JULY/AUGUST 2022 VOL 33 NO 4 • Conscientiousness with cardiac troponin T and stress coping responses • General awareness of heart fai lure and perception of the problem • Bleeding compl ications with new oral anticoagulants in Kenya • Cl inical profi le and outcomes of infective endocarditis in KwaZulu-Natal • The Healthy Aging Adult South Africa report card • Surgical treatment of left atrial dissection caused by PCI CardioVascular Journal of Afr ica (off icial journal for PASCAR) www.cvja.co.za
DOXAZOSIN 4 mg CALM THE FLOW IMPROVE QUALITY OF LIFE 1 For further product information contact PHARMA DYNAMICS Email info@pharmadynamics.co.za CUSTOMER CARE LINE +27 21 707 7000 www.pharmadynamics.co.za CARZIN XL. Each tablet contains 4 mg doxazosin. S3 A41/7.1/0557. NAM NS2 10/34/0376. For full prescribing information, refer to the professional information approved by SAHPRA, 12 June 2015. 1) Pompeo ACL, et al. A randomised, double blind study comparing the efficacy and tolerability of controlled release Doxazosin and tamsulosin in the treatment of benign prostatic hyperplasia. Int J. Clin Pract Oct 2006;60(10):1172-1177. CXLD839/05/2022.
ISSN 1995-1892 (print) ISSN 1680-0745 (online) Cardiovascular Journal of Afr ica www.cvja.co.za CONTENTS INDEXED AT SCISEARCH (SCI), PUBMED, PUBMED CENTRAL AND SABINET Vol 33, No 4, JULY/AUGUST 2022 EDITORS Editor-in-Chief (South Africa) PROF PAT COMMERFORD Assistant Editor PROF JAMES KER (JUN) Regional Editor DR A DZUDIE Regional Editor (Kenya) DR F BUKACHI Regional Editor (South Africa) PROF R DELPORT EDITORIAL BOARD PROF PA BRINK Experimental & Laboratory Cardiology PROF R DELPORT Chemical Pathology PROF MR ESSOP Haemodynamics, Heart Failure & Valvular Heart Disease DR OB FAMILONI Clinical Cardiology DR V GRIGOROV Invasive Cardiology & Heart Failure PROF J KER (SEN) Hypertension, Cardiomyopathy, Cardiovascular Physiology DR J LAWRENSON Paediatric Heart Disease PROF A LOCHNER Biochemistry/Laboratory Science DR MT MPE Cardiomyopathy PROF DP NAIDOO Echocardiography PROF B RAYNER Hypertension/Society PROF MM SATHEKGE Nuclear Medicine/Society PROF YK SEEDAT Diabetes & Hypertension PROF H DU T THERON Invasive Cardiology INTERNATIONAL ADVISORY BOARD PROF DAVID CELEMAJER Australia (Clinical Cardiology) PROF KEITH COPELIN FERDINAND USA (General Cardiology) DR SAMUEL KINGUE Cameroon (General Cardiology) DR GEORGE A MENSAH USA (General Cardiology) PROF WILLIAM NELSON USA (Electrocardiology) DR ULRICH VON OPPEL Wales (Cardiovascular Surgery) PROF PETER SCHWARTZ Italy (Dysrhythmias) PROF ERNST VON SCHWARZ USA (Interventional Cardiology) SUBJECT EDITORS Nuclear Medicine and Imaging DR MM SATHEKGE Heart Failure DR G VISAGIE Paediatric DR S BROWN Paediatric Surgery DR DARSHAN REDDY Renal Hypertension DR BRIAN RAYNER Surgical DR F AZIZ Adult Surgery DR J ROSSOUW Epidemiology and Preventionist DR AP KENGNE Pregnancy-associated Heart Disease PROF K SLIWA-HAHNLE EDITORIAL 167 A report card for healthy aging in South Africa DJ Blom CARDIOVASCULAR TOPICS 169 Associations of conscientiousness with cardiac troponin T and stress coping responses in a teacher cohort: the SABPA prospective cohort study CE Myburgh • L Malan • R von Känel • HS Steyn • NT Malan 180 Awareness of heart failure and perception of the problem in the general population M Kałużna-Oleksy • M Wawrzyniak • M Klimkowska • M Dudek • J Migaj • E Straburzyńska-Migaj 186 Bleeding complications in patients on new oral anticoagulants for venous thromboembolism in Kenya A Obayo • M Ngunga • J Shah • A Sokwala • A Barasa 194 A 10-year retrospective analysis of the clinical profile and outcomes of infective endocarditis at a tertiary hospital in KwaZulu-Natal, South Africa NS Naidoo • S Ponnusamy • DP Naidoo
CONTENTS Vol 33, No 4, JULY/AUGUST 2022 FINANCIAL & PRODUCTION CO-ORDINATOR ELSABÉ BURMEISTER Tel: 021 976 8129 Fax: 086 664 4202 Cell: 082 775 6808 e-mail: elsabe@clinicscardive.com PRODUCTION EDITOR SHAUNA GERMISHUIZEN Tel: 021 785 7178 Cell: 083 460 8535 e-mail: shauna@clinicscardive.com CONTENT MANAGER MICHAEL MEADON (Design Connection) Tel: 021 976 8129 Fax: 0866 557 149 e-mail: michael@clinicscardive.com The Cardiovascular Journal of Africa, incorporating the Cardiovascular Journal of South Africa, is published 10 times a year, the publication date being the third week of the designated month. COPYRIGHT: Clinics Cardive Publishing (Pty) Ltd. LAYOUT: Jeanine Fourie – TextWrap PRINTER: Tandym Print/Castle Graphics ONLINE PUBLISHING & CODING SERVICES: Design Connection & Active-XML.com All submissions to CVJA are to be made online via www.cvja.co.za Electronic submission by means of an e-mail attachment may be considered under exceptional circumstances. Postal address: PO Box 1013, Durbanville, RSA, 7551 Tel: 021 976 8129 Fax: 0866 644 202 Int.: +27 21 976 8129 e-mail: info@clinicscardive.com Electronic abstracts available on Pubmed Audited circulation Full text articles available on: www.cvja. co.za or via www.sabinet.co.za; for access codes contact elsabe@clinicscardive.com Subscription: To subscribe to the online PDF version of the journal, e-mail elsabe@clinicscardive.com • R500 per issue (excl VAT) • R2 500 for 1-year subscription (excl VAT) The views and opinions expressed in the articles and reviews published are those of the authors and do not necessarily reflect those of the editors of the Journal or its sponsors. In all clinical instances, medical practitioners are referred to the product insert documentation as approved by the relevant control authorities. REVIEW ARTICLE 200 The Healthy Aging Adult South Africa report card: a systematic review of the evidence between 2013 and 2020 for middle-aged South African men and women LK Micklesfield • A Kolkenbeck-Ruh • G Mukoma • A Prioreschi • R Said-Mohamed • L Ware • M Motlhatlhedi • S Wrottesley • SA Norris CASE REPORT 220 Surgical treatment of left atrial dissection caused by percutaneous coronary intervention S Wang • J Weng • F He • X Qian • Y Liu • H Chen PUBLISHED ONLINE (Available on www.cvja.co.za and in PubMed)
