UvA-DARE (Digital Academic Repository)

The role of sleep in ethnic inequalities in health: Cardiovascular disease and risk factors

Anujuo, K.O.

Publication date 2018 Document Version Final published version License Other Link to publication

Citation for published version (APA): Anujuo, K. O. (2018). The role of sleep in ethnic inequalities in health: Cardiovascular disease and risk factors.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The . You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:07 Oct 2021 THETHE ROLEROLE OFOF SLEEPSLEEP ININ ETHNICETHNIC INVITATIONINVITATION The role of sleep in ethnic inequalities health The role of sleep in ethnic inequalities health INEQUALITIESINEQUALITIES ININ HEALTH:HEALTH: YouYou are are cordially cordially invited invited to to cardiovascularcardiovascular diseasedisease andand riskrisk factorsfactors attendattend the the public public defense defense ofof the the doctoral doctoral thesis thesis of of

KENNETHKENNETH OKWUDILI OKWUDILI ANUJUO ANUJUO

titledtitled

THETHE ROLE ROLE OF OF SLEEP SLEEP IN IN ETHNIC ETHNIC INEQULITIESINEQULITIES IN IN HEALTH: HEALTH: CARDIOVASCULARCARDIOVASCULAR DISEASE DISEASE ANDAND RISK RISK FACTORS FACTORS

OnOn Friday, Friday, September September 21, 21, 2018 2018 AtAt 12:00 12:00 InIn the the Agnietenkapel Agnietenkapel OudezijdsOudezijds Voorburgwal Voorburgwal 229-231 229-231 AmsterdamAmsterdam ReceptionReception follows follows immediately immediately afterafter the the defense defense in in the the reception reception areaarea of of the the Agnietenkapel Agnietenkapel

PARANYMPHS:PARANYMPHS: K K IkennaIkenna Ebuenyi Ebuenyi ENNETH ENNETH [email protected]@vu.nl BenBen Nieuwenhuizen Nieuwenhuizen O O b.p.vannieuwenhuizenb.p.vannieuwenhuizen@@amc.uva.nlamc.uva.nl KWUDILI KWUDILI A A NUJUO NUJUO

KennethKenneth OkwudiliOkwudili AnujuoAnujuo

Anujuo_omslag_FINAL.inddAnujuo_omslag_FINAL.indd All All Pages Pages 30/07/201830/07/2018 21:13 21:13

THE ROLE OF SLEEP IN ETHNIC INEQUALITIES IN HEALTH: CARDIOVASCULAR DISEASE AND RISK FACTORS

Kenneth Okwudili Anujuo

Anujuo_binnenwerk_FINAL.indd 1 30/07/2018 21:19 COLOFON

The role of sleep in ethnic inequalities in health: Cardiovascular disease and risk factors.

PhD thesis Academic Medical Center – University of Amsterdam.

ISBN: 978-94-6375-046-2

© Kenneth O Anujuo

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the author or the copyright-owning journalists for published chapters.

Layout by Lize Jansen

Printing: Ridderprint BV, www.ridderprint.nl

The studies presented in this thesis were conducted at the Department of Public Health, Academic Medical Center, University of Amsterdam, the Netherlands. The research was mainly from the HELIUS study and partly from the ABCD study. The HELIUS study was funded by Academic Medical Center, Amsterdam, the Public Health Service of Amsterdam, the Dutch Heart Foundation (grant number: 2010T084), the Netherlands Organization for Health Research and Development (ZonMw) (grant number: 200500003), and the European Fund for the Integration of non- EU immigrants (EIF) (grant number: 2013EIF013). The ABCD study was funded by the Netherlands Organization for Health Research and Development (ZonMw) (TOP grant: 40-00812-98-11010).

Anujuo_binnenwerk_FINAL.indd 2 30/07/2018 21:19 THE ROLE OF SLEEP IN ETHNIC INEQUALITIES IN HEALTH: CARDIOVASCULAR DISEASE AND RISK FACTORS

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam

op gezag van de Rector Magnificus

prof. dr. ir. K.I.J. Maex

ten overstaan van een door het College voor Promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel

op vrijdag 21 september 2018, te 12:00 uur

door

Kenneth Okwudili Anujuo

geboren te Okohia, Nigeria

Anujuo_binnenwerk_FINAL.indd 3 30/07/2018 21:19 PROMOTIECOMMISSIE:

Promotor(es): Prof.dr. K. Stronks AMC – UvA Copromotor(es): Dr. C.O. Agyemang AMC – UvA Overige leden: Prof.dr. A.K. Groen AMC – UvA Prof.dr. B.J.C. Middelkoop Universiteit Leiden Prof.dr. S.A. Reijneveld Rijksuniversiteit Groningen Dr. E.J. de Bruin Universiteit van Amsterdam Dr. I.G.M. van Valkengoed AMC – UvA

Faculteit der Geneeskunde

Anujuo_binnenwerk_FINAL.indd 4 30/07/2018 21:19 In loving memory of my late parents: Simeon and Louisa Anujuo

Anujuo_binnenwerk_FINAL.indd 5 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 6 30/07/2018 21:19 CONTENTS

Chapter 1 General introduction 9

Chapter 2 Ethnic differences in self-reported sleep duration 27 in the Netherlands – the HELIUS study

Chapter 3 Ethnic differences in sleep duration at 5 years, 53 and its Relationship with overweight and blood pressure

Chapter 4 Relationship between short sleep duration and 73 Cardiovascular risk factors in a multi-ethnic cohort – the HELIUS study

Chapter 5 Relationship between sleep duration and arterial 99 stiffness in a multi-ethnic population: the HELIUS study

Chapter 6 Contribution of short sleep duration to ethnic 123 Differences in cardiovascular disease: results from a Cohort study in the Netherlands

Chapter 7 Association between depressed mood and sleep 147 duration Among various ethnic groups – the HELIUS study

Chapter 8 General discussion 165

Summary 185

Samenvatting 191

Acknowledgements 197

About the author 201

Lists of publications 205

Portfolio 209

Anujuo_binnenwerk_FINAL.indd 7 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 8 30/07/2018 21:19 CHAPTER 1

GENERAL INTRODUCTION

Anujuo_binnenwerk_FINAL.indd 9 30/07/2018 21:19 Chapter 1

Ethnic inequalities in health has been widely acknowledged across the globe. Different studies both in and the of America (USA) indicate variations in inequalities in health between ethnic minority groups and the European populations [1]. In the Netherlands, various studies have been conducted and reasonable extensive information on inequalities in health status among ethnic minorities both at regional and national level have been well investigated [2,3,4]. A lot of studies show remarkable variations in health outcomes such as CVD and risk factors (e.g. diabetes, hypertension) and mental health among ethnic minority groups and host European populations [5,6,7,8]. Explanations for the ethnic inequalities in health are unclear and with the increasing minority populations, it is relevant to understand the underlying factors to help guide prevention and treatment initiatives. This thesis focuses on the role of sleep in ethnic inequalities in CVD and risk factors. Sleep and its impact on health is increasingly becoming important but very little is known about its influence on ethnic inequalities in health.

THE ROLE OF SLEEP IN HEALTH

By definition, sleep is “a natural periodic state of the body and mind characterized by closing of the eyes, partial or total loss of consciousness and reduced response to external stimuli” [9]. Sleep has a restorative function for the body and mind, and thus helps to maintain health.

Poor sleep quality may result in deleterious health outcomes. The National Sleep Foundation (NSF), American Academy of Sleep Medicine (AASM), and Sleep Research Society, in a joint consensus, recommend 7-9 hours and 10-11 hours of sleep at night as healthy for adults and preschool children, respectively [10]. Inadequate sleep has become a common problem in our modern society. On average, many people fall short of reaching the recommended hours of sleep per night. This can be attributed to the busy schedules that is inherent in our modern society, resulting from long working hours, doing multiple jobs, shift-work, late night use of technology at bed time such as, mobile phones, television, computers and computer games, sedentary lifestyle, lack of physical activity, sleeping with pets; stress, and adoption of some cultural practices which tend to negatively influence sleep duration such as co-sleeping, disregard for napping, and believe that hardworking and successful people require short sleep duration [2,11,12,13,14].

10

Anujuo_binnenwerk_FINAL.indd 10 30/07/2018 21:19 General introduction

Sleep disruption has been linked to several adverse health outcomes due to its effect on the body`s homoeostasis which regulate human cellular, tissue, organ and system activities. Although sleep disruption affects other vital organs of the body such as the brain [15], sleep affects metabolic and 1 cardiovascular health, as have been reflected in previous studies [16]. The studies have shown that sleep deprivation has serious or major metabolic and cardiovascular consequences, and therefore a risk factor for poor health in the future [16].

Studies highlighting the impact of sleep on health have been conducted among various populations. For instance, Steptoe et al (2006) [17] has shown that short sleep was associated with poor health among young adults from 24 different countries. Also, Cappuccio et al`s (2008) [18] meta- analysis demonstrates that short sleep is associated with obesity in children and adults. Other studies conducted in Europe and the USA have shown that sleep duration is associated with increased risk of CVD risk factors such as obesity [19,20,21], type 2 diabetes [22,23,24], hypertension [25,26], and dyslipidaemia [27-33]. In addition, reports from previous studies indicate that short sleep is associated with CVD such as stroke, myocardial infarction, coronary heart disease (CHD) [34-41], and CVD mortality [42,43].

The mechanism linking sleep with cardiovascular disease and diabetes

Knutson (2012) [44] has formulated a model which shows how sleep deprivation is linked to obesity, diabetes and CVD (Figure 1); as well as other factors (social, cultural and environmental) which may directly or indirectly affect sleep and the associated health outcomes (figure 2).

In this model, three pathways which link inadequate sleep to obesity are shown in figure 1. In these pathways, weight gain/obesity is the result of imbalance between food intake and physical activity (energy expenditure) as reflected in the first pathway.

11

Anujuo_binnenwerk_FINAL.indd 11 30/07/2018 21:19 Chapter 1

Fig. 1 Adapted from Knutson 2012 [44]: Schematic representation of possible mechanistic pathways linking inadequate sleep to obesity, diabetes, cardiovascular disease (CVD), and hypertension (HTN)

The second pathway reflects that the more people are awake during the day, the more they tend to eat, hence more likely to be obese. The third pathway indicates a hormonal activity (leptin and ghrelin) in the regulation of appetite resulting in an increased food intake.

The pathways which link inadequate sleep to cardiometabolic diseases (example, type 2 diabetes) partly overlaps with the mechanisms leading to obesity (through insulin resistance and CVD), as obesity is a risk factor for CVD. In addition, the impact might also be more direct. The figure shows that inadequate sleep may lead to reduced utilization of glucose by the brain, which in turn, promotes reduction in glucose tolerance, and insulin resistance resulting in type 2 diabetes. Also, changes in the concentration of growth hormone and cortisol affect glucose metabolism, causing insulin resistance. Increase in the levels of growth hormone triggers transient insulin resistance, with accompanying decrease in glucose uptake, increased blood glucose levels, and insulin resistance and diabetes. On the other hand, increased evening cortisol concentration can lead to reduced morning insulin sensitivity which causes impaired glucose tolerance [26]. Inadequate sleep

12

Anujuo_binnenwerk_FINAL.indd 12 30/07/2018 21:19 General introduction

might also affect cardiometabolic disease through increased sympathetic nervous activity [27,28]. Increased sympathetic nervous activity at the level of pancreas may give rise to a decrease in the secretion of insulin. Also unfavourable changes in the metabolism of glucose may lead to insulin 1 resistance predisposing to diabetes mellitus.

The last pathway illustrates how increased sympathetic nervous activity may result in increased blood pressure and development of hypertension and CVD.

Determinants of inadequate sleep

The factors considered as determinants of inadequate sleep have been well studied by Knutson (2012). As highlighted in figure 2, Knutson illustrated various factors and their respective domains which determine inadequate sleep duration.

Fig. 2 Adapted from Knutson 2012 [44]: Factors associated with biology, culture and environment which can impact and interact with sleep to increase vulnerability to obesity and its health consequences.

The mechanism described in figure 1 depends on three domains illustrated in figure 2, including biology and culture and environment.

