identified using the Massachusetts General Hospital clinical staging algorithm for claims data. The study population was divided into a 70% training sample and a 30% test/validation sample. Leading predictors of TRD were identified using random forest (RF) and logistic regression (LR) approaches, including a total of 42 demographic, health- and treatment-related features. A fully adjusted model and model with top 20 features were built using Caret and randomForest packages for R software. Overall, a total of 23,645 patients were included (mean age 55 years, 73% female; 78% had ≥2 CNPC, 91% joint pain/arthritis and 71% back/neck pain), and 11.4% (N=2,684) had TRD. Compared to patients without TRD, those with TRD were younger and more likely to have anxiety and/or substance use disorder (all p<0.001). Application of LR and RF resulted in an AUC of 0.711 and 0.704, respectively, for the fully-adjusted model (11.47% OOB error), and AUC of 0.704 and 0.701 for the model with 20 features. Mental health specialist visits, polypharmacy (≥5 medications), psychotherapy use, anxiety, and younger age were the top five predictors of TRD, with AORs ranging from 1.9-1.3 (95% CI = 1.13-2.18, P≤0.001). Machine learning identified several modifiable factors that predicted TRD in adults with CNPC. These factors may serve as targets of further investigation or clinical intervention to improve treatment outcomes in this population.

PMU38 ASSOCIATION OF PRESCRIPTION NON-STEROIDAL ANTI-INFLAMMATORY DRUGS (NSAIDS) WITH DEPRESSION AMONG ADULTS WITH INFLAMMATORY CHRONIC CONDITIONS IN THE UNITED STATES

The study objective was to assess the association between prescription NSAIDs and depression among adults with inflammatory chronic conditions. Some randomized clinical trials have demonstrated that non-steroidal anti-inflammatory drugs (NSAIDs) can be beneficial in reducing depressive symptoms among adults with depression. As depression is often considered as an inflammatory disease, NSAIDs may also reduce the risk of depression by reducing inflammation. We used a retrospective cross-sectional design. Data were derived from 2015 Medical Expenditure Panel Survey (MEPS), a nationally representative sample of the US non-institutionalized civilian population. Inflammatory conditions included arthritis, asthma, cancer, cardiovascular disease, chronic obstructive pulmonary disease (COPD), and diabetes. The dependent variable was the presence or absence of depression and the key independent variable was prescription NSAIDs use. Logistic regression models were used to examine the adjusted associations between NSAIDs and depression. In these regressions, other independent variables (biological, sociocultural, socio-economic, access to healthcare services, medical conditions, behavioral and environmental factors) that may affect the relationship between NSAIDs and depression were also included. All analyses accounted for the complex survey design of MEPS. The study sample consisted of 11,714 adults, representing approximately 124.5 million adults with inflammatory conditions. Overall, 18.2% reported depression and 21% used prescription NSAIDs. A higher percentage of adults using NSAIDs reported depression compared to those without NSAIDS use (23.5% vs 16%;OR=1.59;95%CI=1.40,1.82). In the fully-adjusted logistic regression model, with controls for biological (sex, age, race/ethnicity), sociocultural, socio-economic, access to healthcare services, medical conditions including pain, behavioral and environmental factors, adults using prescription NSAIDs had higher odds of depression (AOR=1.24;95%CI=[1.07,1.43]) compared to those without NSAIDs. In this first real-world study of all adults (with and without depression) in the US, we observed a positive association between prescription NSAIDs and the presence of depression.

Cardiovascular disease risk factor burden and cognition: Implications of ethnic diversity within the Hispanic Community Health Study/Study of Latinos.

ObjectiveHispanics/Latinos have some of the highest prevalence rates for cardiovascular disease risk factors, but stark differences exist by self-reported background. Cardiovascular disease risk factors negatively impact cognition in Hispanics/Latinos; less is known about these relationships by Hispanic/Latino backgrounds. We investigated cognitive associations with cardiovascular disease risk factor burden in a diverse cohort, the Hispanic Community Health Study/Study of Latinos.MethodsBaseline data from this observational study of cardiovascular disease and its antecedents was collected from 2008–2011. We included 7,121 participants 45–74 years old from Central American, Cuban, Dominican, Mexican, Puerto Rican, or South American backgrounds. Dichotomous indicators for hypertension, diabetes, hypercholesterolemia, obesity, and smoking were evaluated and totaled, with participants grouped by lowest (0–2), middle (3) or highest (4–5) burden. Cognitive testing included the Brief Spanish English Verbal Learning Test, letter fluency, and digit symbol substitution.ResultsIn separate fully-adjusted linear regression models, lower fluency and digit symbol substitution performance were restricted to the highest compared to the lowest burden group; whereas the middle burden group displayed impaired memory performance compared to the lowest burden group (p-values≤0.05). Background interacted with burden for learning and memory performance. That is, the association of burden level (i.e., lowest, middle, or highest) with cognitive performance was modified by background (e.g., Mexicans vs Cuban).ConclusionsHispanics/Latinos with higher levels of cardiovascular disease risk factor burden displayed lower levels of cognitive performance, with learning and memory performance modified by background.