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 AFRICA 167 Editorial A report card for healthy aging in South Africa Dirk J Blom DOI: 10.5830/CVJA-2022-047 Increasing urbanisation and the changes in lifestyle and diet that accompany the transition from a rural, agrarian lifestyle to urban living, with its ready availability of high-energy-density processed foods and low requirements for physical activity is associated with increasing rates of non-communicable diseases (NCD).1 Changes in lifestyle and diet resulting from urbanisation are followed by the appearance of physiological risk factors such as overweight and obesity, hypertension, diabetes and dyslipidaemia. These risk factors are mostly asymptomatic, but given sufficiently long exposure, clinically overt disease ensues in many individuals. The increasing rates of atherosclerotic cardiovascular disease in South Africans illustrate this trajectory well. Atherosclerosis is a complex, multifactorial disorder but simplistically can be conceptualised as accumulation of atherogenic lipoproteins [represented mainly by low-density lipoprotein cholesterol (LDLC)] in the vascular wall. This accumulation can be quantified by a cholesterol-year score.2 Once the cholesterol-year score reaches a critical threshold, clinical complications, such as myocardial infarction or stroke, may occur. The threshold for disease is lowered by factors that negatively affect the endothelium and increase inflammation, such as smoking, hypertension, lack of exercise and diabetes. As diets change, the concentrations of LDL-C and other atherogenic lipoproteins, such as remnants, increase the rate of cholesterol accumulation. Simultaneously, the threshold at which disease can become clinically manifest is lowered by the factors outlined above. A long asymptomatic period is followed by the often sudden appearance of a major adverse cardiovascular event. This phenomenon is dramatically illustrated by changes in admission diagnoses at Baragwanath Hospital in Soweto. Whereas acute myocardial infarction was seldom seen in the 1970s, the hospital now has a busy coronary care unit, admitting patients with myocardial infarctions every day.3 The demographic transition, with increasing numbers of older individuals, is another important driver of the increase in NCDs, as a greater proportion of the population lives long enough for these disorders to manifest themselves.4 Although metabolic and vascular abnormalities can often be detected early in infancy or in childhood, for example, fatty streaks are found in the arteries of many children, clinically overt disease often occurs only in middle age or later.5 Ideally, one would therefore design population interventions that target lifestyle and dietary factors very early, preventing the development of physiological risk factors and preventing the next generation from ‘paying for the sins of their forefathers’ through epigenetic programming.6 Middle-aged and older adults are however at greatest absolute risk of dying or becoming disabled within the foreseeable future. Because national surveys have shown that middle-aged adults (45–65 years) are at highest risk for many of the NCD risk factors and middle-aged adults are mostly economically active and often support large extended families, Professor Lisa Mickelsfield and colleagues from the SAMRC/Wits Developmental Pathways for Health Research Unit at the University of the Witwatersrand focused their attention on this age group in this first South African report card on healthy aging (page 200)7. In this extensive systematic review of the literature, the authors assign scores for the adequacy of prevalence data and assess whether national policies exist, and are implemented, for a variety of lifestyle and physiological risk factors. The lifestyle risk factors considered are unhealthy diets, physical inactivity, harmful use of alcohol and tobacco use. The physiological risk factors considered are overweight and obesity, hypertension, dyslipidaemia and poor glucose control. This is an important and arduous task. If we do not know the magnitude of a problem and whether it affects certain groups preferentially, divided for instance by gender, geography or race more than others, then it is very hard to design and target interventions correctly. Similarly, a good understanding of current policies and their implementation (or lack of) is required when considering new strategies to deal with NCD risk factors. So how does South Africa score on the report card for prevalence data? Not unexpectedly, our performance is patchy, with significant room for improvement in many areas. For most of the lifestyle and physiological risk factors, there are shards of information. In some areas these shards can be put together to form a coherent mosaic that probably does accurately reflect the current situation in South Africa. In other areas, the available information remains fragmentary, and much is left to the imagination when trying to get a complete picture. Because the information often comes from multiple studies, there are at times significant differences in methodology, for example, in how dietary information was collected and analysed, and to stay with the metaphor, we are therefore often trying to assemble a mosaic using both glass and stone shards. The areas for which we have data graded as A (published national and regional prevalence data available for the 45–65year age group) are alcohol consumption, obesity, diabetes and hypertension. This is perhaps not surprising, given the major societal impact of these disorders. The reported rates of hazardous alcohol consumption do vary between regions and studies, but a consistent pattern is that hazardous drinking is much more Division of Lipidology, Department of Medicine and Cape Heart Institute, University of Cape Town, South Africa Dirk J Blom, MD, dirk.blom@uct.ac.za
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 168 AFRICA common among men than women, with some studies reporting problem drinking rates (as assessed by the CAGE questionnaire) of close to 50% in men. As almost all studies rely on self-reported alcohol use and not on biomarkers, it is likely that the rates of alcohol use and misuse are even higher than those reported. Obesity, hypertension and diabetes can be quantified relatively easily, and these risk factors are therefore included in multiple population surveys. South Africa has not escaped the international epidemic of obesity and we have very high and increasing rates of obesity. When it comes to obesity, women are much more affected than men in South Africa, with rates of female obesity being twice or three times those of male obesity. There are also significant differences by ethnicity, with selfidentified Indian and black women having the highest rates of obesity, with the lowest rates found in black African men. The prevalence of diabetes increases with age and body mass index. The reported prevalence varies significantly by assessment method, generally being lowest for self-reported diabetes and highest for glycated haemoglobin (HbA1c)-based screening. There does not seem to be a clear gender bias, with some studies reporting higher prevalence in males while other studies show the reverse. However, there are consistent differences by ethnicity with Indians, blacks, and mixed-ancestry (coloured) South Africans having significantly higher rates of diabetes than white South Africans. The prevalence of hypertension increases markedly with age, with around 50% of 45–54-year-old persons being hypertensive, while the prevalence is around 75% in those 55–65 years old. Less than half of hypertensive patients are aware of their diagnosis, not all patients with a diagnosis of hypertension are receiving treatment, and in those on treatment, four out of five have uncontrolled hypertension. Dyslipidaemia is diagnosed and treated adequately in even fewer patients. Prevalence data for tobacco use and physical activity is graded B (published national and regional prevalence data available, not specific to the 45–65-year age