While sleep duration decreases as one advances in age [45,46], also variations such as in sex, gynaecological status in women (menstruation cycle/menopause), secretion of melatonin (sleep-promoting hormone), and

13

Anujuo_binnenwerk_FINAL.indd 13 30/07/2018 21:19 Chapter 1

cortisol (from hypothalamo-pituitary-adrenocortical (HPA) axis, have been shown to influence sleep duration [47]. Also, culture, affects sleep owing to its influence on the belief, perception and attitudes of people towards sleep, in a manner likely to reduce sleep duration. Environmental factors including light, noise, temperature, poor air quality, and poor neighborhood have the potential to disrupt sleep directly or indirectly.

Sleep and ethnic inequalities in health

Little is known about ethnic inequalities in sleep duration and their impact on ethnic inequalities in health. The limited evidence, however, seems to suggest that some ethnic minority groups in North America have poor sleep quality than their European counterparts. For instance, in one USA based- study, sleep quality was poorer in African Americans than in [48]. In addition, the same study revealed that African Americans exhibit less night time decrease in blood pressure dipping than Whites Americans. The study posited that excessive sympathetic activity, rather than ethnic differences in sleep quality accounted for the observed dip in blood pressure. In Europe, data on the role of sleep on ethnic inequalities in health are lacking. Europe has different ethnic groups compared to US necessitating the need for more studies in Europe. Besides, the environments in which the ethnic groups live as well as the culture may also differ between the two continents and may influence their sleep patterns. This can lead to different sleep patterns among the various ethnic groups which may contribute to differences in cardiovascular outcomes among the various ethnic groups.

For instance, the pathway of sleep deprivation to obesity (as shown in Fig 1), may differ from one ethnic group to another due to cultural practices such as adoption of earlier or later bedtimes, bed sharing/communal sleeping habits, sleeping with pets, and levels of physical activity. Also factors such as values and expectations attributed to sleep, affecting perception and value of habits attributed to sleep, may differ among the various ethnic groups. These include: the values and attitude which people have regarding sleep (positive or negative), sleep-related rituals such as the belief that less intake of food and water or liquids before bedtime may affect sleep quality, praying before sleeping, sleeping position, attaching significance to dreams etc. For instance, there is the believe by segments of Turkish people that “time spent sleeping is a wasted time, hence sleep time ought to be short” [49], others believe that successful people invest much time being busy,

14

Anujuo_binnenwerk_FINAL.indd 14 30/07/2018 21:19 General introduction

and therefore require short sleep [50], and yet, others disregard the need for napping. Individuals having these beliefs are prone to be short sleepers, and likely to be at risk for short sleep associated problems [49]. On the other hand, individuals who belief that sleep restores or refreshes the body, 1 and that daytime sleep (napping) is important tend to have increased sleep duration, and are less likely to be at risk of short sleep associated health problems. In addition, psychosocial stressors including depression which is more common among ethnic minority groups may affect sleep duration. However, the level of depression could also differ and may influence the sleep patterns between ethnic groups.

According to Knutson (2012) [44], other important factors which determine inadequate sleep lie within the biological domain including certain hormones such as melatonin, and those associated with HPA axis [44,47]. Differences between groups in prevalence of the hormones, as well as hormonal activity levels, which, when at different physiological states, may have differential influence on sleep patterns and may differ across various ethnic groups. Secretion of melatonin (a sleep promoter by the pineal gland of the brain) stimulates sleep, whereas its under-secretion decreases sleep duration. In addition, high level secretion of hormones of the HPA axis, notably cortisol has been shown to impair sleep [47].

Other factors within the environment which may determine inadequate sleep include light, ambient temperature, residential areas (high traffic, noise, pollution (bad air quality), neighborhood crime levels). Ethnic minority groups could be disproportionately affected by these factors as they usually congregate in similar residential areas, usually with low socioeconomic status (SES), and poor neighbourhoods associated with noise and crimes, and these may not only expose them to inadequate sleep, but also provoke anxiety and stress which can impair sleep duration or quality [44].

In sum, there appears to be insufficient evidence for the difference in sleep quality or patterns among ethnic minorities and their role on ethnic inequalities in health outcomes. Hence the purpose of this thesis is to investigate ethnic differences in sleep duration and its role on ethnic differences in CVD and risk factors.

15

Anujuo_binnenwerk_FINAL.indd 15 30/07/2018 21:19 Chapter 1

CONCEPTUAL MODEL OF THIS THESIS

Fig. 3: Conceptual model of this thesis

The conceptual model derived from Knutson (Fig. 1) [44], which guided the studies contained in this thesis, is shown in Fig. 3. This is based on the concept that ethnicity may shape sleep duration, meaning that ethnicity influences whether an individual sleeps short or healthy through a differential distribution of risk factors (A), and ethnicity is an effect modifier for the relationship between sleep duration and CVD risk factors (B), in the sense that this association differs between ethnic groups.

16

Anujuo_binnenwerk_FINAL.indd 16 30/07/2018 21:19 General introduction

OBJECTIVES OF THIS THESIS

The increased prevalence of CVD and risk factors among ethnic minority groups requires that adequate measures are taken to reduce the increased risk 1 through prevention and proper management of the contributing factors. However, the known risk factors of CVD do not entirely explain ethnic inequalities in CVD. It therefore becomes imperative to identify novel modifiable risk factors to contribute to the prevention and management efforts. This thesis aims to assess differences in sleep, and the role of sleep in ethnic differences in CVD and risk factors through the following research objectives:

1. To assess inequalities in sleep duration in the Netherlands between ethnic groups in both adults and pre-school children, and to explore possible factors that explain these inequalities.

2. To assess the associations between short sleep duration and cardiovascular risk factors among ethnic groups in both adults and preschool children.

3. To assess the association between short sleep duration and CVD among ethnic groups, and to quantify how much short sleep contributes to ethnic differences in CVD.

ETHNIC MINORITY GROUPS IN EUROPE AND THE NETHERLANDS

A steady increase in migration flow towards Europe has been observed in the last few decades. The late 1980s witnessed a two times increase in the number of first generation ethnic minority groups [51]. At the beginning of 2016, the number of migrants living in the European Union member (EU 28) states was 20.7 million, equivalent to 4.1% of the total population [52].

The Netherlands is an increasing ethnically diverse country. In 2016, the number of inhabitants in the Netherlands with a migrant background was 3.8 million, which represent 22.1% of the total population [53]. Of the 3.8 million inhabitants, 2.1 million had background from low and middle income countries, whereas 1.7 million had a western background. Nearly half of those with background from low and middle income countries were either second generation or born in the Netherlands [53]. This percentage

17

Anujuo_binnenwerk_FINAL.indd 17 30/07/2018 21:19 Chapter 1

doubled since 1972, when only 9% of the Dutch population belonged to an ethnic minority group [54]. Of the ethnic minorities, 9.8% is of Western origin and 12.3% of low and middle income countries (mainly from Turkey, Morocco, , the Netherland Antilles and other non-Western) [53]. In the upcoming decennial, the ethnic minority population is expected to grow even further. Prognostic analysis show an increase from 21% towards 31% in 2060. The growth will be most profound among ethnic minorities with background from low and middle income countries, from 12% in 2012 towards 18% in 2060. Presently, the age distribution among ethnic minority groups with background from low and middle income countries is relatively low, only 4% of them are ≥65 years of age. Prognostic study estimate that in 2060, 22% of all ethnic minority groups with background from low and middle income countries are ≥65 years of age, which will be only slightly less than within the ethnic Dutch population [55]. As CVD is a condition of the elderly, and seem to be negatively influenced by short sleep duration, reducing ethnic inequalities in CVD will become even more important in the future, hence it also becomes important to investigate the role of sleep in ethnic inequalities in CVD.

Definition of ethnic background

In this thesis, ethnic background is defined by using the country of birth and parental country of birth. The full description has been reported elsewhere [56]. A person from an ethnic minority group is considered as first generation if born abroad with at least one of the parents born abroad, whereas a second generation ethnic minority is born in the Netherlands with at least one of the parents born abroad. Those with both parents born in the Netherlands are considered as ethnic Dutch or Dutch throughout this thesis. We included ethnic minorities from Suriname, Turkish, Moroccan, and Ghanaian origins, as they are the largest groups residing in the Netherlands. Suriname is a previous Dutch colony located in South America. The Surinamese population is ethnically diverse, and made up of people originating from West Africa (30%), India (37%), Java (15%), China (1.5%), and people of mixed origin [57]. They all migrated to the Netherlands due to political instability in Suriname in 1975 and 1980. Ghanaians migrated to the Netherlands between 1974 and 1983 predominantly for economic reasons, and others such as drought, political instability and the deportation of Ghanaians from Nigeria in the early 1990s. Turkish and Moroccans migrated to the Netherlands in the 1960s

18

Anujuo_binnenwerk_FINAL.indd 18 30/07/2018 21:19 General introduction

and early 1970s to fill labour shortages which were lacking in unskilled occupations. In 1970-1980, Turkish and Moroccans guest workers brought their spouses and children to the Netherlands. 1 OUTLINE OF THIS THESIS

Chapter 1 gives a general introduction of the entire contents of this thesis. In chapter 2, ethnic differences in sleep duration among residents in the Netherlands, stratified by ethnicity and gender are discussed. Chapter 3 reports on the ethnic differences in sleep duration among 5 year olds, and its relationship with overweight and blood pressure using the Amsterdam Born Child and their Development (ABCD) Study. Chapter 4 examines the relationship between short sleep duration and cardiovascular risk factors such as obesity, diabetes, hypertension, total cholesterol, high density lipoprotein (HDL) cholesterol, low density lipoprotein (LDL) cholesterol, and triglycerides among ethnic minority groups and the Dutch host population. In chapter 5, the relationship between short, or long sleep duration and arterial stiffness among ethnic groups stratified by gender is investigated. Chapter 6 examines the association between short sleep duration and CVD among the ethnic groups, and the contribution of short sleep to ethnic inequalities in CVD. Chapter 7 examines the association between depression and sleep duration among various ethnic groups, and the contribution of depression to ethnic differences in sleep duration. Chapter 8 is the general discussion in which the main findings of this thesis are discussed. In addition, the implications of the findings as well as the recommendations for further research are suggested. Lastly, a summary of this thesis is presented.

19

Anujuo_binnenwerk_FINAL.indd 19 30/07/2018 21:19 Chapter 1 1 2 2 2 SUB- QUESTIONS 2011-2013 2008 2011-2014 2011-2014 YEAR OF COLLECTION DATA HELIUS, n=6959 ABCD, n=2986 HELIUS, n=12805 HELIUS, n=10994 STUDY STUDY POPULATION Descriptive statistics, logistic regression analysis Descriptive statistics, logistic regression analysis, Linear regression analysis Descriptive statistics, logistic regression analysis Descriptive statistics, linear regression analysis METHOD OF ANALYSIS Sleep duration (short sleep, long sleep) Short sleep, overweight, systolic blood pressure (SBC), diastolic blood pressure (DBP) diabetes, Obesity, hypertension, total cholesterol, HDL cholesterol, LDL cholesterol, triglyceride Pulse wave velocity (PWV) indicator of arterial stiffness DEPENDENT VARIABLE

Ethnic differences in self- Ethnic differences reported sleep duration in the Netherlands – the HELIUS study in sleep Ethnic differences duration at 5 years old, and its relationship with overweight and blood pressure (ABCD) study Relationship between short sleep duration and cardiovascular risk factors in a multi-ethnic cohort – the HELIUS study Relationship between sleep duration and arterial in a multi-ethnic stiffness population: the HELIUS study TOPIC Summary/overview of the thesis showing determinants and outcomes as presented in this 2 3 4 5 CHAPTER Table 1: Table

20

Anujuo_binnenwerk_FINAL.indd 20 30/07/2018 21:19 General introduction 3 2 SUB- QUESTIONS 1 2011-2015 2011-2015 YEAR OF COLLECTION DATA HELIUS, n=20730 HELIUS, n=21130 STUDY STUDY POPULATION Descriptive statistics, logistic regression analysis Descriptive statistics, logistic regression analysis, population attributable fraction (PAF) METHOD OF ANALYSIS Cardiovascular disease (angina pectoris + intermittent claudication + possible myocardial infarction) Short and long sleep duration DEPENDENT VARIABLE