Association between lipoproteins and telomere length in US adults: data from the NHANES 1999\u20132002

BackgroundEvidence regarding the correlation between lipoproteins and telomere length in US adults is limited. We aimed to investigate whether lipoproteins was associated with telomere length using US National Health and Nutrition Examination Survey (NHANES) database.MethodsA total of 6468 selected participants were identified in the NHANES Data Base (1999–2002). The independent and dependent variables were lipoproteins and telomere length, respectively. The covariates included demographic data, dietary data, physical examination data, and comorbidities.ResultsIn fully-adjusted model, we found that 0.1 differences of telomere length were positively associated with HDL-C [0.19 (95% CI 0.07, 0.31)], while the associations between LDL-C [0.19 (95% CI -0.27, 0.65)], TG [− 1.00 (95% CI -2.09, 0.07) and telomere length were not detected. By nonlinearity test, only the relationship between HDL-C and telomere length was nonlinear. The inflection point we got was 1.25. On the left side of the inflection point (telomere length ≤ 1.25), a difference in 0.1 of telomere length was associated with 0.50 difference in HDL-C.ConclusionAfter adjusting for demographic data, dietary data, physical examination data, and comorbidities, telomere length is not associated with LDL-C and TG, but is positively associated with HDL-C when telomere length is less than 1.25.

Hypertension and childhood migration: a nationwide study of 2.7 million adolescents.

Immigration studies can shed light on hypertension development and reveal high-risk populations. To this end, we investigated the association between age at immigration and hypertension occurrence at adolescence among immigrants to Israel. We analyzed cross-sectional data on 2 681 294 adolescents assessed for mandatory military service at approximately 17 years of age between 1967 and 2016. The study population constituted of 410 488 immigrants with origins in Ethiopia, Middle East and North Africa, Former USSR and Western Countries. Age at immigration was categorized into 0-5, 6-11 and 12-19 years. Odds ratios (ORs) for hypertension were calculated according to age at immigration with Israel-born participants as controls. Models were made to account for possible confounders. Additionally, the study population was stratified by country of origin and each immigrant group referenced to Israel-born participants of the same origin. In the fully-adjusted model, immigrants arriving until age 11 years had comparable ORs for hypertension to the Israeli-born reference group, whereas recent immigrants, arriving at age 12-19 years had a marked lower OR of 0.30 (95% CI 0.27-0.33; P < 0.001). The lower hypertension odds among recent immigrants persisted in all models and when the study sample was stratified by sex and origin, with all but those of Western origin showing a graded decrease with increasing age at migration categories. Immigrants arriving earlier in childhood lose their protection against hypertension at adolescence relative to the Israeli-born, likely because of lifestyle acculturation. Prevention programs are needed, beginning upon arrival and placing emphasis on nutritional and physical activity habits.

White matter correlates of scam susceptibility in community-dwelling older adults.

Scam susceptibility places older adults - even those with intact cognition - at great risk. Lower grey matter volumes, particularly within right medial temporal regions, are associated with higher scam susceptibility; however, very little is known about white matter associates. We investigated associations between white matter integrity measured using diffusion tensor imaging (DTI) and scam susceptibility in 302 non-demented older adults (75% female; mean years: age = 81.3+7.5, education = 15.7+2.9). Participants completed comprehensive neuroimaging (including DTI, T1- and T2-weighted imaging), a self-report measure of scam susceptibility, and neuropsychological testing. Tract-Based Spatial Statistics (TBSS) investigated associations of DTI-derived measures of fractional anisotropy (FA), trace of the diffusion tensor, axial and radial diffusivity (separately) with scam susceptibility adjusting for age, sex, education, and white matter hyperintensities (WMH; total volume and voxelwise separately). Statistical significance was determined at p < 0.05, Family Wise Error corrected. TBSS revealed significant negative associations between FA in tracts connecting a number of right hemisphere white matter regions and scam susceptibility, particularly after additional adjustment for global cognitive functioning. The pathways implicated were mainly in right temporal-parietal and temporal-occipital regions. Association of trace, axial, and radial diffusivity with scam susceptibility were not significant in fully-adjusted models. Lower white matter integrity within right hemisphere tracts was associated with higher scam susceptibility independent of relevant confounds including global cognition. Thus, a right hemisphere brain network that includes key structures implicated in multi-sensory processing of immediate and future consequences may serve as a neurobiologic substrate of scam susceptibility in vulnerable older adults.

