Miriam (Norway)

Introduction: Self-reported experiences of discrimination have been associated with a range of important clinical disease outcomes; however, most studies have focused on discrimination assessed using a single scale which may fail to capture the pervasive nature of experiences of discrimination in the lives of certain groups. We sought to examine associations between reports of “pervasive” discrimination—conceptualized by combining multiple scales assessing discriminatory experiences-- and an indicator of homeostatic dysfunction, allostatic load (AL), among African-Americans and Whites. We also examined whether associations were stronger among African-Americans than Whites. AL was chosen as it is a measure of cumulative “wear and tear” on the body caused by stressors (i.e., discrimination) and has been associated with cardiovascular-related outcomes. Hypothesis: We hypothesized that pervasive discrimination would be associated with higher allostatic load, and that associations would be stronger among African-Americans compared to Whites. Methods: We conducted linear regression models using data from 215 African-American and 985 White adults (57% female; mean age=54.7 years) in the Midlife in the United States II (MIDUS II) Biomarker Project. A pervasive discrimination score was created by combining three discrimination scales (Everyday Discrimination, Lifetime Discrimination, and Workplace Discrimination), resulting in a score measuring 0-3 domains of discrimination, and an AL score was created based on a set of 24 biomarkers representing 7 biological systems (including inflammation, lipid metabolism, and cardiovascular, among others). Results: African-Americans had higher pervasive discrimination and AL scores than Whites (p<0.001). In models fully-adjusted for demographics and other relevant covariates, raceхpervasive discrimination interaction was observed (p<0.05). A pervasive discrimination score of 2 vs. 0 was associated with a greater increase in AL among African-Americans (b=0.32; p<0.001) compared to Whites (b=0.18; p<0.001). Conclusions: African-Americans report more pervasive discrimination than Whites. This pervasive discrimination in turn, has a greater impact on multisystemic physiological dysregulation in African-Americans compared to Whites. Findings suggest that measuring discrimination by combining multiple forms of discriminatory experiences may be particularly important for studying the health effects of discrimination among African-Americans.

Background: Due to demographic changes a growing number of elderly patients present with mitral valve (MV) disease. Mitral Valve Repair (MV-repair) is the gold-standard treatment option for patients with MV disease. Nevertheless, in elderly patients there is still controversy if MV-repair improves outcomes of patients compared to mitral valve replacement (MVR). We reviewed results after MV surgery in elderly patients treated at our institution over the past 20 years. Methods: Our in-hospital database was explored for patients who underwent MV surgery between 1994 and 2015. A comparative survival-analysis was performed for MV-repair vs. MVR patients. Cox-regression was applied to adjust for age, valve pathology, ejection fraction (EF) and combined procedures. Results: Of the 1776 patients identified, 356 (20.0%) were aged ≥75 years and received either MV-repair (n=224, 62.9%) or MVR (n=132, 37.1%). Patients with repeat cardiac surgery, endocarditis and concomitant aortic valve replacement were excluded. In 55.8% of MV-repair and 46.2% of MVR procedures were combined (p=0.10). Mean age was 79.3±3.1yrs (MV-repair) vs 78.4±2.9yrs (MVR: p=0.02). Logistic EuroSCORE was 14.8±1.3% (MV-repair) vs. 15.3±16.1% (MVR: p=0.77). Median survival of all patients was 7.4years (95%CI: 6.8-8.1) compared to a survival of 8.5years (95%CI: 8.2-9.4) in the age matched UK population (Ratio 0.87; p=0.01). Thirty-day mortality was 6.3% (MV-repair) vs. 10.6% (MVR, p=0.16). Length of ICU-stay was 1.1±1.6 days vs. 1.0±1.0days (p=0.43). Overall 1-, 2-, 5- and 8-year survival after MV-repair (90.7%, 89.0%, 77.0% and 45.9% vs. 79.5%, 76.5%, 58.9% and 41.4%, p[logrank]=0.02) as well as adjusted survival at 1-, 2-, 5- and 8-years were significantly improved after MV-repair (91.1%, 90.2%, 81.0% and 52.5% vs. 77.7%, 76.1%, 56.2%, 37.0%, p<0.01). After MV-repair, impaired EF was an independent predictor of adverse outcome (p<0.01). Conclusion: Mitral valve surgery is a reliable treatment option with excellent long-term outcomes in elderly patients. Long-term outcomes can be improved in this patient group if mitral valves are repaired.