group). Not unexpectedly, the rates of physical inactivity are high, with most studies showing women to be more inactive than men. Smoking rates are also worryingly high with all studies consistently showing higher smoking rates in men. The data for diet and dyslipidaemia is graded C (only regional prevalence data for the 45–65-year age group). Most of the dietary data is based on food recall and is therefore subject to recall inaccuracies as well as over and under reporting. Although there are significant differences in the macronutrient composition and micronutrient content of the diets reported by various studies, a common theme is a high intake of processed foods, simple carbohydrates, sugar and salt. The picture painted is therefore not rosy. For many and complex reasons, including poverty, social deprivation, slick advertising and the convenience of ready-to-go, processed foods, unhealthy lifestyles are widespread in South Africa. These unhealthy behaviours such as lack of physical activity, smoking, poor diet and alcohol misuse contribute to the development of physiological risk factors such as obesity, hypertension, dyslipidaemia and diabetes. The high rates of cardiovascular disease in South Africa are therefore not surprising. What can be done to improve the health of South Africans? Many and varied interventions are required, some at the population level, such as tobacco regulations, and others targeted at individuals, for example, adequate diagnosis and treatment of hypertension. This takes us to the policy report card. In the policy arena, the most successful interventions probably have been tobacco regulations, which have resulted in measurable declines in smoking rates, and food policies such as the tax on sugarsweetened beverages and micronutrient fortification of foods. Tobacco, food and alcohol regulation are promulgated at a national level and, given sufficient governmental will to overcome pushback from industry, implementation is relatively easy. The effectiveness of pushback from industry should, however, not be underestimated and is illustrated by the issue of alcohol. Many policy changes have been proposed, for example, increasing the age at which alcohol can be bought and reducing the legal driving limit to zero, but none of these policies have been implemented yet. Dealing with other lifestyle factors, such as lack of physical activity, is much more difficult because it not only requires providing a safe environment conducive to physical activity, but also requires motivating the individual. Similarly, physiological risk factors such as diabetes, hypertension and dyslipidaemia require significant improvements to our healthcare system. We need to have a healthcare system that effectively engages with the population in screening, correctly identifies risks, and provides the correct advice and treatment reliably and regularly without impacting on the ability of middle-aged adults to continue working due to day-long clinic visits. Here the task will not so much be policy formulation, although there are still some areas, such as diabetes management in which policies can be improved, but effective implementation of policies. This report card should serve as a wake-up call to all interested in the health of South Africans. As a nation we have done reasonably well in a few isolated areas, but we need to improve our overall performance considerably to maintain the health of the nation. References 1. Mayosi BM, Flisher AJ, Lalloo UG, Sitas F, Tollman SM, Bradshaw D. The burden of non-communicable diseases in South Africa. Lancet 2009; 374(9693): 934–947. 2. Gallo A, Charriere S, Vimont A, Chapman MJ, Angoulvant D, Boccara F, et al. SAFEHEART risk-equation and cholesterol-year-score are powerful predictors of cardiovascular events in French patients with familial hypercholesterolemia. Atherosclerosis 2020; 306: 41–49. 3. Stewart S, Carrington M, Pretorius S, Methusi P, Sliwa K. Standing at the crossroads between new and historically prevalent heart disease: effects of migration and socio-economic factors in the Heart of Soweto cohort study. Eur Heart J 2011; 32(4): 492–499. 4. Solanki G, Kelly G, Cornell J, Daviaud E, Geffen L. Population ageing in South Africa: trends, impact, and challenges for the health sector. S Afr Health Rev 2019; 2019(1): 175–182. 5. Restrepo C, Tracy RE. Variations in human aortic fatty streaks among geographic locations. Atherosclerosis 1975; 21(2): 179–193. 6. Lejarraga H. Perinatal origin of adult diseases. Arch Argent Pediatr 2019; 117(3): e232–e242. 7. Micklesfield LK, Kolkenbeck-Ruh A, Mukoma G, Prioreschi A, SaidMohamed R, Ware L, et al. The Healthy Aging Adult South Africa report card: a systematic review of the evidence between 2013 and 2020 for middle-aged South African men and women. S Afr J Cardiovasc Dis 2022; 33(4): 200–219.
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 AFRICA 169 Cardiovascular Topics Associations of conscientiousness with cardiac troponin T and stress coping responses in a teacher cohort: the SABPA prospective cohort study Catharina Elizabeth Myburgh, Leoné Malan, Roland von Känel, Hendrik Stefanus Steyn, Nicolaas Theodor Malan Abstract Background: In a South African bi-ethnic cohort, defensive (DefS)/social support/avoidance coping strategies have been shown to influence cardiac troponin T (cTnT) levels through different stress signalling pathways. Personality traits (extraversion, neuroticism, conscientiousness, openness to experience, agreeableness) partially control stress coping responses and may affect prospective cardiac responses. Hence in this cohort, we aimed to examine relationships between personality traits and coping strategies, and to assess associations between cTnT changes over time, personality traits and coping strategies. Methods: A cohort of African and Caucasian male and female teachers (n = 359) participating in both phases of the Sympathetic activity and Ambulatory Blood Pressure in Africans (SABPA) study, was prospectively followed for three years. Personality traits (Basic Traits Inventory) and coping (Coping Strategy Indicator) scores were determined. Fasting serum samples for cTnT determination were collected. Established hypertension-related cTnT cut-off points of 4.2 pg/ml (Africans) and 5.6 pg/ml (Caucasians) were applied. Results: Higher neuroticism and lower conscientiousness scores were found in the Africans than in the Caucasians (p < 0.05). Both traits correlated with all three coping strategies in Caucasians, but only with DefS and avoidance coping in Africans. Over a period of three years, cTnT levels decreased in both races. Compared to Africans, Caucasians showed a greater recovery from the ethnic-specific cTnT cut-off point over time. In the Africans with high DefS scores, cTnT level changes were inversely associated with conscientiousness (adjusted R2 = 0.14; β = –0.26). In Caucasians scoring high in avoidance coping, conscientiousness (odds ratio 0.84) and neuroticism (odds ratio 0.90) showed a lower likelihood of predicting the cTnT cut-off