Death in Transitional Asia: 11-Year All-Cause Mortality in the Thai Cohort Study

OBJECTIVE: Thailand is experiencing a substantial reduction in overall mortality, an ageing society and increasing prevalence of non-communicable diseases. There is an urgent need to understand locally important risk factors for this new disease burden and their distribution. We investigated risk factors for mortality in a large cohort of Thai adults and report on key trends.&#x0D; &#x0D; PARTICIPANTS: A nationwide cohort of 87,151 Thai adults followed up since 2005 with their data records linked to the Thai civil registration system to monitor mortality up to the end of 2016. &#x0D; &#x0D; METHODS: We used logistic regression models to measure associations between a large range of socio-demographic, health behaviour and health status variables and all-cause mortality. &#x0D; &#x0D; RESULTS: 1402 cohort members died between 2005 and 2016. In fully-adjusted models higher income, female sex, and higher education had the strongest protective effects against mortality. Normal body weight also protected (AOR 0.71 [0.52-0.96] with Obese as reference). Heavy smoking (AOR 1.48 [1.29-1.70]), and regular alcohol consumption (AOR 1.37 [1.12-1.68]) were associated with the highest mortality. Experiencing injury in the year proceeding the baseline survey also associated with increased mortality, while urbanising since childhood had a protective effect. &#x0D; &#x0D; CONCLUSION: This study adds to evidence regarding risks for all-cause mortality in Thailand. Results indicate the need for Thailand to maintain successful tobacco control programs and to address the effects of increased alcohol consumption. The protective effect of higher education is particularly important in Thailand given the growing proportion of the population who are finishing high school and moving to higher education.

Abstract MP02: Urinary Level of Trimethylamine-N-Oxide Predicts an Increased Risk of Coronary Heart Disease Among Urban Chinese Adults

Introduction: Trimethylamine-N-oxide (TMAO), a diet-derived, gut microbial-host co-metabolite, has been associated with adverse cardiovascular outcomes in patient populations. However, the evidence is lacking from prospective studies conducted in general populations and in non-Western populations. Hypothesis: TMAO level is associated with risk of coronary heart disease (CHD) in general populations. Methods: We examined associations of urinary TMAO and its precursors (i.e., choline, betaine, and carnitine) with risk of CHD in a case-control study nested within two prospective cohorts of Chinese adults, including 275 incident CHD cases and 275 individually matched controls. Results: Urinary TMAO, but not its precursors, was associated with risk of CHD. Odds ratio (OR, 95% CI) for the highest vs. lowest quartiles of TMAO was 1.91 (1.08-3.35; p-trend=0.008) after adjusting for CHD risk factors including obesity, diet, lifestyles, and metabolic diseases and 1.75 (0.96-3.18; p-trend=0.03) after further adjusting for potential confounders/mediators including central obesity, dyslipidemia, low-grade inflammation, and intakes of seafood and deep-fried meat/fish, which were associated with TMAO level in our study. OR was 1.30 (1.03-1.63) per standard deviation increase in log-TMAO in the fully-adjusted model. History of diabetes modified the TMAO-CHD association. A high TMAO level (≥ vs. &lt;median) showed ORs of 6.21 (1.64-23.6) and 1.56 (1.00-2.43) among diabetic and non-diabetic participants, respectively (p-interaction=0.02). History of diabetes also modified the associations of choline, betaine, and carnitine with risk of CHD with significant positive associations being found among diabetics but null associations among total and non-diabetic participants. Conclusions: Our study suggests that TMAO may play a role in the development of CHD, highlighting the importance of diet-gut microbiota-host interplay in cardiometabolic health.