The contribution of short sleep duration to in ethnic differences cardiovascular disease – the HELIUS study Association between depressed mood and sleep duration among various ethnic groups – the HELIUS study TOPIC 6 7 CHAPTER Table 1 - continued Table

21

Anujuo_binnenwerk_FINAL.indd 21 30/07/2018 21:19 Chapter 1

REFERENCES

1. James Y. Nazroo. The Structuring of ethnic inequalities in health: Economic position, racial discrimination, and racism. Am J Public Health. 2003; 93(2):277–284. 2. Stronks K, Kunst AE. The complex interrelationship between ethnic and socio- economic inequalities in health. J Public Health (Oxf). 2009;31(3):324-5 3. Agyemang C, Van Oeffelen AA, Norredam M,et al. Ethnic disparities in ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage incidence in the Netherlands. Stroke 2014, 11:3236-42. 4. Stronks K, Raveilla AC, Reijneveld SA. Immigrants in the Netherlands: Equal access for equal needs? J Epidemiol Comm Health. 2001;55:701-707. 5. Kurian AK, Cardarelli KM: Racial and ethnic differences in cardiovascular disease risk factors: a systematic review. Ethn Dis. 2007;17(1):143-52. 6. Agyemang C, Addo J, Bhopal R, Aikins Ade G, Stronks K: Cardiovascular disease, diabetes and established risk factors among populations of sub-Saharan African descent in Europe: a literature review. Global Health. 2009;11:5-7. 7. Ramon S. Inequality In Mental Health: The relevance of current research and understanding to potentially effective social work responses.Radical Pschology. 2007;6:1. 8. Fassaert T, de Wit, Verhoeff AP, Tuinebreijer WC, Gorissen WH, Beekman AT, Dekker J. Uptake of health services for common mental disorders by first- generation Turkish and Moroccan migrants in The Netherlands.BMC Public Health. 2009;23:307. 9. Sleep: What it is – how it works – why we do it – what can go wrong. http:// www.howsleepworks.com/what_definition.html. Assessed on March 09, 2017. 10. Watson NF, Badr MS, Belenky G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep 2015, 6:843–844. 11. Sadeh A, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychol. 2004;23(5):542–545. 12. CultureGram 2012 World Editions – Republic of Surinamese. http://www. culturegrams.com/images/samples/SurinameWorld.pdf accessed on March 21 2016. 13. Amira S. ``External factors that influence sleep``, Healthy Sleep Division of Sleep Medicine at Havard Medical School,18 December 2007; Web 15 September 2013. Healthysleep.med.havard.edu/healthy/science/how/ external-factors. 14. Hollan D. Sleeping, dreaming, and health in rural Indonesia and the U.S. urban: A cultural and experiential approach. Soc Sci Med. 2013;79:23-30.

22

Anujuo_binnenwerk_FINAL.indd 22 30/07/2018 21:19 General introduction

15. Swaminathan N. Can a lack of sleep cause psychiatric disorders? Scientific American (published October 23, 2007 online article from website: https:// www.scientificamerican.com/article.can-a-lack-of-sleep-cause/ Accessed on Nov.11, 2016. 1 16. Von Ruesten A, Weikert C, Fietze I, et al. Association of sleep duration with chronic diseases in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Postdam study. PLoS One 2012, 1:e30972. 17. Steptoe A, Peacey V, Wardle J. Sleep duration and health in young adults. Arch Intern Med. 2006;166(16):1689-92. 18. Cappuccio FP, Taggart FM, Kandala NB, Currie A, Peile E, Stranges S, Miller MA. Meta-analysis of short sleep duration and obesity in children and adults. Sleep. 2008;31(5):619-26. 19. Taheri S, Lin L, Austin D, et al. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med 2004, 3:e62. 20. Gangwisch JE, Malaspina D, Boden-Albala B, et al. Inadequate sleep as a risk factor for obesity: analyses of the NHANES I. Sleep 2005, 10:1289-96. 21. Patel SR, Malhotra A, White DP, et al. Association between reduced sleep and weight gain in women. Am J Epidemiol 2006, 10:947-54. 22. Spiegel K, Knutson K, Leproult R, et al. Sleep loss: a novel risk factor for insulin resistance and Type 2 diabetes. J Appl Physiol 2008, 5:2008-19. 23. Gottlieb DJ, Punjabi NM, Newman AB, et al. Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med 2005, 8:863-7. 24. Yaggi HK, Araujo AB, McKinlay JB. Sleep duration as a risk factor for the development of type 2 diabetes. Diabetes Care 2006;29(3):657- 661. 25. Gangwisch JE, Heymsfield SB, Boden-Albala B,et al. Short sleep duration as a risk factor for hypertension: analyses of the first National Health and Nutrition Examination Survey. Hypertension 2006, 5:833-9. 26. Cappuccio FP, Stranges S, Kandala NB, et al. Gender-specific associations of short sleep duration with prevalent and incident hypertension: the Whitehall II Study. Hypertension 2007, 4:693-700. 27. Schwartz J, Allison MA, Ancoli-Isreal S, et al. Sleep, type 2 diabetes, dyslipidemia, and hypertension in elderly Alzheimers caregivers. Arch Gerontol Geriatr 2013, 1:70-7. 28. Kaneita Y, Uchiyama M, Yoshiike N, et al. Associations of usual sleep duration with serum lipid levels. Sleep 2008, 5:645-652. 29. Gangwisch JE, Malaspina D, Babiss LA, et al. Short sleep duration as a risk factor for hypercholesterolemia: analyses of the National Longitudinal Study of Adolescent Health. Sleep 2010, 7: 956–961.

23

Anujuo_binnenwerk_FINAL.indd 23 30/07/2018 21:19 Chapter 1

30. Bjorvatn B, Sagen IM, Øyane N, et al. The association between sleep duration, body mass index and metabolic measures in the Hordaland Health Study. J Sleep Res 2007, 1:66-76. 31. Choi KM, Lee JS, Park HS, et al. Relationship between sleep duration and the metabolic syndrome: Korean National Health and Nutrition Survey 2001. Int J Obes 2008, 7:1091-7. 32. Williams CJ, Hu FB, Patel SR, Mantzoros CS. Sleep duration and snoring in relation to biomarkers of cardiovascular disease risk among women with type 2 diabetes. Diabetes Care 2007, 5:1233-40. 33. Kerkhofs M, Boudjeltia KZ, Stenuit P, et al. Sleep restriction increases blood neutrophils, total cholesterol and low density lipoprotein cholesterol in postmenopausal women: A preliminary study. Maturitas 2007, 2:212-5. 34. Amagai Y, Ishikawa S, Gotoh T, et al. Sleep duration and incidence of cardiovascular events in a Japanese population: the Jichi Medical School cohort study. J Epidemiol 2010, 2:106-10. 35. Meisinger C, Heier M, Löwel H, et al. Sleep duration and sleep complaints and risk of myocardial infarction in middle-aged men and women from the general population: the MONICA/KORA Augsburg cohort. Sleep 2007, 9:1121-7. 36. Hoevenaar-Blom MP, Spijkerman AM, Kromhout D, et al. Sleep duration and sleep quality in relation to 12-year cardiovascular disease incidence: the MORGEN study. Sleep 2011, 11:1487-92. 37. Westerlund A, Bellocco R, Sundström J, et al. Sleep characteristics and cardiovascular events in a large Swedish cohort. Eur J Epidemiol 2013, 6:463- 73. 38. Wingard DB, Berkman LF. Mortality risk associated with sleeping patterns among adults. Sleep 1983, 6:102–7. 39. Ayas NT, White DP, Manson JE, et al. A prospective study of sleep duration and coronary heart disease in women. Arch Intern Med 2003, 163:205–9. 40. Chandola T, Ferrie JE, Perski A, et al. The Effect of Short Sleep Duration on Coronary Heart Disease Risk is Greatest Among Those with Sleep Disturbance: A Prospective Study from the Whitehall II Cohort. Sleep 2010, 6:739-44. 41. Aggarwal S, Loomba RS, Arora RR, et al. Association between sleep duration and prevalence of cardiovascular events. Clin Cardiol 2013, 11:671-6. 42. Ferrie JE, Shipley MJ, Cappuccio FP, Brunner E, Miller MA, Kumari M, Marmot MG. A prospective study of change in sleep duration: associations with mortality in the Whitehall 11 cohort. Sleep. 2007;30(12):1659-66.

24

Anujuo_binnenwerk_FINAL.indd 24 30/07/2018 21:19 General introduction

43. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiatry. 2002;59(2):131-6. 44. KL: Does Inadequate Sleep Play a Role in Vulnerability to Obesity?: American 1 Journal of Human Biology 2012; 24:361-371. 45. Unruh ML, Redline S, An MW, Buysse DJ, Nieto FJ, Yeh JL, Newman AB. Subjective and objective sleep quality and aging in the sleep heart health study. J Am Geriatr Soc. 2008 Jul; 56(7):1218-27. 46. Vitiello MV. Recent Advances in Understanding Sleep and Sleep Disturbances in Older Adults: Growing Older Does Not Mean Sleeping Poorly. Current Directions in Psychological Science. 2009;18(6):316–320. 47. Steiger A. Neurochemical regulation of sleep. J Psychiatr Res. 2007 Oct; 41(7):537-52. 48. Hughes JW, Kobayashi I, Deichert NT: Ethnic differences in sleep quality accompany ethnic differences in night-time blood pressure dipping:Am J Hypertens. 2007;20(10):1104-10. 49. Arslan S, Kocoglu D, Durmus M. Cultural beliefs affecting sleep duration. Sleep Biol Rhythms. 2015;13(9):287-296. 50. Hollan D. Sleeping, dreaming, and health in rural Indonesia and the U.S. urban: A cultural and experiential approach. Soc Sci Med. 2013;79:23-30. 51. Salt J. Trends in Europe`s international migration. In: Rechel B, Mladovsky P, Devillè W, Rijks B, Petrova-Benedict R, Mckee M, editors. Migration and health in the European Union. 1 ed. England: Open University Press; 2012. p. 23. 52. Migration and migrant population statistics. Eurostat. http://ec.europa. eu/eurostat/statistics-explained/index.php/Migration_and_migrant_ population_statistics. Assessed on 28/03/2018 53. Annual report on integration 2016 – summary assessed on 28/03/2018 from: file:///H:/Downloads/JI2016S_web.pdf. 54. Verweij A, Sanderse C, de Beer J. Etniciteit: Wat waren de belangrijkste ontwikkelingen in het verleden? In: Volksgezondheid Toekomst Verkenning, National Kompas Volksgezondheid. Bilthoven: RIVM, 2012. 55. Sanderse C, Verweij A, de Beer J. Etniciteit: Wat zijn de verwachtingen voor de toekomst? In: Volksgezondheid Toekomst Verkenning, Nationaal Kompas Volksgezondheid. Bilthoven: RIVM, 2012. 56. Stronks K, Kulu-Glasgow I, Agyemang C. The utility of ‘country of birth’ for the classification of ethnic groups in health research: the Dutch experience. Ethn Health. 2009;14: 255-269. 57. Bos V, Kunst AE, Keij-Deerenberg IM, Garssen J, Mackenbach JP. Ethnic inequalities in age-and cause-specific mortality in the Netherlands. Int J Epidemiol 2004;33(5):1112-9.

25

Anujuo_binnenwerk_FINAL.indd 25 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 26 30/07/2018 21:19 CHAPTER 2

ETHNIC DIFFERENCES IN SELF-REPORTED SLEEP DURATION IN THE NETHERLANDS – THE HELIUS STUDY

Published

Kenneth Anujuo, Karien Stronks, Marieke B. Snijder, Girardin Jean-Louis, Gbenga Ogedegbe, Charles Agyemang. Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS study. Sleep Med. 2014;15:1115-1121.

Anujuo_binnenwerk_FINAL.indd 27 30/07/2018 21:19 Chapter 2

ABSTRACT

Background: We investigated ethnic differences in sleep duration, and the contribution of socio-economic status (SES) to the observed differences in Amsterdam, the Netherlands.

Methods: 6959 participants (aged 18-71 years) from the multi-ethnic HELIUS cohort were studied. Outcome variables were short sleep (<7hours/ night) and long sleep (≥9hours/night). Comparisons among groups were made using Prevalence Ratios (PRs).