point. Conclusion: In both races, conscientiousness may contribute to healthier stress coping responses and protect against cardiac ischaemia and risk of hypertension. Keywords: personality traits, coping, cardiac ischaemia, troponin T Submitted 18/1/21, accepted 6/11/21 Published online 9/12/21 Cardiovasc J Afr 2022; 33: 169–178 www.cvja.co.za DOI: 10.5830/CVJA-2021-058 Stress has an impact on many bodily systems through neuroendocrine changes [sympathetic adrenal–medullary (SAM) and the hypothalamic–pituitary–adrenal (HPA) axes] that accompany the stress response.1-6 The heart is especially vulnerable to the effects of the stress response, as the autonomic nervous system directly innervates and controls the heart.7,8 Peripheral vascular changes in response to SAM- and HPA-axis activity may also alter loading conditions of the heart.7,8 Associations between myocardial ischaemia and injury, left ventricular dysfunction, coronary perfusion deficits and mental stress have been reported.1,7-9 Cardiac troponin T (cTnT) is a cardiac-specific protein found in the contractile apparatus of the cardiomyocytes.10-12 Detection of this protein in blood samples has been ascribed to myocardial ischaemia with or without necrosis, left ventricular wall stress and a catecholamine overload following severe stress.10-12 In a South African cohort of Africans and Caucasians, a cTnT cut-off point of 4.2 pg/ml in the Africans and a higher cut-off point of 5.6 pg/ml in the Caucasians were predictive of 24-hour systolic hypertension.13 Moreover, in the African men, cTnT levels as low as 4.2 pg/ml were related to silent cardiac ischaemia and were proposed as a potential biomarker thereof.13 According to the transactional stress theory, the magnitude of the stress response and consequent health outcomes are greatly influenced by cognitive appraisal of a stressor.14,15 Primary Hypertension in Africa Research Team (HART), Faculty of Health Sciences, Centre of Excellence, North-West University, Potchefstroom, South Africa Catharina Elizabeth Myburgh, MHSc, BHSc Hons, BSc Leoné Malan, PhD, RN, leone.malan@nwu.ac.za Roland von Känel, MD, PhD Nicolaas Theodor Malan, DSc Department of Consultation-Liaison-Psychiatry and Psychosomatic Medicine, University Hospital Zurich, University of Zurich, Zurich, Switzerland Roland von Känel, MD, PhD Statistical Consultation Services, North-West University, Potchefstroom, South Africa Hendrik Stefanus Steyn, PhD
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 170 AFRICA appraisal of a stressor as either a threat or a challenge is followed by a secondary appraisal where coping options are evaluated to overcome the stressor.14,15 The three main coping strategies usually employed to deal with stress are active problem solving or defensive coping (DefS), seeking social support coping and avoidance coping.16 Effective management of stress and health promotion is usually associated with DefS and seeking social support coping.16 DefS involves the direct confrontation of the stressor in a highly action-orientated manner,14,16 whereas seeking social support coping usually involves approaching another individual or group for advice and comfort in the midst of stress.16,17 Avoidance coping on the other hand is more passive, directed at ignoring the stressful situation, facilitating negative health outcomes.14,16 Avoidance coping, but not DefS or seeking social support coping, has been associated with greater incidence of andmortality from various cardiovascular diseases, including ischaemic heart disease.18 On the other hand, chronic DefS utilisation and sympatho-adrenal dysregulation responses in Africans predicted the hypertension-related cTnT cut-off point, thus indicating possible ineffectiveness of this coping strategy.4 Similar findings were not evident in their Caucasian counterparts who applied effective DefS.4,6,13 DefS is said to be ineffective when an individual loses control over the stressor, resulting in exaggerated stress responses, with subsequent negative health outcomes.19 Coping or stress appraisal is partly determined by personality traits.19-23 The most widely accepted model used in personality trait research is the big five model of personality, which includes extraversion, neuroticism, conscientiousness, openness to experience and agreeableness.21,24,25 Viewing a stressor as a challenge and engaging in DefS have been ascribed to conscientiousness, extraversion and openness to experience,23 whereas threat appraisal of a stressor and engaging in avoidance coping have been related to higher scores in neuroticism.23 Both extraversion and agreeableness were also positively related with seeking social support coping.19,26 Furthermore, personality traits have been found to be prospectively linked to health outcomes, including cardiac health.27,28 Despite a controversy regarding the specific relationship between personality traits and cardiac health, neuroticism and conscientiousness have proven to be more consistent in outcome prediction.27 Neuroticism, characterised by affect instability, anxiety and depression,29 was found to be a risk factor for cardiac morbidity, especially in relation to depressed heart rate variability30 and ischaemic heart disease.31,32 By contrast, high conscientiousness is usually associated with better cardiovascular health31 because these individuals experience higher levels of control over life stressors by finding taxing situations less demanding.33 To the best of our knowledge, the effect of personality traits on stress coping responses and future cardiac health has not been investigated in a South African population. This investigation may also shed light on previous findings of DefS contributing to cardiac stress and ischaemia in Africans but not in Caucasians. As cTnT levels are indicative of cardiac morbidity risk in Africans and Caucasians from South Africa,4-6,13 studying this biomarker in relation to stress coping and personality traits may provide potential novel brain–heart link mechanisms. Therefore, in a bi-ethnic South African cohort, we aimed to (1) examine the relationships between personality traits and coping strategies, (2) assess associations between baseline personality traits and longitudinal changes in cTnT levels, and (3) examine hypertension-related cTnT cut-off points with regard to coping strategies. Methods The current study is nested in the SABPA prospective cohort study, of which a detailed protocol has been published elsewhere.34 Phase I of the SABPA study was conducted in 2008/2009 and the follow-up (phase II) data collection commenced three years later (2011/2012). The study complied with the guidelines of the Declaration of Helsinki on research in humans (2004). The SABPA study included urban-dwelling African and Caucasian teachers of both genders, between the ages of 20 and 65 years at baseline. The teachers were recruited from the Dr Kenneth Kaunda district in the North-West province of South Africa. Except for cultural diversity, socio-economic status and occupational environment were