Abstract P239: Sugar-Sweetened Beverage Consumption and Risk of Type 2 Diabetes in African Americans: Findings From the Jackson Heart Study

Background: There is a paucity of data investigating whether consumption of sugar-sweetened beverages (SSBs) is associated with an increased risk of type 2 diabetes (T2D) among African Americans. OBJECTIVE: To determine, in a prospective analysis, whether SSB consumption is associated with increased risk of T2D among African Americans. RESEARCH DESIGN AND Methods: Using a community-based prospective investigation of Jackson Heart Study adult participants from 2004 to 2010, we estimated risk of T2D by frequency of SSB consumption adjusting for age, sex, marital status, CVD, total energy intake, physical activity, smoking status, alcohol consumption, socioeconomic status, and body mass index (BMI) as a potential mediator. Results: Among 3,223 study participants, the mean age was 53±12.5 years, 63% were women, and 66% had at least a college education. Total energy intake, and intakes of carbohydrates, fat, sugar and caffeine increased with increasing SSB consumption, and 21% of participants consumed ≥1 SSB per day. Mean follow-up was 7.5 years with 24,124 person-years, and there were 561 cases of incident diabetes. After age and sex adjustment, SSB consumption was not significantly associated with T2D risk: hazard ratio (HR): 1.61 [95% confidence interval (CI): 0.58-4.48] for 1-4 times/month, HR: 1.77 [95%CI: 0.66-4.76] for 2-6 times/week, and HR: 1.81 [95%CI: 0.67-4.85] for ≥ 1/day (p for trend=0.291) compared to &lt;1/month. In fully-adjusted models, associations remained positive but non-significant and adjustment for BMI attenuated the associations minimally (≤3%). Conclusions: Although HRs tended to increase with higher intake, the frequency of SSB intake was not significantly associated with risk of T2D in African Americans. Larger studies in this population with consideration of other aspects of the diet are warranted.

Abstract P025: Urinary Level of Trimethylamine-N-Oxide Predicts an Increased Risk of Coronary Heart Disease Among Urban Chinese Adults

Introduction: Trimethylamine-N-oxide (TMAO), a diet-derived, gut microbial-host co-metabolite, has been associated with adverse cardiovascular outcomes in patient populations. However, the evidence is lacking from prospective studies conducted in general populations and in non-Western populations. Hypothesis: TMAO level is associated with risk of coronary heart disease (CHD) in general populations. Methods: We examined associations of urinary TMAO and its precursors (i.e., choline, betaine, and carnitine) with risk of CHD in a case-control study nested within two prospective cohorts of Chinese adults, including 275 incident CHD cases and 275 individually matched controls. Results: Urinary TMAO, but not its precursors, was associated with risk of CHD. Odds ratio (OR, 95% CI) for the highest vs. lowest quartiles of TMAO was 1.91 (1.08-3.35; p-trend=0.008) after adjusting for CHD risk factors including obesity, diet, lifestyles, and metabolic diseases and 1.75 (0.96-3.18; p-trend=0.03) after further adjusting for potential confounders/mediators including central obesity, dyslipidemia, low-grade inflammation, and intakes of seafood and deep-fried meat/fish, which were associated with TMAO level in our study. OR was 1.30 (1.03-1.63) per standard deviation increase in log-TMAO in the fully-adjusted model. History of diabetes modified the TMAO-CHD association. A high TMAO level (≥ vs. &lt;median) showed ORs of 6.21 (1.64-23.6) and 1.56 (1.00-2.43) among diabetic and non-diabetic participants, respectively (p-interaction=0.02). History of diabetes also modified the associations of choline, betaine, and carnitine with risk of CHD with significant positive associations being found among diabetics but null associations among total and non-diabetic participants. Conclusions: Our study suggests that TMAO may play a role in the development of CHD, highlighting the importance of diet-gut microbiota-host interplay in cardiometabolic health.

Abstract MP34: Associations Muscle Quality, Muscle Strength, and Muscle Mass With All-Cause Mortality in Elderly Men and Women