Results: Ethnic minority groups were more likely than ethnic-Dutch to report short sleep with prevalence ranging from 15.1-49.7% in men and 16.3-41.4% in women. Among men, the age-adjusted PRs ranged from 2.15 (95%CI 1.72-2.69) in Turkish to 3.31 (2.75-3.99) in Ghanaians; and among women, from 1.62 (1.30-2.01) in Turkish to 2.52 (2.15-2.95) in African-Surinamese, respectively. The prevalence of long sleep was significantly higher only in Moroccan men; and all the ethnic minority women than in ethnic-Dutch women except for African-Surinamese. Adjustment for SES explains the ethnic difference in long sleep, but not for short sleep.

Conclusion: Ethnic minority groups reported more short sleep than ethnic- Dutch, while there were no ethnic differences in long sleep. Further study is needed to investigate how this finding on short sleep may contribute to ethnic differences in health outcomes.

28

Anujuo_binnenwerk_FINAL.indd 28 30/07/2018 21:19 Ethnic differences in sleep patterns

INTRODUCTION

Evidence demonstrates clear ethnic differences in health outcomes such as CVD and its related risk factors [1-6], mental health [7,8], and health care utilization [9] across the globe. The drivers of the ethnic differences in health outcomes are poorly characterized. Sleep is increasingly being recognized as a factor that might have an impact on health, and may play an important role in the observed differences in health outcomes between ethnic minority groups and the host majority populations. Both short and 2 long sleep duration are established correlates of poor health outcomes [10]. Very little is, however, known about the differences in sleep duration among ethnic minorities and their potential role in health. One of the reasons may be that data on ethnic differences in sleep patterns are limited, and the existing data mainly from the USA show conflicting results. For example, recent studies found distinct variations in sleep duration between ethnic minority groups and White Americans [11,12]. A similar study also found that African Americans had a longer sleep duration (≥8 hours per night) [13] than White Americans. On the other hand, some reports indicate that African Americans, Hispanics and Asians in the USA also have shorter sleep duration (≤7 hours/night) than White Americans [14-15].

Socio-economic differences in sleep is important for understanding how habitual sleep can be influenced. There is also discrepancy in reports concerning the association of socio-economic status (SES) with sleep (duration) among various ethnic groups. For instance, it has been reported that ethnic groups with a lower SES report less sleep duration and/or poorer sleep quality than those with higher SES [16-18]. In addition, in a study involving African Americans, Hispanics, Asians and White Americans, Grandner and colleagues [19] found that lower income and educational attainment and unemployment were associated with more sleep complaints. A similar study by Mezick and colleagues [20] indicated that African Americans in lower SES group may be more at risk for sleep disturbances and the associated health consequences. However, one study found that education was not related to sleep in a study population consisting of African American, White American and Chinese women [21]. Put together, it appears that differences in SES in relation to sleep duration may differ between ethnic groups.

Shift work is also known to affect sleep, particularly night shift [22]. Studies indicate that ethnic minority groups appears to do more shift work than

29

Anujuo_binnenwerk_FINAL.indd 29 30/07/2018 21:19 Chapter 2

the host population [23], hence shift work may also contribute to ethnic differences in sleep duration.

In Europe, there is no data on ethnic differences in sleep duration. Thus, the purpose of the present study was to examine ethnic differences in self-reported sleep duration between the ethnic-Dutch and several ethnic minority groups, using a Dutch population-based study. Further, we assessed the contribution of socio-economic indicators to the observed differences in sleep duration among the ethnic groups.

STUDY POPULATION AND METHODS

Study population

The current study was based on baseline data from the HELIUS (Healthy Life in an Urban Setting) study. The aims and design of the HELIUS study have been described elsewhere [24]. In brief, HELIUS is a large-scale prospective cohort study on health and health care among different ethnic groups living in Amsterdam. The study started in 2011 and it includes individuals aged 18-70 years from the six major ethnic groups in Amsterdam (African- Surinamese, South Asian-Surinamese, Turkish, Moroccan, Ghanaian, and European descent Dutch ( henceforth referred to as ethnic-Dutch). It focuses on three major disease categories: cardiovascular disease, mental health and infectious diseases. Participants were randomly sampled from the municipal registers, stratified by ethnicity. The study protocols were approved by the AMC Ethical Review Board.

Baseline data collection is still ongoing. For the current study, baseline data that were collected until June 2013 were used, including 7307 participants in whom data from the HELIUS questionnaire were available. Participants with a background (n=113), other/unknown Suriname background (n=109) or other/unknown background (n=15) were excluded because of the small sample sizes. In addition, individuals with no data on sleep duration were also excluded from the analysis (111). This resulted in a dataset of 6959 participants, including 1767ethnic Dutch, 1812 African- Surinamese, 1156 South-, 1117 Ghanaians, 844 Turkish and 263 Moroccans. Information on demographic, sleep duration, socio- economic indicators and shift work were collected through a questionnaire.

30

Anujuo_binnenwerk_FINAL.indd 30 30/07/2018 21:19 Ethnic differences in sleep patterns

Measurements

Sleep Participants were asked to provide information on the average number of hours they usually sleep at night. Sleep duration was assessed using the item: “How many hours do you sleep on average per night?” Sleep duration was categorized according to the standard recommendations of the National Sleep Foundation [25]. For adults, 7-9 hours per night is recommended. Short sleep is defined as having less than 7 hours of sleep per night, and long 2 sleep as having 9 or more hours of sleep per night.

Ethnicity Participant’s ethnicity was defined according to the country of birth of the participant as well as that of his/her parents. Specifically, a participant is considered as an ethnic minority if he/she fulfils either of the following criteria: 1) he or she was born abroad and has at least one of his/her parents born abroad; or 2) he or she was born in the Netherlands but has both his/ her parents born abroad [26]. The Surinamese are migrants from Suriname which is a former Dutch colony in South America. who have an African-Caribbean background are mostly the descendants of West African, while those with a South Asian background have their origin in North India. They all migrated from Suriname to the Netherlands due to the political instability in Suriname in 1975 and 1980. Ghanaians migrated to the Netherlands between 1974 and 1983 predominantly for economic reasons, and others such as drought, political instability and the deportation of Ghanaians from Nigeria in the early 1990s. The Turkish and Moroccans migrated to the Netherlands in the 1960s and early 1970s to fill labour shortages which were lacking in unskilled occupations. In 1970-1980, Turkish and Moroccans guest workers brought their spouses and children to the Netherlands.

Socio-economic indicators Socio-economic indicators were assessed through participant’s highest level of education (either in the Netherlands or in the country of origin) and the current employment status. Participants were categorized into those who have never been to school or had elementary schooling only (1st category), those with lower vocational schooling or lower secondary schooling (2nd category), those with intermediate vocational schooling or intermediate/ higher secondary education schooling (3rd category), and those with higher vocational schooling or university (4th category). For the current analyses,

31

Anujuo_binnenwerk_FINAL.indd 31 30/07/2018 21:19 Chapter 2

the first two categories were combined because of small sample size. Employment was categorized into those who were employed and those who were not.

Shift work Information on shift work was obtained by asking the participants if they do shift in their work schedules and if they have irregular working hours including night work. Shift work was assessed using the item “Do you work irregular hours, including services during night hours?”

Data Analysis

Baseline data were expressed as percentages or means with 95% CI. Chi-square tests were used to test the differences in categorical variables between ethnic groups. Comparisons among ethnic groups were made using Prevalence Ratios (PRs). To assess the contribution of socio-economic indicators to the ethnic differences in sleep, education and employment, were included in the regression models. The change in PRs, before and after adjustment was used to assess the relative contribution of socio-economic indicators to the ethnic differences in sleep duration. The same was done for shift work. Interaction effect between ethnicity and socio-economic indicators was also assessed. All analyses were performed using STATA 11.0 (Stata Corp, Station, TX). A p-value of <0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

Table 1 shows the characteristics of the study population by ethnic group and sex. Turkish and Moroccan men and women, and Ghanaian women were younger, less employed and had lower educational levels than the ethnic-Dutch and other ethnic groups. South-Asian Surinamese men and African-Surinamese women have higher prevalence of shift work compared with other ethnic groups.

32

Anujuo_binnenwerk_FINAL.indd 32 30/07/2018 21:19 Ethnic differences in sleep patterns

Sleep duration

Figure 1 shows mean sleep duration (hours per night) in men and women by ethnic group. Among men, the mean sleep duration levels were significantly lower in all the ethnic minority groups (ranging from 6.6 to 7.1 hours per night) than in ethnic-Dutch (7.3 hours per night), with Ghanaians and African Surinamese reporting the lowest mean sleep duration. Among women, the mean sleep duration levels ranges from 6.7-7.3 hours per night in ethnic minority groups and 7.3 hours per night in the ethnic-Dutch, 2 with African-Surinamese reporting the lowest mean sleep duration.

Short sleep

Figures 2a shows the prevalence of short sleep in men and women by ethnic group. The prevalence of short sleep was lower in ethnic-Dutch compared to all ethnic minority groups in both men and women. Table 2 shows that these ethnic differences were significant after adjusting for age, socio- economic indicators (education and employment), and shift work with PRs ranging from 2.31 (95% CI, 1.25-4.26) in Moroccan men to 3.49 (95% CI, 2.33-5.22) in Ghanaian men compared with ethnic-Dutch men, and 1.64 (95% CI, 1.07-2.50) in Turkish women to 2.27 (95% CI, 1.71-3.00) in African-Surinamese women compared with ethnic-Dutch women.

When <5 hourr and 5-6 hours were used as cutoff for very short sleep and short sleep, and after adjusting for SES, observed ethnic differences in very short sleep persisted except for South-Asian Surinamese and Turkish men, and for Turkish and Moroccan women; the ethnic differences in short sleep also persisted after adjusting for SES, except for Moroccan men and women (Annex table 1)

Long sleep

Figures 2b shows the prevalence of long sleep among ethnic groups as compared with ethnic-Dutch. The prevalence of long sleep was low among all ethnic groups, ranging from 3.4% to 9.4% in men and from 5.6% to 12.0% in women. Table 3 shows that in men, compared with ethnic-Dutch, the age-adjusted PR of long sleep was significantly higher only in Moroccan men (PR=1.91 95% CI 1.05-3.48), but the difference disappeared after

33

Anujuo_binnenwerk_FINAL.indd 33 30/07/2018 21:19 Chapter 2

adjusting for education, employment and shift work (PR=1.04 95% CI 0.40-2.71). In women, compared with ethnic-Dutch, the age-adjusted PR of long sleep was significantly higher in South-Asian Surinamese (PR=1.58, 1.06-2.36), Ghanaian (PR=1.77, 1.20-261), Turkish (PR=1.97, 1.24- 3.13) and Moroccans (PR=2.92, 1.61-5.29) respectively. The differences disappeared after further adjustment for SES

Association between Socio-economic indicators and sleep

We observed a negative association between short sleep and educational level for the ethnic-Dutch only, whereas for the other ethnic groups, no association was observed (Figure annex 3a & b). For long sleep, there was significant positive association between education attainment and long sleep only in the African Surinamese men (Figure annex 4a & b). Whereas a significant positive association between employment and short sleep was observed in South-Asian Surinamese men and African Surinamese women (Figure annex 5a & b), there was a significant positive association between employment and long sleep, but only in the ethnic-Dutch women (Figure annex 6a & b). For short sleep, interaction was formally tested and there was interaction effect between education and the ethnic groups, but only in the women category. Interaction results show that all ethnic minority women with secondary and tertiary education were more likely to report short sleep than ethnic-Dutch with same educational level.