similar for both races. During phase I, there were 409 participants after excluding individuals who were pregnant, lactating, had tympanum temperatures ≥ 37.5°C and were vaccinated or had donated blood three months prior to the commencement of the study. Furthermore, individuals who abused or were dependent on psychotropic substances were excluded. All of the participants were invited three months prior to phase II and a high follow-up rate of 87.8% was achieved. For the current study, only participants who participated in both phases of the SABPA study (n = 359) were included. Additional exclusion criteria were β-blocker use, history of myocardial infarction or stroke and left ventricular hypertrophy (n = 11) at baseline. Participants with missing data (n = 8) and outliers for three-year percentage change (%∆) in cTnT level (n = 4) were also excluded. The final study sample included 336 participants. From Monday to Thursday between 07:00 and 08:00, four participants were fitted daily with 24-hour blood pressure (BP) and electrocardiogram (ECG) (Cardiotens CE120®, Meditech, Budapest, Hungary) as well as 24-hour physical activity (Actical® accelerometers, Montréal, Québec) apparatuses at the school. Thereafter participants continued with their normal daily activities. They had to report any irregularities on the issued 24-hour BP diary cards. Participants were transported to the Metabolic Unit Research Facility of North-West University that afternoon, where they were each allocated a private bedroom. They were requested to refrain from smoking, the intake of caffeine or alcohol, as well as exercise within 12 hours prior to data sampling. In addition, they were introduced to the experimental set-up, completed the general health and demographic questionnaires and received HIV/AIDS pre-counselling. The participants were served a standardised evening meal and completed the Coping Strategy Indicator (CSI), as well as the Basic Traits Inventory (BTI) questionnaire, all of which were done under the supervision of registered clinical psychologists. The participants were requested to go to bed at 22:00, fasting overnight. At 06:00 the next day, the 24-hour apparatuses were disconnected after the last BP recording. Anthropometric measurements were taken, and participants remained in a semirecumbent resting position for 30 minutes, with a 12-lead ECG
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 AFRICA 171 and blood sampling starting after the resting phase. Feedback was given to the participants on immediately available clinical measures in the privacy of their rooms, by a registered nurse, and referrals were made if deemed necessary. After breakfast the participants were transported back to their respective schools. The CSI questionnaire was validated for use in ethnic groups and used to determine the coping strategies utilised during stressful situations.16,35 Before the participants started with the written questionnaire, they had to recall a stressful problem or incident they had encountered in the past six months, while bearing in mind their manner of coping therewith. Each participant’s coping strategy was assessed using both deductive and inductive methodologies. The three coping strategies that formed part of the 33-item CSI questionnaire included: active problem solving or DefS, avoidance or loss of control, and seeking social support. The three sub-scales were assessed by dividing the 33 items into three sets of 11 questions. These questions were randomly ordered in the questionnaire. According to the answer, every item was assigned a numerical value, namely: a lot (three points), a little (two points), or not at all (one point). A maximum score out of 33 was calculated for each sub-scale. The above-median coping scores included:16 26–33 for DefS, 23–25 for social support coping and 19–22 for avoidance coping. High scores within each coping strategy sub-scale (≥ 31 for DefS, ≥ 28 for social support, ≥ 23 for avoidance coping) indicated preferred use of that strategy.16 The reliabilities, as measured by Cronbach α-coefficients for the SABPA study, were 0.83 for DefS, 0.84 for seeking social support coping and 0.69 for avoidance coping. The BTI questionnaire was developed by Taylor and de Bruin24 for a South African context and measures the ‘big five’ factors of personality.25 The use of the BTI questionnaire was validated across different culture and language groups in South Africa.36 The five factors consist of certain facets and include the following: extraversion (positive affectivity, gregariousness, excitement seeking, ascendance, liveliness), neuroticism (anxiety, selfconsciousness, depression, affective instability), conscientiousness (self-discipline, order, effort, dutifulness, prudence), openness to experience (actions, aesthetics, ideas, values, imagination) and agreeableness (compliance, straightforwardness, tendermindedness, modesty, pro-social tendencies). The questionnaire includes 193 written statements covering the different facets for each factor. The participants had to indicate the degree to which they agreed with the statement that was measured on a five-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’. These items were alternated with a 13-item social desirability scale, which also forms part of the BTI. Scores of 40–60 for individual personality traits were considered to be average, while low and high scores were reported as any score < 40 and > 60, respectively.24,37 The Cronbach α-coefficients calculated for the SABPA study to indicate reliability of personality traits included: extraversion (0.81); neuroticism (0.88); conscientiousness (0.90); openness to experience (0.85) and agreeableness (0.83). Level II anthropometrists used standardised procedures when they obtained three anthropometric measurements, using the mean for analysis. Waist circumference was taken at the midpoint between the lower costal rib and the iliac crest, vertical to the long axis of the trunk. The Actical® omnidirectional accelerometer (Montreal, Quebec, Canada) measured total energy expenditure of the participants as an indication of physical activity. A registered nurse obtained fasting blood samples from the participants’ antebrachial vein branches. Standardised procedures were followed regarding preparation and storage of samples at –80°C until analysis. Serum cotinine and gammaglutamyl transferase (GGT) were respectively used as biochemical markers for smoking status and alcohol consumption.38,39 Cotinine was measured by means of a modular Roche automised (Switzerland) apparatus using a homogeneous immunoassay. Cotinine, ametabolite of nicotine, defined habitual smokers when levels were ≥ 14.99 ng/ml.38 Serum GGT, total cholesterol and high-density lipoprotein (HDL) cholesterol were determined with the Unicel DXC 800 (Beckman and Coulter, Germany) and the KonelabTM 20I sequential multiple analyser computer (Thermo Scientific, Vantaa, Finland) at baseline, and with the Integra 400 (Roche, Switzerland) apparatus during follow up. The total cholesterol to HDL