Although muscle mass, muscle strength, and muscle quality are important risk factors for disability and mortality in elderly men and women, the associations of these variables with all-cause mortality remain less explored. PURPOSE: We investigated the associations of muscle quality, muscle strength, and muscle mass with all-cause mortality in elderly US men and women. We also compared the strength of these associations with all-cause mortality. Methods: Cox proportional hazards regression was used to examine the associations of muscle quality, muscle strength, and muscle mass with all-cause mortality. We followed 1,361 men and women, aged 60 to 85 years, who participated in the National Health and Nutrition Examination Survey 1999-2002. All participants completed baseline lifestyle factors, and muscle strength and muscle mass measurements. Body composition was estimated by Dual-energy X-ray absorptiometry. Appendicular skeletal muscle mass (SMS) was computed by combining lean tissues in both arms and legs, and relative muscle mass was computed as SMS divided by height in meters squared (AMS). Leg muscle strength (MST) was estimated using the Kin Com MP dynamometer. Leg muscle quality (MQI) was computed as the peak force (Newton) divided by lean mass in the right leg. We further categorized AMS, MST, and MQI as sex-specific quartile categories. The significance of AMS, MST, and MQI with all-cause mortality was also tested by the Cox hazards models with the likelihood ratio statistics with and without adjustment for multiple risk factors. Results: During an average of 9.8 years of follow-up (9,4687 person-years), there were a total of 372 all-cause deaths (85 CVD, 202 chronic diseases). After adjustment for multiple risk factors, the risks of all-cause mortality across MQI quartile categories were (95% CI) 1.00 (Q1, referent), 0.72 (0.55, 0.95), 0.66 (0.49, 0.90), and 0.50 (0.31, 0.80) (P for trend &lt;0.001). The risks of all-cause mortality across MST categories were (95% CI) 1.00 (Q1, referent), 0.85 (0.59, 1.21), 0.50 (0.33, 0.75), and 0.46 (0.31, 0.67) (P for trend &lt;0.001). The risks of all-cause mortality across AMS categories were (95% CI) 1.00 (Q1, referent), 0.77 (0.56, 1.07), 0.75 (0.52, 1.10), and 0.65 (0.47, 0.90) (P for trend = 0.01). As a single measure, after adjustment for multiple risk factors, the hazards for all-cause mortality across MQI, MST, and AMS were 0.97 (p&lt;0.001), 0.99 (p&lt;0.001), and 0.84 (p = 0.001), respectively. When we included all these variables in the fully-adjusted model, only MQI (p = 0.03) was a strong risk factor for all-cause mortality, but not for MST (p = 0.84) and AMS (p = 0.05). Conclusion: Muscle quality was a strong risk factor for all-cause mortality in elderly men and women when compared with muscle strength or relative muscle mass. The American Heart Association should emphasize the importance of improving muscle quality in this geriatric population.

Abstract P084: Neighborhood Deprivation is Associated With Diabetes Complications: Results From the UCSF Population Health Data Initiative

Introduction: Although neighborhood-level factors influence diabetes and diabetes complications, these determinants are often ignored in clinical practice. The widespread use of electronic medical records (EMR) represents a unique opportunity to incorporate these factors into clinical decision making through geocoding of a patient’s residential address. However, despite increasing interest, few examples exist of how to leverage geospatial data within a clinical care system to inform patient care. The objective of this study is to examine the association between neighborhood deprivation and diabetes complications using geocoded address and clinical data extracted from patients’ EMR. Methods: Using retrospective EMR data from UCSF Health (01/01/2015-06/30/2018), we identified a cohort of patients with diabetes using a combination of labs, diagnosis codes and medications (n=9,876). Patients’ residential addresses were extracted from EMR, geocoded, and assigned to census block/track. Area deprivation index (ADI) was calculated for each census block/track, categorized as high neighborhood-level deprivation (ADI ≥80 th percentile) or not (ADI &lt;80 th percentile), and linked to patient-level data. ADI includes 17 measures of education, employment, housing-quality, and poverty and incorporates information from the Census and American Community Survey. We used logistic regression models to examine the association between ADI and the following diabetes-related outcomes: poor glycemic control (HbA1c ≥9%), severe hypoglycemia (ICD-9/10 codes for hypoglycemia-related ED visit or hospitalization), and retinopathy (ICD-9/10 codes). Models were adjusted for age, sex, race/ethnicity, and insurance coverage. Hypoglycemia and retinopathy models were additionally adjusted for last available HbA1c measure. Results: In fully-adjusted models, high neighborhood-level deprivation was associated with increased risk of poor glycemic control (OR: 1.18, 95% CI: 1.01, 1.40) and severe hypoglycemia (OR: 1.35, 95% CI: 1.04, 1.76). Conversely, high neighborhood-level deprivation was associated with decreased risk of retinopathy (OR: 0.78, 95% CI: 0.63, 0.96). Conclusions: Using EMR data, after controlling for several patient-level factors, neighborhood-level deprivation was associated with increased risk of poor glycemic control and severe hypoglycemia and decreased risk retinopathy. Understanding the patient’s broader social context could help inform risk stratification and tailoring of diabetes management. Future analyses will more fully characterize patients’ clinical and utilization-related characteristics and will include additional diabetes complications such as nephropathy, neuropathy and chronic kidney disease. This study represents an example of how neighborhood-level data obtained through patient’s EMR could be used to inform clinical practice.