34

Anujuo_binnenwerk_FINAL.indd 34 30/07/2018 21:19 Ethnic differences in sleep patterns n = 138 Moroccan 40.8 (38.7, 42.8) 40.8 (38.7, 34.9 (27, 43) 83.7 (73, 94) 27.0 (19.5, 34.5) 27.0 (19.5, 66.7) 58.4 (50.1, 14.6 (8.7, 20.5) n = 125 41.0 (38.9, 43.2) 41.0 (38.9, 15.9 (8, 24) 48.1 (37, 59) 39.2 (30.6, 47.8) 39.2 (30.6, 57.6) 48.8 (40.0, 17.7) 12.0 (6.28, 2 Turkish n = 396 40.2 (39.1, 41.4) 30.8 (26, 36) 6.3 (69, 84) 28.4 (23.9, 32.8) 57.5 (52.6, 62.4) 14.2 (10.7, 17.6) n = 448 39.7 (38.5, 40.8) 8.1 (5, 11) 44.5 (39, 51) 40.7 (36.2, 45.3) 45.9 (41.2, 50.5) 13.4 (10.3, 16.6) n = 455 Ghanaian 47.4 (46.3, 48.6) 27.5 (23, 32) 87.5 (82, 93) 12.9 (9.81, 16.0) 68.6 (64.3, 72.9) 18.5 (14.9, 22.1) n = 662 43.5 (42.6, 44.4) 15.2 (12, 18) 80.8 (77, 85) 35.4 (31.8, 39.1) 59.7 (55.9, 63.4) 4.91 (3.25, 6.57) African Surinamese n = 685 46.8 (45.8, 47.8) 46.8 (45.8, 47.8) 90.8 (87, 95) 7.67 (5.66, 9.68) 75.7 (72.4, 78.9) 16.7 (13.9, 19.5) n = 1127 46.8 (46.2, 47.6) 22.5 (20, 25) 86.3 (83, 90) 5.35 (4.03, 6.67) 68.0 (65.3, 70.7) 26.7 (24.1, 29.2) South-Asian Surinamese n = 549 44.9 (43.8, 46.1) 37.5 (33, 42) 81.8 (76, 87) 16.6 (13.5, 19.8) 63.1 (59.0, 67.1) 20.3 (16.9, 23.7) n = 607 46.9 (46.0, 48.0) 17.9 (15, 21) 78.9 (74, 84) 17.0 (14.0, 20.0) 63.5 (59.6, 67.3) 19.5 (16.3, 22.7) Dutch n = 816 47.4 (46.5, 48.4) 24.7 (22, 28) 90.3 (86, 94) 3.43 (2.18, 4.68) 39.1 (37.7, 42.4) 57.5 (54.1, 60.9) n = 951 45.9 (45.0, 46.8) 16.7 (14, 19) 86.8 (83, 91) 2.65 (1.62, 3.68) 38.7 (35.6, 41.8) 58.6 (55.5, 61.8) Characteristics of the study population by sex and ethnicity None or Primary (%) Secondary (%) (%) Tertiary None or Primary (%) Secondary (%) (%) Tertiary Men Age Shift work (% yes) Employed (% yes) Education Women Age Shift work (% yes) Employed (% yes) Education Table 1: Table Data are presented as mean and percentages with 95% CI

35

Anujuo_binnenwerk_FINAL.indd 35 30/07/2018 21:19 Chapter 2 n = 138 2.23 (1.65-3.01)* 2.69 (1.54-4.69)* 2.31 (1.25-4.26)** n = 125 1.71 (1.24-2.37)** (1.40-3.18)* 2.11 1.81 (1.05-3.12)** Moroccan PR (95%CI) n = 396 2.15 (1.72-2.69)* 2.75 (1.77-4.28)* 2.72 (1.71-4.33)** n = 448 1.62 (1.30-2.01)* 1.64 (1.15-2.33)** 1.64 (1.07-2.50)** Turkish PR (95%CI) n = 455 3.31 (2.75-3.99)* 3.38 (2.29-5.00)* 3.49 (2.33-5.22)* n = 662 2.43 (2.04-2.88)* 2.36 (1.75-3.19)* 2.23 (1.61-3.08)* Ghanaian PR (95%CI) n = 685 3.16 (2.64-3.79)* 3.12 (2.14-4.53)* 3.07 (2.08-4.52)* n = 1127 2.52 (2.15-2.95)* 2.27 (1.74-2.98)* 2.27 (1.71-3.00)* African Surinamese PR (95%CI) n = 549 2.59 (2.13-3.14)* 2.60 (1.92-4.17)* 2.98 (1.99-4.44)* n = 607 2.12 (1.77-2.53)* 2.18 (1.63-2.93)* 2.17 (1.59-2.96)* South-Asian Surinamese PR (95%CI) Dutch n = 816 1.00 1.00 1.00 n = 951 1.00 1.00 1.00 Prevalence ratio(s) for short sleep among ethnic minority groups compared with Dutch Men Short sleep (<7 h/ night) Model 1 Model 2 Model 3 Women Short sleep (<7 h/ night) Model 1 Model 2 Model 3 Model 1: adjusted for age, Model 2: Age, education, and employment, Model 3: Age, education, employment and shift work. (*: p < 0.001, **: Age, education, and employment, Model 3: Model 1: adjusted for age, 2: 0.05) Table 2: Table

36

Anujuo_binnenwerk_FINAL.indd 36 30/07/2018 21:19 Ethnic differences in sleep patterns 2.04 (0.84-4.96) 0.97 (0.45-2.08) 1.99 (1.16-3.40)** n = 125 1.04 (0.40-2.71) 1.28 (0.55-2.97) 1.91 (1.05-3.48)** n = 138 Moroccan PR (95%CI) 2 1.62 (0.79-3.31) 1.19 (0.70-2.01) 1.87 (1.29-2.70)** n = 448 0.68 (0.28-1.70) 0.91 (0.43-1.95) 1.43 (0.89-2.29) n = 396 Turkish PR (95%CI) 1.25 (0.66-2.36) 1.01 (0.59-1.72) 1.54 (1.08-2.20)** n = 662 0.85 (0.40-1.82) 0.84 (0.41-1.75) 0.92 (0.54-1.59) n = 455 Ghanaian PR (95%CI) 0.88 (0.47, 1.63) 0.76 (0.45-1.27) 0.98 (0.69-1.39) n = 1127 0.47 (0.22-1.03) 0.55 (0.27-1.13) 0.73 (0.44-1.22) n = 685 African Surinamese PR (95%CI) 1.25 (0.65-2.40) 1.09 (0.63-1.89) 1.47 (1.02-2.12)** n = 607 0.56 (0.25-1.26) 0.78 (0.40-1.51) 1.09 (0.68-1.76) n = 549 South-Asian Surinamese PR (95%CI) 1.00 1.00 1.00 n = 951 1.00 1.00 1.00 Dutch n = 816 Prevalence ratio(s) for long sleep among ethnic minority groups compared with Dutch Model 3 Model 2 Long sleep (>=9 h/ night) Model 1 Women Model 3 Model 2 Long sleep (>=9 h/ night) Model 1 Men Model 1: adjusted for age, Model 2: Age, education, and employment, Model 3: Age, education, employment and shift work. (*: p < 0.001, **: Age, education, and employment, Model 3: Model 1: adjusted for age, 2: 0.05) Table 3: Table

37

Anujuo_binnenwerk_FINAL.indd 37 30/07/2018 21:19 Chapter 2

Men Women

Fig 1: Mean sleep duration (hours per night) with 95% CI among ethnic groups by gender.

Men Women

Fig. 2a: Prevalence short sleep with 95% CI among ethnic groups by gender

Men Women

Fig. 2b: Prevalence long sleep with 95% CI among ethnic groups by gender

38

Anujuo_binnenwerk_FINAL.indd 38 30/07/2018 21:19 Ethnic differences in sleep patterns

DISCUSSION

Our study findings indicate that all ethnic minority groups were more likely to report shorter sleep duration than the ethnic-Dutch, in both men and women. Contribution of ethnic differences in educational level to ethnic differences in short sleep was small. There were ethnic differences in long sleep for Moroccan men and women, as well as in South-Asian Surinamese, Ghanaian, and Turkish women compared with ethnic-Dutch. SES explains the observed differences in long sleep but contribution of SES to ethnic 2 differences in short sleep was small.

Our present finding on the higher prevalence of short sleep among ethnic minority groups compared with the ethnic-Dutch confirms some previous studies in the United States which also found a higher prevalence of short sleep duration (<7 hours per night) among ethnic minority groups, including African Americans, Hispanics and Asian Americans compared to White Americans [14-15]. The reasons for the high prevalence of short sleep found among ethnic minority groups are unclear, but may be attributed to other factors such as migration stress, environmental and cultural factors rather than socio-economic indicators. It is a known fact that ethnic minority groups usually perform different kinds of jobs, and sometimes multiple jobs which may deprive them of the ideal time for having sufficient sleep. Since socio-economic indicators (education and employment) differences contributed little to the observed ethnic differences in short sleep, it seems that the observed difference may be more type of job-related in some ethnic minority groups than in others. Short sleep can be caused by undue stress which accompanies rigorous tasks in work places and the environment in which ethnic minority groups live. The negative effect of stress on sleep has been documented [27, 28]; thus psychosocial stress due to migration and perceived discrimination may have undesirable effects on sleep in migrant and ethnic minority populations. Congested and noisy environment arising from various sources negatively affect night sleep of residents living in those areas [29-30]. In addition to work and stress-related factors, cultural practices such as communal sleeping habits notably common among African Surinamese [30], beliefs and attitudes towards sleep [28, 29] can affect normal sleep schedules, and therefore may contribute to high prevalence of short sleep among ethnic minority groups. For instance, it has been suggested that, although people may report short sleep duration, they may seem contented owing to their belief that hard working, and successful people require short sleep duration [28, 31]. Also other factors such as inherent values ascribed

39

Anujuo_binnenwerk_FINAL.indd 39 30/07/2018 21:19 Chapter 2

to sleep [32], bedtimes [33-36], and napping [34, 37-40] have been shown to differ among adults in various parts of the globe and may also differ among ethnic minority groups in our study. It could be that ethnic minority groups value sleep less, have delayed bedtimes and nap more, which may result in shorter sleep duration compared with ethnic-Dutch. This needs to be further investigated. In addition, literature indicates that acculturation seems to play a protective role in several health behaviours, for example, in greater exercise and leisure-time physical activity [40]. However, using primary language spoken as basis for assessing acculturation, Whinnery et al. [42] demonstrated that exclusively Spanish-speaking households were 2-3 times less likely to report very short sleep compared to primary English- speaking households. Another study conducted on younger Hispanics living in the USA indicated that acculturation was associated with decreased likelihood of health-promoting behaviour such as hour of sleep per night [43], which seem to suggest that acculturation did not have a protective role in these studies. Thus, it could be said that the role of acculturation may differ from one ethnic group to another. However, we did not investigate the role of acculturation in our study.

Shift work is known to be highly related to sleep duration, and ethnic minority groups tend to do more shift work than the host populations [22,23,44]. Shift work (night shift or other unconventional hours as in multiple jobs) has the potential to disrupt circadian rhythm, thereby affecting physiological, biochemical, behavioural and other related processes of the body, and making normal sleeping difficult [45]. Previous studies show that ethnic minority groups have higher prevalence of short sleep due to shift work when compared with non-Hispanic Whites [23,46]. However, the ethnic differences in short sleep persisted after adjustment for ethnic differences in shift work in our study.

Overall, the contribution of socio-economic indicators on ethnic differences in short sleep was small. This finding is consistent with a previous study which found that though, substantial ethnic differences in sleep exists, yet the differences were not accounted for by SES [21]. On the other hand, this finding is opposed to some other studies which found that ethnic groups with lower SES report lower sleep duration and sleep quality than those with higher SES [17-20]. The reasons for these inconsistent results are not clear. It could be that SES measures vary between studies, or the extent of the variation in SES differs between studies. On one hand, we found that socio-economic indicators were associated with ethnic differences in

40

Anujuo_binnenwerk_FINAL.indd 40 30/07/2018 21:19 Ethnic differences in sleep patterns

long sleep particularly in women. This is in accordance with previous study which suggests that long sleep is associated with lower SES [20]. On the other hand, this finding is opposed to another study which suggests that SES (education) was not related to sleep [21]. In the association between education and short sleep, education was inversely related to short sleep but only in the ethnic-Dutch group. This finding agrees with previous studies by Hale [44,45] and Grandner et al. [20], that people in lower educational category are more likely to have short sleep duration compared to those in higher education categories. For the ethnic minority groups, irrespective of 2 educational level, education does not relate to short sleep. The reason for the observed difference in the association between education and short sleep in the ethnic-Dutch and ethnic minority groups is not clear, but may be attributed to less awareness and disregard of the implications of short sleep by ethnic minority groups compared with ethnic-Dutch.