cholesterol ratio was calculated at both baseline and follow up. Serum cTnT, N-terminal pro-brain natriuretic peptide (NT-proBNP) and oestradiol were analysed using the electrochemiluminescence immunoassay on the Cobas e411® (Roche, Basel, Switzerland). In the current cohort, 83 (24.7%) cTnT values were below the detection limit of 3 pg/ml and these values were substituted with log-transformed values, which were calculated using the Croghan and Egeghy method for underthe-limit-of-detection values.40 The turbidimetric inhibition immunoassay method was used to measure whole blood EDTA glycated haemoglobin A1c (HbA1c) (Cobas Integra 400 plus; Roche, Basel, Switzerland). The Cardiotens® CE120 (Meditech, Budapest, Hungary), recording 24-hour BP and ECG, was fitted to the non-dominant arm of each participant. This device was programmed to measure BP every 30 minutes between 08:00 and 22:00 and every 60 minutes between 22:00 and 06:00.41 The 24-hour data were analysed using the CardioVisions 1.19 personal edition software (Meditech, Budapest, Hungary). A mean 24-hour systolic blood pressure (SBP) ≥ 130 mmHg and/or diastolic blood pressure (DBP) ≥ 80 mmHg was regarded as hypertensive.42 Statistical analyses All data analyses were done with Statistica version 13.3 (TIBCO Software Inc, Palo Alto, USA, 2018). Visual inspection of Q-Q plots, followed by Kolmogorov–Smirnov and Lilliefors tests, were performed to test normality of data, and non-normally distributed data were Box–Cox transformed. Before continuing with any analyses, interactions were computed to determine how groups should be divided. Interactions on main effects [race × gender × high DefS/seeking social support coping/ avoidance coping] for each of the personality traits and threeyear percentage change (%Δ) in cTnT level, independent of a priori selected covariates were tested with a three-way analysis of covariance (ANCOVA). The a priori covariates included age, and cotinine and GGT levels at baseline.42-45 Independent t-tests were performed to compare general and clinical characteristics between the two races. Pearson’s chi-squared (χ2) test was used for determining prevalence and proportions. Mean differences in cTnT levels between baseline
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 172 AFRICA and three-year follow up were determined by means of dependent sample t-tests within African and Caucasian groups. Effect sizes for independent and dependent sample t-tests were determined by calculating Cohen’s d (0.2 = small effect, 0.5 = medium effect and 0.8 = large effect), while phi (φ)-values for Pearson’s chi-squared tests indicated effect sizes (0.1 = small effect, 0.3 = medium effect and 0.5 = large effect).46,47 McNemar case–control tests determined three-year incidence (negative at baseline becomes positive at follow up) and recovery (positive at baseline recovers to negative at follow up) frequencies for ethnic-specific cTnT cut-off points in Africans and Caucasians.4,13 The following formula was used to calculate %∆ over a period of three years: %∆ = (follow-up value – baseline value) __________________________ baseline value × 100. Pearson’s correlations determined relationships between coping scores and personality traits in Africans and Caucasians. In an attempt to gain more insight into the role of personality traits and genetic influences, paired samples correlations were also used to determine linear relationships between the BTI traits and available gene data [maternal lineage/mitochondrial DNA,48 catecholamine-synthesis trigger enzyme/tyrosine hydroxylase C-824T single nucleotide polymorphism (SNP)49 and accelerated aging/telomere length50]. Thereafter, multiple linear regression analyses were performed to determine associations between three-year %Δ in cTnT level and personality traits, independent of a priori selected covariates. These analyses were performed in Africans and Caucasians with and without considering specific coping styles. Personality Table 1. Baseline characteristics of a bi-ethnic South African cohort Variables Africans (n = 155) Caucasians (n = 181) p-value Cohen’s d φ-value Demographic factors Age (years) 44.25 ± 7.40 46.46 ± 9.95 0.023 –0.25 Men, n (%) 79 (50.97) 88 (48.62) 0.931 0.02 Women, n (%) 76 (49.03) 93 (51.38) 0.668 0.02 Lifestyle and biochemical measurements Cotinine (ng/ml) 26.11 ± 60.52 24.72 ± 82.03 0.861 0.02 GGT (U/l) 42.94 (27.67–74.54) 18.00 (12.00–30.00) < 0.001 0.63 Waist circumference (cm) 92.90 ± 14.86 93.43 ± 16.18 0.754 –0.03 Physical activity (kcal/day) 2584.94 (2156.28–3118.10) 2966.95 (2384.84–3528.59) 0.002 –0.35 Total cholesterol:HDL cholesterol 4.01 (3.22–5.03) 4.79 (3.67–6.25) 0.002 –0.35 HbA1c (measured as %) 5.70 (5.50–6.20) 5.50 (5.20–5.70) < 0.001 0.63 Oestradiol (pmol/l) 101.30 (65.75–164.300) 79.13 (45.68–135.30) 0.286 0.12 Coping scores Defensive coping 28.28 ± 4.13 28.93 ± 3.87 0.134 –0.16 Defensive coping ≥ 31, n (%) 61 (39.35) 78 (43.09) 0.488 0.03 Seeking social support coping 25.82 ± 4.88 18.79 ± 4.91 < 0.001 1.44 Seeking social support coping ≥ 28, n (%) 61 (39.35) 8 (4.42) < 0.001 0.43 Avoidance coping 21.34 ± 3.74 23.70 ± 5.12 < 0.001 –0.52 Avoidance coping ≥ 23, n (%) 55 (35.48) 105 (58.01) < 0.001 0.23 Personality trait scores Extraversion 43.26 ± 7.62 44.64 ± 6.64 0.077 –0.19 Neuroticism 32.54 ± 9.30 29.30 ± 8.70 0.001 0.36 Conscientiousness 44.90 ± 7.87 47.87 ± 6.70 < 0.001 –0.40 Openness to experience 44.96 ± 7.32 44.92 ± 6.63 0.966 0.01 Agreeableness 52.83 ± 6.45 52.29 ± 5.59 0.833 0.09 Cardiac risk markers cTnT (pg/ml) 4.14 (2.99–5.42) 4.87 (3.19–7.03) 0.001 –0.37 cTnT ≥ 4.2 pg/ml, n (%) 73 (47.71) 111 (61.67) 0.011 0.14 cTnT ≥ 5.6 pg/ml, n (%) 33 (21.57) 70 (38.89) 0.001 0.19 NT-proBNP (pg/ml) 28.88 (16.61–51.61) 34.86 (20.83–55.28) 0.591 –0.06 Haemodynamic measurements 24-h SBP (mmHg) 131 ± 14.53 125 ± 12.52 < 0.001 0.50 24-h DBP (mmHg) 83 ± 10.03 77 ± 8.16 < 0.001 0.65 24-h hypertension, n (%) 100 (64.52) 78 (43.09) < 0.001 0.21 Medication Anti-hypertensive drugs, n (%) 51 (32.90) 23 (12.71) < 0.001 0.24 Thiazides, n (%) 19 (12.26) 9 (4.97) 0.016 0.13 ACE inhibitor, n (%) 18 (11.61) 4 (2.21) 0.001 0.19 Diabetes medication, n (%) 9 (5.81) 2 (1.10) 0.016 0.13 Results of t-tests of independent groups are displayed as arithmetic mean ± standard deviation or as median (interquartile range) for variables that were not normally distributed. Chi-squared (χ2) tests were used to determine proportions and prevalence and are indicated as frequencies (%). GGT, gamma glutamyl transferase; total cholesterol:HDL cholesterol, total cholesterol to high density lipoprotein cholesterol ratio; HbA1c, haemoglobin A1c; cTnT, cardiac troponin T; NT-proBNP, N-terminal pro-brain natriuretic peptide; SBP, systolic blood pressure; DBP, diastolic blood pressure; ACE, angiotensin converting enzyme. Effect sizes are indicated by Cohen’s d or φ-values. Cohen’s d for independent t-tests: 0.2 = small effect, 0.5 = medium effect and 0.8 = large effect. φ-values for chi-squared tests: 0.1 = small effect, 0.3 = medium effect and 0.5 = large effect.