Facility ownership and mortality among older adults residing in care homes.

Background and objectivesNursing or care home characteristics may have a long-term impact on the residents’ mortality risks that has not been studied previously. The study’s main objective was to assess the association between facility ownership and long-term, all-cause mortality.Research design and methodsWe conducted a mortality follow-up study on a cohort of 611 nursing-home residents in the city Madrid, Spain, from their 1998–1999 baseline interviews up to September 2013. Residents lived in three types of facilities: public, subsidized and private, which were also sub-classified according to size (number of beds). Residents’ information was collected by interviewing the residents themselves, their caregivers and facility physicians. We used time-to-event multivariable models and inverse probability weighting to estimate standardized mortality risk differences.ResultsAfter a 3728 person-year follow-up (median/maximum of 4.8/15.2 years), 519 participants had died. In fully-adjusted models, the standardized mortality risk difference at 5 years of follow-up between medium-sized private facilities and large-sized public facilities was -18.9% (95% confidence interval [CI]: -33.4 to -4.5%), with a median survival (95% CI) of 3.6 (0.5 to 6.8) additional years. The fully-standardized 5-year mortality difference (95% CIs) between for-profit private facilities and not-for-profit public institutions was -15.1% (-31.1% to 0.9%), and the fully-standardized median survival difference (95% CIs) was 3.0 (-1.7 to 7.7) years.Discussion and implicationsThese results are compatible with an association between factors related with the ownership of facilities and the long-term mortality risk of their residents. One of these factors, the facility size, could partly explain this association.

Associations of insulin resistance with cognition in individuals without diagnosed diabetes: Results from the Hispanic Community Health Study/Study of Latinos

AimsInsulin resistance (IR) adversely impacts memory and executive functioning in non-Hispanic whites without diabetes. Less is known in Hispanics/Latinos, despite the fact that Hispanics/Latinos have higher rates of insulin resistance than non-Hispanic whites. We investigated the association between IR and cognition and its variation by age. MethodsData from 5987 participants 45–74 years old without diabetes from the Hispanic Community Health Study/Study of Latinos. IR was considered continuously using homeostasis model assessment for insulin resistance (HOMA-IR) and also dichotomized based on clinically relevant thresholds for hyperinsulinemia (fasting insulin > 84.73 pmol/L or HOMA-IR > 2.6) and sample-based norms (75th percentile of fasting insulin or HOMA-IR). Cognitive testing included the Brief Spanish English Verbal Learning Test (B-SEVLT), Verbal Fluency, and Digit Symbol Substitution. ResultsThere was 90% overlap in participant categorization comparing clinically relevant and sample-based thresholds. In separate fully-adjusted linear regression models, age modified the association between HOMA-IR and Digit Symbol Substitution (p = 0.02); advancing age combined with higher HOMA-IR levels resulted in higher scores. Age also modified the association between clinically relevant hyperinsulinemia and B-SEVLT recall (p = 0.03); with increasing age came worse performance for individuals with hyperinsulinemia. ConclusionThe relationship of IR with cognition in Hispanics/Latinos without diabetes may reflect an age- and test-dependent state.

Abstract WP458: Blood Pressure at Hospital Arrival Does Not Reliably Distinguish Hypertensive Intracerebral Hemorrhage

Introduction: Hypertension is a known risk factor for intracerebral hemorrhage (ICH), but it is unclear whether blood pressure (BP) elevation at hospital arrival can reliably distinguish ICH caused by hypertensive angiopathy from other potential etiologies. Methods: We performed a retrospective single-center cohort study using data from consecutive patients admitted with ICH from February-June 2018. Presumed ICH etiology, location, and other clinical predictors were prospectively adjudicated by two attending neurologists. We compared patients’ first recorded systolic BP (SBP) and mean arterial pressure (MAP) at hospital arrival stratified by hypertensive vs. non-hypertensive primary ICH etiology, with further adjustment for demographics and initial arrival location (direct arrival vs. transfer) using linear regression models. In a sensitivity analysis, we used ICH location (deep, lobar, or infratentorial) rather than etiology in our models to account for potential subjectivity in adjudication. Results: There were 110 ICH patients in our cohort (mean age 69.1 [SD 17.7], 55% male, median ICH score 1.5 [IQR 1-2]). The most frequent ICH etiologies were hypertension (58%), cerebral amyloid angiopathy (19%), vascular lesions (6%), and malignancy (5%); 43% of hemorrhages occurred in deep subcortical locations, 37% were lobar, and 14% were infratentorial. Mean SBP and MAP for patients with hypertensive ICH was 158.8 (SD 29.7) and 109.0 (SD 18.3), respectively, compared to 157.1 (SD 33.4) and 108.2 (SD 21.2) in patients with ICH caused by other etiologies (p=0.78; p=0.82). Fully-adjusted regression models showed that, relative to hypertensive ICH, those caused by non-hypertensive etiologies had similar arrival BP (mean SBP and MAP difference [95% CI], -0.2 [-12.3-11.8] and -0.8 [-8.4-6.8], respectively). In a sensitivity analysis, we found that lobar and infratentorial ICH had similar arrival BP relative to deep ICH (mean SBP and MAP difference [95% CI] 2.1 [-11.1-15.3] and 0.2 [-8.3-8.6] for lobar ICH; 6.5 [-12.0-25.0] and 3.8 [-8.0-15.6] for infratentorial ICH). Conclusion: BP at hospital arrival should not be used as a primary determinant of likely ICH etiology, as hypertension may be implicated in various subtypes of ICH.