The strength of our study lies in the large sample size used in most of the ethnic groups, hence a more reliable result. Also multiple ethnic groups were investigated together in a similar manner. However, the study has some weaknesses such as use of self-reported data, which can be prone to subjective bias; the participants may have under- or over- reported short or long sleep duration. Being a cross-sectional study, the relationship between ethnic groups, socio-economic indicators and sleep duration may not be generalised, even though associations were observed. Also, information about daytime sleeping and cultural attitudes such as inherent values ascribed to sleep, bedtimes, and napping, which may affect sleep were lacking in the original dataset. Despite these shortcomings, our study shows a distinct and relevant result on ethnic differences in sleep duration among residents in Europe.

In conclusion, our current findings show that ethnic minority groups report shorter sleep duration compared with their ethnic-Dutch counterparts. Hence, it might be interesting to further investigate how these ethnic differences in short sleep may be related to ethnic differences in mental health and possibly contribute to ethnic inequalities in health outcomes such as cardiovascular disease.

41

Anujuo_binnenwerk_FINAL.indd 41 30/07/2018 21:19 Chapter 2

ANNEXES

Fig. 3a: Association between education and short sleep among ethnic groups in Amsterdam (men)

Fig. 3b: Association between education and short sleep among ethnic groups in Amsterdam (women)

42

Anujuo_binnenwerk_FINAL.indd 42 30/07/2018 21:19 Ethnic differences in sleep patterns

2

Fig. 4a: Association between education and long sleep among ethnic groups in Amsterdam (men)

Fig. 4b: Assiciation between education and long sleep among ethnic groups in Amsterdam (women)

43

Anujuo_binnenwerk_FINAL.indd 43 30/07/2018 21:19 Chapter 2

Fig. 5a: Association between employment and short sleep among ethnic groups in Amsterdam (men)

Fig. 5b: Association between employment and short sleep among ethnic groups in Amsterdam (women)

44

Anujuo_binnenwerk_FINAL.indd 44 30/07/2018 21:19 Ethnic differences in sleep patterns

2

Fig. 6a: Association between employment and long sleep among ethnic groups in Amsterdam (men)

Fig. 6b: Association between employment and long sleep among ethnic groups in Amsterdam (women)

45

Anujuo_binnenwerk_FINAL.indd 45 30/07/2018 21:19 Chapter 2 1.59 (0.76-3,35) 2.03 (1.05-3.92)** 1.83 (1.31-2.56)* 1.68 (1.19-2.34)* 22.2 (2.21-224)** 11.3 (1.80-70.1)** 11.3 19.5 (5.16-73.9)* 15.8 (4.23-58.7)* n = 138 Moroccan PR (95%CI) 2.28 (1.39-3.76)** 2.55 (1.59-4.09)* 2.02 (1.60-2.55)* 1.84 (1.46-2.32)* 15.4 (1.69-140)** 5.62 (0.93-34.1) 6.97 (1.82-26.7)** 5.49 (1.47-20.6)** n = 396 Turkish PR (95%CI) 3.10 (2.03-4.74)* 3.15 (2.08-4.76)* 2.99 (2.47-3.64)* 2.99 (2.46-3.63)* 14.0 (1.61-122)** 6.71 (1.29-34.8)** 15.9 (4.81-52.3)* 16.5 (4.72-51.1)* n = 455 Ghanaian PR (95%CI) 2.57 (1.71-3.87)* 2,72 (1.83-4.05)* 2.76 (2.28-3.33)* 2.73 (2.26-3.29)* 16.9 (2.13-134)** 8.60 (1.86-39.9)** 19.5 (6.09-62.6)* 19.1 (5.96-60.9)* n = 685 African Surinamese PR (95%CI) 2.71 (1.78-4.11)* 2.71 (1.80-4.07)* 2.43 (1.99-2.97)* 2.35 (1.92-2.88)* 10.6 (1.24-90.5)** 4,54 (0.85-24.2) 8.56 (2.49-29.4)** 7.93 (2.32-27.1)** n = 549 South-Asian Surinamese PR (95%CI) Prevalence ratio(s) for very short sleep among ethnic minority groups compared with Dutch 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 n = 816 Dutch Model 3 Model 2 Model 1 Crude Model 3 Model 2 Model 1 Crude Short sleep (5-6 h/ night) Very short sleep Very (<5 h/ night) Men Supplementary Table 1: Supplementary Table

46

Anujuo_binnenwerk_FINAL.indd 46 30/07/2018 21:19 Ethnic differences in sleep patterns 1.64 (0.88-3.05) 2.00 (1.27-3.17)** 1.48 (1.03-2.12)** 1.38 (0.96-1.99) 3.37 (0.64-17.8) 3.13 (0.79-12.3) 8.16 (2.55-26.1)* 6.34 (1.96-20.5)** n = 125 Moroccan PR (95%CI) 2 1.65 (1.05-2.60)** 1.56 (1.06-2.29)** 1.45 (1.15-1.82)** 1.34 (1.06-1.69)** 1.22 (0.22-6.65) 2.49 (0.78-7.96) 6.23 (2.39-16.2)* 4.59 (1.76-12,0)** n = 448 Turkish PR (95%CI) 1.99 (1.39-2.87)* 2.08 (1.49 (2.91)* 2.11 (1.75-2.53)* 2.11 2.02 (1.68-2.44)* 4.32 (1.47-12.7)** 5.11 (1.86 (14.1)** 5.11 11.1 (4.75-25.8)* 11.1 9.34 (3.98-21.9)* n = 662 Ghanaian PR (95%CI) 2.04 (1.49-2.77)* 2.01 (1.49-2.70)* 2.27 (1.92-2.68)* 2.29 (1.94-2.70)* 4.30 (1.66-11.7)** 4.85 (1.92-12.2)** 8.90 (3.89-20.4)* 8.86 (3.85-20.4)* n = 1127 African Surinamese PR (95%CI) 2.03 (1.44-2.86)* 2.08 (1.51-2.87)* 1.97 (1.63-2,37)* 1.99 (1.64-2.40)* 3.44 (1.22-9.65)** 3.19 (1.14-8.93)** 5.96 (2.45-14.5)* 6.01 (2.46-14.7)* n = 607 South-Asian Surinamese PR (95%CI) 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 n = 951 Dutch Model 3 Model 2 Model 1 Crude Model 3 Model 2 Model 1 Crude Short sleep (5-6 h/ night) Very short sleep Very (<5 h/ night) Women Supplementary Table 1 - continued Supplementary Table Age, education, employment and shift work. (*: p < 0.001, **: Age, education, and employment, Model 3: Model 1: adjusted for age, 2: 0.05)

47

Anujuo_binnenwerk_FINAL.indd 47 30/07/2018 21:19 Chapter 2

REFERENCES

1. Agyemang C, Addo J, Bhopal R, Aikins Ade G, Stronks K. Cardiovascular disease, diabetes and established risk factors among populations of sub-Saharan African descent in Europe: a literature review, Global Health 2009;5:7. 2. Kurian AK,Cardarelli KM. Racial and ethnic differences in cardiovascular disease risk factors: a systematic review, Ethn Dis. 2007;17(1):143-152. 3. Jonkers M, Richters A, Zwaart J, et al. Ethnic Differences in Maternal Morbidity in the Netherlands: Anthropological Reflections on Socio-Ethnic Determinants. Int J Psychosom Obstet Gynecol. 2007;28:48. 4. Mackenbach JP, Bos V, Garssen MJ, et al. Mortality among non-western migrants in The Netherlands,Ned Tijdschr Geneeskd. 2005;149(17):917-923. 5. van Oeffelen AA, Vaartjes I, Stronks K,et al. Incidence of acute myocardial infarction in first and second generation minority groups: Does the second generation converge towards the majority population? Int J Cardiol. 2013;168(6):5422-5429. 6. Finkelstein EA, Khavjou OA, Mobley LR, et al. Racial/ethnic differences in coronary heart disease risk factors among WISEWOMEN enrollees, J Womens Health. (Larchmt). 2004;13(5):503-518. 7. Fassaert TJL, De Wit MAS, Verhoeff AP,et al. Uptake of Health Services for Common Mental Disorders by First-Generation Turkish and Moroccan Migrants in The Netherlands,BMC Public Health. 2009;23:307. 8. Ramon S. Inequality In Mental Health. The Relevance of Current Research and Understanding to Potentially Effective Social Work Responses,Radical Psychol. 2007;6 (1):1-20. 9. Stronks K, Ravelli AC, Reijneveld SA. Immigrants in the Netherlands. Equal access for equal needs? J Epidemiol Community Health. 2001;55(10):701-707. 10. Shankar A, Charumathi S, Kalidindi S. Sleep Duration and Self-Rated Health: the National Health Interview Survey. Sleep. 2008;34(9):1173-1177. 11. Profant J, Ancoli-Isreal S, Dimsdale JE. Are there ethnic differences in sleep Architecture? Am J Hum Biol. 2002;14(3):321-326. 12. Department of Health and Human Services 2003 sleep disorders research plan. Washington, D.C; Department of Health and Human Services. 13. Quereshi AI, Giles WH, Croft JB, et al. Habitual sleep patterns and risk for stroke and coronary heart disease: a 10-year follow-up from NHANES 1. Neurology. 1997;48(4):904-911. 14. Kripke DF, Brunner R, Freeman R, et al. Sleep Complaints of Postmenopausal Women. Clin J Womens Health. 2001;1(5):244-252. 15. Alice Park. ``How Well Do You Sleep? The Answer May Depend on Your Race``,Time, Time Inc. 2012, Web. 10 August 2013.

48

Anujuo_binnenwerk_FINAL.indd 48 30/07/2018 21:19 Ethnic differences in sleep patterns

16. Jean-Lous G, Magai CM, Cohen CI, et al. Ethnic Differences in Self-Reported Sleep Problems in Older Adults. Sleep. 2001;24(8):926-933. 17. Gellis LA, Lichstein KL, Scarinci IC, et al. Socioeconomic status and insomnia, J Abnorm Psychol. 2005;114(1):111–118. 18. Patel NP, Gooneratne N, Xie D, Branas CC. Poverty effects on population sleep, Sleep. 2008;31:A353 . 19. Grandner MA,Patel NP, Gehrman PR, Xie D, Sha D, Weaver T, Gooneratne N. Who gets the best sleep? Ethnic and socioeconomic factors related to sleep complaints, Sleep Med. 2010 May;11(5):470-478. 2 20. Mezick EJ, Matthews KA, Hall M, et al. Influence of race and socioeconomic status: Pittsburgh Sleep SCORE Project. Psychosom Med. 2008;70(4):410- 416. 21. Hall MM, Matthews KA, Kravitz HM, et al. Race and financial strain are independent correlates of sleep in midlife women: The SWAN Study.Sleep . 2009;32 (1):73-82. 22. Luckhaupt SE. Short Sleep Duration Among Workers – United States, 2010. Weekly. 2012;61(16):281-285. 23. U.S. Department of Labour Bureau of Statistic. (2005). Workers on Flexible and Shift Schedules in May 2004. Retrieved December 22, 2013 from http:// www.bls.gov/news.release/flex.toc.htm 24. Stronks K, Snijder MB, Peters RJ, et al. Unravelling the impact of ethnicity on health in Europe: the HELIUS study. BMC Public Health. 2013;13:402. 25. Sleep guidelines for adults. Retrieved April 19, 2012 from http://www. sleepfoundation.org/alert/new-research-identity-and-treat-insomnia-early- reduce-risk-other-illness. 26. Stronks K, Kunst AE. The complex interrelationship between ethnic and socio- economic inequalitiesin health, J Public Health (Oxf). 2009;31(3):324-5. 27. Sadeh A, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychology. 2004;23(5):542–545. 28. Amira S. ``External factors that influence sleep``, Healthy Sleep Division of Sleep Medicine at Havard Medical School,18 December 2007; Web 15 September 2013.Healthysleep.med.havard.edu/healthy/science/how/ external-factors. 29. Knutson KL. Does Inadequate Sleep Play a Role in Vulnerability to Obesity? Am J Hum Biol. 2012;24:361-371. 30. CultureGram 2012 World Editions – Republic of Surinamese. http://www. culturegrams.com/images/samples/SurinameWorld.pdf assessed on March 21 2014. 31. Hollan D. Sleeping, dreaming, and health in rural Indonesia and the U.S. urban: A cultural and experiential approach. SocSci Med. 2013;79:23-30.