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 AFRICA 173 traits were never added simultaneously into the models to avoid co-linearity. Logistic multiple regressions were used to indicate the probability or odds of personality traits (exposure) to increase or decrease the likelihood of meeting ethnic-specific cTnT cut-off points (outcome) with and without considering specific coping styles.4,13 An odds ratio (OR) > 1 is indicative of a higher likelihood of an outcome upon exposure, whereas an OR < 1 is indicative of a lower likelihood of an outcome upon exposure.51 An OR = 1 indicates that the outcome is not affected by the exposure.51 A statistical level of p < 0.05 indicated significance in all analyses. To ensure that the outcome of our regression analyses was not influenced by cases with atrial fibrillation (n = 4) and other lifestyle factors (waist circumference, physical activity), we repeated the regressions by excluding these cases and adjusting for waist circumference and physical activity. Results All three coping strategies and race showed interaction effects for neuroticism [race × DefS ≥ 31, (F(1,325) = 4.60; p = 0.033)]; [race × seeking social support coping ≥ 28, (F(1,325) = 4.18; p = 0.042]; [race × avoidance coping ≥ 23 (F(1,325) = 21.73; p < 0.001]. An interaction term was fitted for conscientiousness in DefS ≥ 31 groups (F(1,325) = 6.43; p = 0.012). No significant interactions were evident between gender and personality traits. We therefore stratified our groups according to race, but also divided each race into the respective coping strategies when associations were assessed in regression analyses, additionally adjusting for gender. In Table 1, Cohen’s d showed small to large effect sizes when comparing lifestyle and biochemical markers, mean coping and personality scores, as well as cardiac risk markers and medication usage. Only those variables showing statistically significant differences are now reported. Compared to Caucasians, Africans were younger (d = –0.25), consumed more alcohol (d = 0.63), engaged in less physical activity (d = –0.35), had less dyslipidaemia (d = –0.35), albeit higher hyperglycaemic values (d = 0.63). Although the number of Africans and Caucasians scoring high in DefS was similar, more Africans scored high on the seeking social support coping scale (φ = 0.43), whereas a greater number of Caucasians scored high on the avoidance coping scale (φ = 0.23). Africans also scored higher in neuroticism (d = 0.36), but lower in conscientiousness (d = –0.40) than Caucasians. When comparing cardiac and haemodynamic risk markers, Africans demonstrated lower cTnT levels (d = –0.37) and more were hypertensive (φ = 0.21), with higher 24-hour SBP (d = 0.50) and DBP (d = 0.65) than the Caucasians. Almost half of the African participants (48%) and 39% of the Caucasian participants showed cTnT values similar to or greater than the ethnic-specific cut-off points (Africans: 4.2 pg/ml, Caucasians: 5.6 pg/ml) that were previously shown to be predictive of 24-hour hypertension.4,13 Furthermore, φ-values for medication use showed low to medium effect sizes, which included greater usage of anti-hypertensives (φ = 0.24), thiazides (φ = 0.13), angiotensin converting enzyme inhibitors (φ = 0.19) and diabetes medication (φ = 0.13) in the Africans. Unadjusted longitudinal changes are presented in Table 2, where cTnT levels over time decreased in both races. Africans showed no significant three-year frequency variance in the ethnicspecific cTnT cut-off point of 4.2 pg/ml,4,13 while Caucasians showed low incidence (7%), with greater recovery (18%) from the 5.6 pg/ml cut-off point (OR: 2.67, p = 0.003). Table 3 presents correlations between coping scores and personality traits. In both Africans and Caucasians, DefS correlated positively with conscientiousness but inversely with neuroticism. A positive correlation was evident between DefS and openness to experience in Africans. The seeking social support coping score correlated positively with extraversion in the African group, but inversely in the Caucasian group. Furthermore, seeking social support coping in Caucasians was positively correlated with neuroticism, but inversely with conscientiousness. A positive correlation in the African group but an inverse relationship in the Caucasian group was evident between avoidance coping and neuroticism. In both the Africans and Caucasians, linear relationships did not exist (p > 0.05; r = 0) between personality traits and gene data (mitochondrial DNA, tyrosine hydroxylase C-824T SNP and telomere length). Caucasians showed no significant associations between %Δ cTnT and the independent variables in Table 4. In Africans, %Δ cTnT was inversely associated with conscientiousness (adjusted R2 = 0.14; β = –0.26, p = 0.034) in individuals scoring high in DefS. These coefficients show that when conscientiousness Table 3. Pearson correlations between personality traits and coping strategy scores in a bi-ethnic South African cohort Personality traits Defensive coping Seeking social support coping Avoidance coping r p r p r p Africans (n = 155) Extraversion 0.13 0.099 0.17 0.038 –0.03 0.718 Neuroticism –0.24 0.003 0.08 0.312 0.48 < 0.001 Conscientiousness 0.26 0.001 0.12 0.136 –0.10 0.210 Openness to experience 0.20 0.013 0.02 0.766 0.06 0.490 Agreeableness 0.15 0.063 0.15 0.065 0.05 0.504 Caucasians (n = 181) Extraversion 0.04 0.623 –0.18 0.018 0.09 0.213 Neuroticism –0.24 0.001 0.40 < 0.001 –0.15 0.049 Conscientiousness 0.17 0.025 –0.15 0.042 0.01 0.884 Openness to experience 0.04 0.584 0.07 0.350 0.01 0.906 Agreeableness 0.10 0.194 0.00 0.965 0.02 0.754 The values are displayed as the correlation coefficient and p-value to indicate significance. Table 2. Presenting longitudinal changes in cardiac troponin T in a bi-ethnic South African cohort Africans (n = 155) Caucasians (n = 181) Baseline; follow up (difference) p d Baseline; follow up (difference) p d cTnT (pg/ml) 4.50; 4.15 (–0.35) 0.020 –0.19 5.63; 5.08 (–0.56) < 0.001 –0.25 cTnT ≥ 4.2 pg/ml cTnT ≥ 5.6 pg/ml Incidence, n (%) 30 (19) 12 (7) Recovery, n (%) 31 (20) 32 (18) OR (95% CI), p 1.03 (0.63–1.71), 0.898 2.67 (1.37–5.18), 0.003 Data presented as unadjusted mean differences between baseline and follow up (change over time) and were determined using dependent sample t-tests. cTnT, cardiac troponin T; NT-proBNP, N-terminal pro-brain natriuretic peptide. Effect sizes are indicated by Cohen’s d-values for dependent sample t-tests: 0.2 = small effect, 0.5 = medium effect and 0.8 = large effect. Incidence, cTnT ≥ 4.2 pg/ml or cTnT > 5.6 pg/ml (negative at baseline becomes positive at follow up) Recovery, cTnT ≥ 4.2 ng/l or cTnT > 5.6 pg/ml (positive at baseline recovers to negative at follow up). OR value effects: 1.5 = small; 2.5 = medium; 4.25 = large.