Trends in female genital mutilation/cutting in Senegal: what can we learn from successive household surveys in sub-Saharan African countries?

Trends in female genital mutilation/cutting in Senegal: what can we learn from successive household surveys in sub-Saharan African countries?

Is Cannabis Use Associated with Sedentary Behavior during Leisure Time? A Study in Canada, 2011–2012

Background: Sedentary behavior, defined as lying or sitting, is a global health concern. As researchers continue to identify modern-day risk factors for sedentary behavior, few have explored the role of illicit drug use. Objective: To examine the association between leisure-time sedentary behavior and cannabis use, in adolescents and adults. Methods: We conducted a cross-sectional study relating cannabis use to total leisure-time sedentary behavior (hr/wk using the computer, playing video games, watching television or videos, and reading for leisure) using data from the 2011–2012 Canadian Community Health Survey (CCHS), a population-based survey of Canadians age ≥12 years. Analyses were possible for 48,240 respondents in Saskatchewan, Ontario, and Nunavut who reported on sedentary behavior. We used logistic regression modeling to relate frequency of cannabis use (never, occasional, heavy) to high-risk sedentary behavior (<35 versus ≥35 hr/wk) overall and stratified by sex, age, and rural location. Results: Approximately 80% of respondents were ≥25 years old. In the fully-adjusted model, the odds of ≥35 hr/wk of sedentary behavior were 80% higher for heavy cannabis users versus never users (OR = 1.8, 95% CI: 1.4–2.3); in occasional cannabis users, the odds were 30% higher (OR = 1.3, 95% CI: 1.1–1.5). In stratified analyses, odds ratios were statistically significant among adults age 25–44 years and people living in nonrural settings. Conclusions/Importance: Our findings support dual lifestyle interventions targeting heavy cannabis use and excessive sedentary behavior simultaneously. Moreover, leisure-time sedentary behavior should be considered as a covariate in future epidemiologic models relating cannabis to health.

Salivary characteristics and dental caries experience in remote Indigenous children in Australia: a cross-sectional study

BackgroundWhile associations between salivary characteristics and dental caries have been well studied, we are not aware of this being assessed in a remote Indigenous child population, where lifestyles may be different from urban children. Our aim was to assess associations between caries experience and putative biomarkers in saliva, accounting for oral hygiene and dietary habits.MethodsChildren attending schools in an Indigenous community in remote north Queensland, Australia were invited to an oral examination by qualified and calibrated examiners. Salivary flow rate, pH, buffering capacity and loads of mutans streptococci (MS), lactobacilli (LB) and yeasts were determined. Also, data on tooth brushing frequency and soft drinks consumption were obtained via a questionnaire. Caries experience was recorded by the International Caries Detection and Assessment System (ICDAS-II), and quantified as decayed, missing and filled surfaces. Relationships between the salivary variables and the cumulative caries experience (dmfs+DMFS) in the deciduous and permanent dentitions were examined by multivariate analyses to control the effect of confounders.ResultsThe mean cumulative decayed (DS + ds), missing (MS + ms) and filled (FS + fs) surfaces were 3.64 (SD: 4.97), 1.08 (4.38) and 0.79 (1.84) respectively. Higher salivary MS and LB counts, low tooth brushing frequency and daily soft drink consumption were significantly related to greater caries experience. Caries experience was about twice in those with ≥10^5 CFU/ml saliva counts of MS (mean = 6.33, SD: 8.40 vs 3.11, 5.77) and LB (7.03, 7.49 vs 4.41, 8.00). In the fully-adjusted multivariate model, caries experience in those with higher counts of MS and LB were 51 and 52% more than those with lower counts.ConclusionsAs with studies in other populations, childhood salivary counts of MS and LB were significantly associated with greater caries experience in this remote Indigenous community. To address the serious burden of oral disease, we are researching ways to promote a healthy oral environment by encouraging good dietary habits, and emphasising the importance of daily tooth brushing with a fluoridated toothpaste. Our ongoing longitudinal studies will indicate the success of measures employed to reduce the counts of bacteria closely associated with cariogenesis and their impact on caries increment.Trial registrationAustralian New Zealand Clinical Trials Registry (ANZCTR), No: ACTRN12615000693527; date of registration: 3rd July 2015.