49

Anujuo_binnenwerk_FINAL.indd 49 30/07/2018 21:19 Chapter 2

32. Worthman CM, Brown RA. Sleep budgets in a globalizing world: biocultural interactions influence sleep sufficiency among Egyptian families. Soc Sci Med. 2013;79:31-39. 33. Knutson K. Sociodemographic and cultural determinants of sleep deficiency: implications for cardiometabolic disease risk. Soc Sci Med. 2013;79:7-15. 34. Saldatos CR, Allaert FA, Ohta T, Dikeos et al. How do individuals sleep around the world? Results from a single-day survey in ten countries. Sleep Med. 2005; 6(1):5-13. 35. LeBourgeois MK, Giannotti F, Cortesi F, et al. The relationship between reported sleep quality and sleep hygiene in Italian and American adolescents. Pediatrics. 2005;115(1 suppl):257-265. 36. Morselli L, Leprout R, Balbo M, Spiegel K. Role of sleep duration in the regulation of glucose metabolism and appetite. Best Pract Res Clin Endocrinol Metab. 2010;24(5):687-702. 37. Reimao R, Souza JC, Gaudioso CE, et al. Siestas among Brazilian Native Terena adults: a study of daytime napping. Arq Neuropsiquiatr. 2000;58(1)39- 44. 38. National Sleep Foundation. 2005 Sleep in America poll. Washington, DC; 2005. Available at http://www/sleepfoundation.org. 39. Valencia-Flores M, Castano VA, Campos RM, et al. The siesta culture concept is not supported by the sleep habits of urban Mexican students. J Sleep Res. 1998;7(1):21-9. 40. Paraskakis E, Ntouros T, Ntokos M, et al. Siesta and sleep patterns in a sample of adolescents in Greece. Pediatr Int. 2008;50(5):690-693. 41. Crespo CJ, Smith E, Carter-Pokras O, et al. Acculturation and leisure-time physical inactivity in Mexican Americans adults: results from NHANES III, 1988-1994. Am J Public Health. 2001;91:1254-1257. 42. Whinnery J, Jackson N, Rattanaumpawan P, et al. Short and long sleep duration associated with race/ethnicity, sociodemographics, an socioeconomic position. Sleep 2014;37(3):601-11. 43. Ebin VJ, Sneed CD, Morisky DE, et al. Acculturation and interrelationship between problem health-promoting behaviours among Latino adolescents. J Adolesc Health. 2001;(1):62-72. 44. Hale L, Do DP: Racial differences in self-reported sleep duration in a population-based study, Sleep. 2007;30(9):1096-103. 45. Hale L. Who has time to sleep? J Public Health (Oxf).2005;27(2):205-11. 46. Ertel KA, Berkman LF, Buxton OM. Socioeconomic Status, Occupational Characteristics, and Sleep Duration in African/Caribbean Immigrants and US White Health Care Workers. Sleep. 2011;34(4): 509-518.

50

Anujuo_binnenwerk_FINAL.indd 50 30/07/2018 21:19 Ethnic differences in sleep patterns

2

51

Anujuo_binnenwerk_FINAL.indd 51 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 52 30/07/2018 21:19 CHAPTER 3

ETHNIC DIFFERENCES IN SLEEP DURATION AT 5 YEARS, AND ITS RELATIONSHIP WITH OVERWEIGHT AND BLOOD PRESSURE

Published

Kenneth O. Anujuo, Tanja G.M. Vrijkotte, Karien Stronks, Girardin Jean-Louis, Charles Agyemang. Ethnic differences in sleep duration at 5 years, and its relationship with overweight and blood pressure. Eur J Public Health. 2016;26:1001-1006.

Anujuo_binnenwerk_FINAL.indd 53 30/07/2018 21:19 Chapter 3

ABSTRACT

Background: Studies on adult population indicate shorter sleep duration in ethnic minority groups than host populations. We examined ethnic differences in sleep duration and its relationship with overweight and blood pressure (BP) among children living in Amsterdam.

Methods: Participants include 2384 children (aged 5 years) and their mothers from the Amsterdam-based longitudinal study. Sleep was categorised into short sleep (<10hours/night) and normal sleep (10-11hours/night). Linear regressions (β) were used to study association between sleep duration and systolic BP (SBP) and diastolic BP (DBP). Prevalence ratios (PRs) were used to study ethnic differences in sleep duration and its association with overweight and raised BP.

Results: Minority groups reported shorter sleep duration compared to native Dutch, with prevalence ranging from 11.3% in Dutch to 53.1% in Ghanaians. Age-adjusted PRs ranged from 3.38 (95%CI 2.63-4.34) in Moroccans to 4.78 (95%CI 3.36-6.82) in Ghanaian compared with Dutch children. Increased prevalence of overweight was observed among children with short sleep in Dutch and Moroccans only, but this risk was no longer statistically significant after further adjustment for socioeconomic status. Short sleep was not related to SBP and DBP in all groups. No relationship was observed between short sleep and raised BP except for African Surinamese (3.65, 95% CI 1.23-10.8).

Conclusion: Like adults, children from ethnic minority populations sleep less hours than Dutch children. Efforts to improve ethnic inequalities in sleep hygiene should also include children at younger age. Associations as reported in adults with overweight and BP could not consistently be replicated in children, however.

54

Anujuo_binnenwerk_FINAL.indd 54 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

INTRODUCTION

A wide range of studies on sleep duration among adults from various ethnic minority groups have been conducted in many countries with results showing that ethnic minority groups have short sleep duration compared to their host populations.1-3 Studies on sleep duration among children of various ethnic groups show mixed results, however. For instance, it has been reported that ethnic minority children (African American, Asian, Native American, Hispanic and biracial) sleep less than nonminority children (White or European American),4 whereas, Biggs et al5 reported there was no significant difference in sleep duration between Southeast Asian and Europeans in a study conducted in Australia. Another cross-country comparative study,6,7 alongside other studies, indicated that children from Asia have shorter sleep 8-15 3 duration than their American and European counterparts. A recent study in Netherlands showed that migrant children sleep less than their Dutch counterparts.16 Unfortunately, this study did not clearly take into account the children with African background (such as African Surinamese and Ghanaians), and did not explore the association of sleep with blood pressure among the ethnic groups probably due to lack of data.

Sleep duration may contribute to increase in the prevalence of chronic health conditions in both children and adult populations.17-19 Previous studies have shown that short sleep duration was independently associated with increased overweight/obesity in children and adolescents.20-23 A recent study also indicated that short sleep duration was associated with overweight and obesity in migrant children, compared with European Dutch.24 However, this study did not clearly explore observations among children with African background such as the African Surinamese and Ghanaians. Studies in children have demonstrated that sleep duration was related to blood pressure (BP), but these studies also show inconsistent results, for example, whereas some studies reported that short sleep duration was related to high B P, 25 another study found that short and long sleep durations were related with high BP.19 However, one study found no significance difference in the relationship between sleep duration and BP.26

The purpose of the present study was to examine self-reported sleep duration among 5 years old children in a multi-ethnic population in the Netherlands, using the Amsterdam Born Child and their Development study (ABCD), a Dutch-population based study. In addition, we assessed the relationship between sleep duration and overweight, and BP among these children.

55

Anujuo_binnenwerk_FINAL.indd 55 30/07/2018 21:19 Chapter 3

STUDY POPULATION AND METHODS

Study population

The main goal of the ABCD study is to examine and determine factors in early life (during pregnancy and infancy) that might explain the later health of the child with specific attention paid to ethnic inequalities. Approval was obtained from the Academic Medical Center Medical Ethical Committee, the Vrije Universiteit Medical Center Medical Committee and the Registration Committee of Amsterdam. All participating mothers gave written informed consent for themselves and their children.

The design and rationale of the ABCD study have been described previously.27 In brief, between January 2003 and March 2004, 8,266 pregnant women were included in the study after their first antenatal visit to an obstetric caregiver (phase 1). Of these respondents, 7,863 women gave birth to a viable singleton infant and 6,575 women gave permission to collect information obtained from the Youth Health Care (phase 2). Phase 3 of the study started in the summer of 2008. Around two weeks after their ABCD-child’s fifth birthday, 6,161 mothers who initially gave permission for follow-up (93.7% of 6,575) were sent a questionnaire (in Dutch, English or Turkish) in which they were also asked for permission on participation of their child in the physical examination. The questionnaire, returned by 4,488 mothers, provided information on their child’s health, development and behaviour. Various physical measurements, including height, bodyweight, and blood pressure (BP), were made in 3,321 children. Reasons for lack of follow-up included withdrawal from the study, infant or maternal death, and loss to follow-up due to unknown address or emigration.

This study includes 2986 participants who completed the questionnaire, including singleton children (aged 5 years) from the main ethnic groups living in the Netherlands: Dutch, Turkish, Moroccan, African Surinamese and Ghanaian, in which the 5-year questionnaire and physical examination were completed. Ethnicity of the child was based on the country of birth of the child’s mother and her mother (self-reported). Because of heterogeneity among the Western and non-Western minority groups, they were excluded from the study (n=602), leaving a total of 2384 participants (2025 Dutch, 142 Moroccans, 92 Turkish, 93 African Surinamese and 32 Ghanaians).

56

Anujuo_binnenwerk_FINAL.indd 56 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

Measurements

Sleep The mothers were asked to estimate their children`s average night sleep duration on school days and during weekends. Sleep duration was assessed using the item: `How many hours does the child sleep on average per night on school days and during weekends?` Although sleep duration did not differ significantly between school days or during the weekend, we calculated the average sleep duration by taking 5 times the average school day sleep duration plus 2 times the weekend sleep duration and then divided by seven. Sleep duration was categorised using the most recent sleep guide from National Sleep Foundation.28 For children 3-5 years, 10-11 hour sleep per night is recommended as normal. Short sleep was defined as having less 3 than 10 hours of sleep per night.

BP and overweight Blood pressure (systolic and diastolic) was measured with Omron 705 IT (Omron Healthcare Inc, Bannockburn, IL, USA) with its appropriate cuff size (arm circumference 17–22 cm). First, a test measure was made (to comfort/relax) the child followed by a 10- min rest period. Then, BP was measured twice on the right arm in sitting position, with the arm supported at heart level. These two measurements were considered valid if they did not differ by more than 10 mmHg, otherwise a third measurement was made. Systolic BP (SBP) and diastolic BP (DBP) were calculated by taking the mean value of the two valid measures. Raised BP (hypertension and prehypertension: SBP or DBP>90th percentile or both) was based on sex, age, and height specific-percentiles.29 To calculate body mass index (BMI), height was measured to the nearest millimetre using a Leicester portable height measure (Seca), and weight to the nearest 100 gram using a Marsden weighing scale, (model MS-4102). Overweight was defined based on pooled international data for body mass index and linked to the widely used adult obesity cut off of 30kg/m2 as proposed by Cole et al, (2000).30

Information on educational level of the mothers - years of education after primary school (continuous) - was obtained from the pregnancy questionnaire. Sedentary time (computer and television) hours per day was obtained from the 5-year questionnaire.

57

Anujuo_binnenwerk_FINAL.indd 57 30/07/2018 21:19 Chapter 3

Data analysis

Baseline data were calculated in percentages and means with their 95% confidence interval. Chi square test and ANOVA were used to calculate differences in categorical and continuous variables. Comparisons of differences in sleep duration among ethnic groups and the association between sleep duration and overweight and raised BP were made using prevalence ratios (PRs). Linear regressions (β) were used to examine the association between sleep duration and BP respectively, adjusting for potential confounders known to influence sleep duration, overweight and BP including age, gender, SES (education) and height.19 All analyses were performed using STATA 11.0 (Stata Corp, Station, TX). A p-value of <0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

Table 1 shows the characteristics of the study population by ethnicity. Dutch children were on average, younger, weigh more at birth and have lower BMI, SBP and DBP, and were less frequently overweight than the children of the ethnic minority groups. Turkish and Moroccan mothers were youngest, and Turkish had the lowest level of education compared to the other ethnic groups. Ethnic minority groups spent more hours on sedentary activities compared to the Dutch.