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 33, No 4, July/August 2022 174 AFRICA increases by one standard unit, %Δ cTnT will decrease by 0.26 (i.e. 0.26 standard deviations on the %Δ cTnT scale). Logistic multiple regression analyses (Table 5) in the Caucasian group showed that conscientiousness was associated with a lower likelihood of meeting a previously defined cTnT cut-off point of 5.6 pg/ml [OR 0.91 (0.84–0.97), p = 0.007], thereby indicating that conscientiousness lowered the probability of meeting the cut-off point by 9%. In Caucasians with high avoidance coping scores, both neuroticism [OR 0.90 (0.83–0.98), p = 0.021] and conscientiousness [OR 0.84 (0.74–0.96), p = 0.013] also lowered the probability of meeting this cut-off point by 10 and 16%, respectively. No significant associations were seen in the African group. Multiple linear and logistic regressions were not performed in Caucasians seeking social support coping at scores ≥ 28, because the sample size was too small (n = 8). Excluding atrial fibrillation cases and adjusting for waist circumference and physical activity did not change the results of the regression analyses. Discussion In a bi-ethnic South African cohort, we aimed to examine relationships between personality traits and coping strategies, and to assess associations between established hypertension-related cTnT cut-off points, personality traits and coping strategies. We Table 4. Forward multiple regression analyses indicating associations between three-year changes in cardiac troponin T and personality traits in a bi-ethnic cohort with and without considering coping style %∆ cTnT Total Defensive coping score ≥ 31 Seeking social support coping score ≥ 28 Avoidance coping score ≥ 23 Africans (n = 155) (n = 61) (n = 61) (n = 55) Adjusted R2 < 0.10 0.14 < 0.10 < 0.10 β (95% CI) p β (95% CI) p β (95% CI) p β (95% CI) p Extraversion NS NS NS NS Neuroticism NS NS NS NS Conscientiousness NS –0.26 (–0.50; –0.03) 0.034 NS NS Openness to experience –0.16 (–0.32; –0.00) 0.050 –0.24 (–0.48; 0.00) 0.056 –0.22 (–0.47; 0.03) 0.083 NS Agreeableness NS NS NS NS Caucasians (n = 181) (n = 78) (n = 8) (n = 105) Adjusted R2 < 0.10 < 0.10 – < 0.10 β (95% CI) p β (95% CI) p β (95% CI) p β (95% CI) p Extraversion NS NS – NS Neuroticism NS NS – NS Conscientiousness NS NS – NS Openness to experience NS NS – NS Agreeableness NS NS – NS Forward stepwise regression analyses showing standardised regression coefficients β (95% confidence interval). We applied case-wise deletion with F to enter 2.5 and F to remove 0.5. Additional independent variables included gender, age, cotinine and log gamma-glutamyl transferase. %∆, percentage change; cTnT, cardiac troponin T; NS, not significant. Table 5. Probability of personality traits to predict hypertension-related cTnT cut-off points in Africans and Caucasians with and without considering coping style cTnT cut-off point of 4.2 pg/ml Without considering coping style Defensive coping score ≥ 31 Seeking social support score ≥ 28 Avoidance coping score ≥ 23 Africans (n = 155) (n = 61) (n = 61) (n = 55) Nagelkerke R2 0.31 0.35 0.29 0.37 OR (95% CI) p OR (95% CI) p OR (95% CI) p OR (95% CI) p Extraversion 0.97 (0.91–1.04) 0.244 0.99 (0.89–1.10) 0.860 0.94 (0.86–1.04) 0.235 0.93 (0.85–1.03) 0.164 Neuroticism 1.01 (0.97–1.06) 0.515 0.94 (0.85–1.03) 0.175 1.02 (0.95–1.09) 0.649 1.05 (0.97–1.13) 0.272 Conscientiousness 0.99 (0.93–1.06) 0.807 0.90 (0.77–1.05) 0.170 0.97 (0.86–1.10) 0.663 0.99 (0.88–1.12) 0.898 Openness to experience 0.98 (0.91–1.06) 0.618 1.00 (0.86–1.16) 0.972 1.06 (0.92–1.21) 0.457 1.08 (0.91–1.29) 0.398 Agreeableness 1.07 (1.00–1.15) 0.064 1.06 (0.93–1.21) 0.383 1.04 (0.92–1.17) 0.541 0.94 (0.79–1.12) 0.501 cTnT cut-off point of 5.6 pg/ml Caucasians (n = 181) (n = 78) (n = 8) (n = 105) Nagelkerke R2 0.30 0.50 – 0.40 OR (95% CI) p OR (95% CI) p OR (95% CI) p OR (95% CI) p Extraversion 0.97 (0.90–1.04) 0.376 0.92 (0.79; 1.07) 0.275 – 0.93 (0.83–1.04) 0.227 Neuroticism 0.96 (0.90–1.01) 0.082 0.91 (0.81; 1.03) 0.129 – 0.90 (0.83–0.98) 0.021 Conscientiousness 0.91 (0.84–0.97) 0.007 0.93 (0.82; 1.07) 0.307 – 0.84 (0.74–0.96) 0.013 Openness to experience 1.05 (0.97–1.14) 0.225 1.13 (0.95–1.35) 0.157 – 1.00 (0.87–1.15) 0.963 Agreeableness 1.02 (0.93–1.11) 0.692 1.02 (0.86–1.22) 0.800 – 1.10 (0.96–1.26) 0.186 Data adjusted for gender and a priori covariates (age, cotinine and gamma-glutamyl transferase). Odds ratio (OR) > 1: higher likelihood of an outcome upon exposure; OR < 1: lower likelihood of an outcome upon exposure; OR = 1: outcome is not affected by the exposure.51 cTnT, cardiac troponin T.
RkJQdWJsaXNoZXIy NDIzNzc=