Heart rate reserve during pharmacological stress is a significant negative predictor of impaired coronary flow reserve in women.

Evidence to date has failed to adequately explore determinants of cardiovascular risk in women with coronary microvascular dysfunction (CMVD). Heart rate responses to adenosine mirror autonomic activity and may carry important prognostic information for the diagnosis of CMVD. Hemodynamic changes during adenosine stress were analyzed in a propensity-matched cohort of 404 patients (202 women, mean age 65.9 ± 11.0) who underwent clinically indicated myocardial perfusion 13N-ammonia Positron-Emission-Tomography (PET) at our institution between September 2013 and May 2017. Baseline heart rate (HR) was significantly higher in patients with abnormal coronary flow reserve (CFR, p< 0.001 vs normal CFR). Accordingly, a blunted HR response to adenosine (=reduced heart rate reserve, %HRR) was seen in patients with abnormal CFR, with a most pronounced effect being observed in female patients free of myocardial ischemia (45.9 ± 34.9 vs 26.5 ± 18.0, p< 0.001 in women and 29.1 ± 16.9 vs 24.3 ± 21.7, p= 0.15 in men). Hence, a fully-adjusted multivariate logistic regression model identified HRR as the strongest negative predictor of reduced CFR in women free of myocardial ischemia, but not in men. Accordingly, receiver operating characteristics (ROC) curves for the presence of reduced CFR revealed that a %HRR <35 was a powerful predictor for abnormal CFR with a sensitivity of 81% and a specificity of 60% in women. A blunted HRR <35% is associated with abnormal CFR in women. Taking into account HR responses during stress test in women may help to risk stratify the heterogeneous female population of patients with non-obstructive coronary artery disease (CAD).

Suicidality in People With Obsessive-Compulsive Symptoms or Personality Traits.

Objective: Obsessive-Compulsive Disorder (OCD) and Obsessive-Compulsive Personality Disorder (OCPD) have been reported to be associated with mood instability (MI), depression and suicide-related outcomes. We examined whether obsessive-compulsive symptoms and personality traits as well as obsessional thoughts of death, are associated with suicidal thoughts, non-suicidal self-injury and attempted suicide.Methods: We used data from 7,839 people from the 2000 British Adult Psychiatric Morbidity Survey that elicited symptoms of OCD with a computerized version of the Clinical Interview Schedule—Revised (CIS-R) and traits of OCPD with a self-completed version of the SCID-II. We created a series of logistic regression models, first entering only OCD symptoms and OCPD traits in separate models, to which depression and mood instability (MI) were added. We also examined the relation of obsessional thoughts about death with self-harm in a network analysis model that included the main symptoms of mood instability and depression.Results: OCD symptoms were associated with suicidal thoughts (OR: 1.23, 95% CI: 1.14–1.32), and suicide attempts (OR: 1.13, 95% CI: 1.04–1.24) in the fully-adjusted model. OCPD traits were associated with suicidal thoughts (OR: 1.14, 95% CI: 1.10–1.19), non-suicidal self-injury (OR: 1.14 95% CI: 1.03–1.26), and suicide attempts (OR: 1.09; 1.01–1.17). Depression and MI were both associated with all three suicide-related outcomes. In the network analysis, MI was the most prominent correlate of suicide-related outcomes, being associated with suicidal ideas (partial r = 0.15) and non-suicidal self-injury (partial r = 0.07).Limitation: This was a cross-sectional study that used a single-item measure for mood instability.Conclusions: Obsessive-compulsive symptoms and personality traits are related to suicide-related outcomes independently of depressive symptoms and mood instability. This relationship is not accounted for by obsessional thoughts of death alone.