Background: The high prevalence of hypertension in the US adult population represents a serious public health challenge. High hypertension rates result in substantial cardiovascular morbidity and mortality and in elevated health care expenditures. A better understanding of risk factors for development of hypertension in adults may contribute to hypertension prevention efforts. Currently, it is unclear whether veteran status is associated with hypertension risk in women. The goal of this project was to assess the association between veteran status and risk of developing hypertension in women. Methods: The Integrated Health Interview Series (IHIS) is a harmonized dataset for the U.S. National Health Interview Survey (NHIS). Using the IHIS 1997-2010 dataset, population-based prevalence of hypertension among female veterans was compared to non-veterans. For the purpose of analysis, only Whites and African Americans were included in the sample dataset. Veterans were participants who answered ‘Yes’ to the question, “Have you ever been honorably discharged from active duty in the U.S. Army, Navy, Air Force, Marine Crops, or Coast Guard?” Diagnose of hypertension was self-reported. Regression models were performed to evaluate the association between veteran status and the risk of hypertension. Socio-demographic factors and health behaviors were accounted for in the analysis. Sampling weights that adjust for the complex design of the IHIS were applied to all analyses presented. Statistical analyses were performed using SAS version 9.0. Results: Veteran African American women aged 18-40 years of age had a hypertension prevalence rate of 14.8% while non-veterans had a prevalence rate of 14.2%. In White women the prevalence of hypertension was 8.7% among veterans and 7.9% in non-veteran civilians. Hypertension prevalence in African American women older than 40 was 39.1% among veterans and 56.4% in non-veterans (p<0.0001). In White women of the same age-group the prevalence of hypertension was 37.7% among veterans and 39.6% in non-veteran civilians (p<0.005). After adjusting for age, marital status, poverty level, Body Mass Index (BMI), education, smoking, alcohol drinking, and exercise, African American veterans were 24% less likely to have hypertension than non-veteran civilians with statistical significance (OR 0.76, CI 0.6-0.9). There was a 2% risk reduction for hypertension for White veterans but without statistical significance. Conclusion: The prevalence of hypertension was significantly lower in female veterans than in non-veteran females. African American female veterans were 24% less likely to have hypertension than non-veteran African American females, and this association remained significant after adjusting for socio-demographic and health behavior factors.

We thank Kavsak and MacRae for taking interest in our recent clinical research assessing absolute versus relative concentration changes in cardiac troponin for the diagnosis of acute myocardial infarction (AMI).1 We found that the diagnostic accuracy for AMI of absolute cardiac troponin changes is very high and markedly superior in comparison with corresponding relative changes. The method used for comparison was a simple receiver operating characteristic curve, which is an established test to measure the diagnostic accuracy of a test. Kavsak and MacRae questioned “whether focusing solely on either the absolute change or on a relative change is appropriate or practical in the context of the diagnosis of AMI,” and therefore further suggested combining …

We present the parallelization of a linear programming solver using a primal-dual interior point method on one of the heterogeneous processors, namely the Cell BE processor. Focus is given to Cholesky factorization as it is the most computationally expensive kernel in interior point methods. To make it easier to develop and port to other heterogeneous systems, we propose a two-phase implementation procedure where we first develop a shared-memory multithreaded application that executes only on the main processor, and then offload the compute-intensive tasks to execute on the synergistic processors (Cell accelerator cores). We used parent–child supernode amalgamation to increase sizes of the blocks, but we noticed that the existence of many small blocks cause significant performance degradation. To reduce the overhead of small blocks, we extend the block fan-out algorithm such that small blocks are aggregated into large blocks without adding extra zeros. We also use another type of amalgamation that can merge any two consecutive supernodes and use it to avoid having very small blocks in a composed block. The suggested block aggregation method is able to speedup the whole LP solver of up to 2.5 when compared to using parent–child supernode amalgamation alone.