Sleep duration

All ethnic minority groups had significantly lower mean sleep duration compared to the Dutch, with the Ghanaians having the lowest mean sleep duration (9.8 hours) compared with (10.7 hours) in Dutch (Table 1).

Short sleep

The prevalence of short sleep was lower in the Dutch compared to all ethnic minority groups. Table 2 shows that these ethnic differences were significant after adjusting for age, gender, and education with PRs ranging from 2.37

58

Anujuo_binnenwerk_FINAL.indd 58 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

(95% CI, 1.68-3.35) in Turkish to 3.56 (95% CI, 2.44-5.19) in Ghanaian compared with Dutch children.

Association between sleep duration and blood pressure, and overweight

There was no significant association between sleep duration and SBP, and DBP in any of the ethnic groups after adjusting for age, gender, height and education (Table 3). Figure 2 shows the association between sleep duration and raised BP among ethnic groups. We observed a significant association between sleep duration and raised BP, but only in African Surinamese. The association remained significant after adjusting for SES (education) 3 (Table 3). The prevalence of raised BP was lower in the African Surinamese normal sleepers than the short sleepers. In the unadjusted model, there was a significant association between sleep duration and overweight in the Dutch and Moroccans. The PRs slightly decreased after further adjustment for socioeconomic status [SES (education)], and become statistically non- significant (Table 3).

Although the focus of our study was on the various ethnic groups, however, when all ethnic minority groups were combined as one group, the results of the analysis essentially remain the same with ethnic minority children having a higher odds of shorter sleep duration than native Dutch even after adjustment for potential confounders (PR 2.64, 2.08-3.35). There was no significant relationship between sleep duration and mean BP and raised BP, and overweight in all groups.

59

Anujuo_binnenwerk_FINAL.indd 59 30/07/2018 21:19 Chapter 3 n = 32 5.9 (5.7-6.1) 120.3 (118.3-122.4) 3.4 (3.1-3.5) 24.4 (22.6-26.3) 9.8 (9.45-10.1) 53.1 (35.5-70.7) 16.7 (15.8-17.6) 31.3 (14.9-47.6) 104.3 (101.3-107.3) 63.1 (60.8-65.4) 12.5 (1.00-24.2) 25.0 (9.75-40.3) 2.7 (2.2-3.2) 62.5 (45.5-79.6) 34.4 (17.7-51.1) 3.13 (3.00-9.25) 29.2 (27.3-31.1) 5.1 (3.9-6.2) Ghanaian n = 93 5.8 (5.7-5.9) (117.6-120.0) 118.8 3.3 (3.2-3.4) 22.6 (21.8-23.4) 10.0 (9.78-10.3) 37.6 (27.7-47.5) 15.9 (15.6-16.3) 18.3 (10.4-26.2) 100.0 (98.7-101.4) 59.7 (58.2-61.1) 5.56 (1.00-10.3) 13.3 (6.27-20.4) 2.6 (2.3-2.9) 60.2 (50.2-70.2) 26.9 (17.8-35.9) 12.9 (6.05-19.8) 29.6 (28.3-30.9) 7.9 (7.2-8.6) African-Surinamese n = 142 5.9 (5.9-6.0) (116.2-118.3) 117.3 3.3 (3.3-3.4) 22.7 (22.1-23.4) 10.1 (9.87-10.3) 37.3 (29.3-45.3) 16.4 (16.1-16.7) 24.7 (17.5-31.8) 100.9 (99.5-102.5) 59.3 (57.9-60.6) 12.9 (7.35-18.6) 18.0 (11.6-24.4) 2.4 (2.3-2.6) 59.2 (51.0-67.3) 35.2 (27.3-43.1) 5.63 (1.83-9.44) 27.8 (26.9-28.6) 5.4 (4.8-6.1) Moroccan n = 92 5.8 (5.7-5.9) (115.3-117.2) 116.2 3.4 (3.3-3.4) 22.2 (21.4-22.9) 10.0 (9.82-10.2) 38.0 (28.1-48.0) 16.3 (15.9-16.8) 25.00 (16.1-33.9) 101.5 (99.7-103.2) 58.8 (57.4-60.2) (4.69-18.0) 11.4 16.9 (9.03-24.7) 2.4 (2.1-2.6) 52.2 (41.9-62.4) 40.2 (30.1-50.3) 7.61 (2.16-13.1) 27.2 (25.9-28.5) 4.6 (3.9-5.3) Turkish n = 2025 5.7 (5.6-5.7) (116.3-116.8) 116.5 3.5 (3.4-3.5) 20.9 (20.8-21.0) 10.7 (10.6-10.7) (9.93-12.7) 11.3 15.3 (15.3-15.4) 6.37 (5.31-7.43) 98.9 (98.6-99.2) 56.6 (56.3-56.8) 5.46 (4.46-6.46) 10.9 (9.59-12.4) 1.2 (1.1-1.2) 12.7 (11.2-14.1) 44.5 (42.4-46.7) 42.8 (40.6-44.9) 32.8 (32.6-32.9) 10.7 (10.6-10.9) Dutch Characteristics of the study population by ethnicity >2 hours/day 1-2 hour/day <1 hour/day Child Age (years) Height (cm) Birth weight (kg) Child weight (kg) Mean sleep duration (hrs) Short sleep <10 hrs/night BMI (kg/m2) Overweight (kg/m2) mm/Hg Systolic BP m/Hg Diastolic BP Hypertension (%) (%) Raised BP Sedentary time (hours) Sedendary categories Maternal age (yrs) Education (yrs) BMI: body mass index; BP: Blood pressure, Data are presented as mean and percentages with 95% confidence interval (CI) Table 1: Table

60

Anujuo_binnenwerk_FINAL.indd 60 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers : p < 0.001, **: * 3.56 (2.44-5.19)* 4.81 (3.37-6.85)* PR (95%CI) 4.78 (3.36-6.82)* Ghanaian 4.69 (3.32-6.65)* n = 32 3 2.79 (2.05-3.81)* 3.38 (2.53-4.52)* PR (95%CI) 3.37 (2.52-4.51)* African Surinamese 3.33 (2.49-4.44)* n = 93 2.47 (1.82-3.34)* 3.37 (2.62-4.33)* PR (95%CI) 3.38 (2.63-4.34)* Moroccan 3.30 (2.58-4.22)* n = 142

2.37 (1.68-3.35)* 3.39 (2.54-4.55)* 3.40 (2.55-4.54)* PR (95%CI) Turkey 3.36 (2.52-4.49)* n = 92 1.00 1.00 1.00 PR (95%CI) 1.00 Dutch n = 2025 A Prevalence ratio(s) for short sleep among children at 5 years compared with the Dutch. : Model 1: Adjusted for age, model 2: adjusted for age and gender, model 3: adjusted for age, gender and education. . ( Adjusted for age, model 2: adjusted age and gender, : Model 1: Model 3 Model 2 Model 1 Short sleep (<10 h/ night) Crude PR: prevalence ratio(s); *: p-value (p) A 0.05) Table 2: Table

61

Anujuo_binnenwerk_FINAL.indd 61 30/07/2018 21:19 Chapter 3 (95%CI) (95%CI) Ghanaian β n = 32 0.889 (-5.822, 7.600) -0.100 (-7.186, 6.985) 1.665 (-3.465, 6.795) 0.360 (-4.597, 5.318) PR 0.88 (0.26-2.99) 0.63 (0.17-2.35) (95%CI) (95%CI) African Surinamese β n = 93 1.801 (-0.985, 4.588) 1.956 (-0.874, 4.786) 0.619, (-2.301, 3.540) 0.772 (-2.210, 3.754) PR 3.29 (1.07-10.2)** 3.65 (1.23-10.8)** (95%CI) (95%CI) Moroccan β n = 142 0.736 (-2.439, 3.911) 0.429 (-2.776, 3.633) -1.505 (-4.188, 1.179) -1.593 (-4.313, 1.128) PR 0.97 (0.46-2.04) 0.98 (0.46-2.09) (95%CI) (95%CI) Turkey β n = 92 -0.395 (-4.036, 3.247) -0.348 (-4.083, 3.388) -0.913 (-3.894, 2.068) -0.813 (-3.870, 2.244) PR 0.89 (0.33-2.39) 0.90 (0.33-2.48) (95%CI) (95%CI) Dutch β n = 2025 0.162 (-0.808, 1.132) 0.074 (-0.897, 1.045) -0.160 (-0.947, 0.626) -0.292 (-1.077, 0.494) PR 0.88 (0.58-1.34) 0.87 (0.57-1.32) Relationship between sleep duration and SBP and DBP, raised BP and overweight among children at 5 years raised BP and DBP, Relationship between sleep duration and SBP Systolic BP Short sleep (<10 h/ night) Model 1 Model 2 Diastolic BP Short sleep (<10 h/ night) Model 1 Model 2 Raised BP Short sleep (<10 h/ night) Unadjusted Model 1 Table 3: Table

62

Anujuo_binnenwerk_FINAL.indd 62 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers (95%CI) Ghanaian β n = 32 0.59 (0.20-1.72) 0.47 (0.14-1.98)

(95%CI)

African Surinamese β n = 93 1.86 (0.79-4.40) 1.89 (0.79-4.53) 3

(95%CI) Moroccan β n = 142 1.78 (1.01-3.15)** 1.69 (0.95-3.05) adjusted for age, gender, height, and education. adjusted for age, gender, Model 2: (95%CI) Turkey β n = 92 0.71 (0.33-1.56) 0.72 (0.31-1.66) : adjusted for education. **: p < 0.05 Model 1 adjusted for age and gender, (95%CI) Dutch β n = 2025 1.61 (1.04-2.49)** 1.49 (0.97-2.29) Model 1: Overweight Short sleep (<10 h/ night) Crude Model 1 β: regression coefficient, PR: prevalence ratio(s), CI: confidence interval, *: p-value (p) and DBP, For SBP and overweight, For raised BP Table 3 - continued Table

63

Anujuo_binnenwerk_FINAL.indd 63 30/07/2018 21:19 Chapter 3

Fig. 1: Association between sleep duration and overweight among children at 5 years in Amsterdam

Fig. 2: Association between sleep duration and raised BP among children at 5 years in Amsterdam

64

Anujuo_binnenwerk_FINAL.indd 64 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

DISCUSSION

Key findings

The current study examined ethnic differences in sleep duration and its relationship with overweight and BP among children from various ethnic groups living in Amsterdam. The result of our study showed that ethnic minority children were more likely than Dutch children to report short sleep. With a few exceptions, short sleep was not associated with overweight or blood pressure in any of the ethnic groups.

Discussion of key findings 3 The result of our study demonstrated that ethnic minority children sleep less than their Dutch counterparts. Our finding is consistent with a recent study in Netherlands that showed that migrant children sleep less than their Dutch counterparts.16 Our finding is also in line with other previous studies, which showed that ethnic minority children from African American, Asian, Native American, Hispanic and biracial origins sleep less than White American children.4,8-15,31 The reason for the higher prevalence of short sleep observed in ethnic minority children is less clear, but may be due to socio-cultural and environmental factors that may differ among various ethnic groups. Culture, beliefs and attitudes of adult parents towards sleep may be transferred to their children and this can affect attitudes and practices surrounding their sleep behavior.31,32 Factors such as bedtime routines, co-habitation (number of older children sleeping in the same room) and sedentary activities before bedtime may influence ethnic differences in sleep duration in children. For instance, previous study has shown that communal sleeping habits are more common among African Surinamese, and higher rates of bed sharing has been reported among Turkish and Moroccans.33,34 Previous studies indicated that increased sedentary activities (TV and computer screen time) predict short sleep duration.35 Our data show that a higher proportion of the ethnic minority groups spent more than 2 hours on sedentary activities and this may contribute to their higher prevalence of short sleep compared to the Dutch. In addition, other studies suggest that ethnic minority parents were more likely to be short sleepers because they usually do shift work, and sometimes, multiple jobs, with little time left to sleep.1,8 It is likely that parents were not able to enforce early bedtime schedules in their children because they arrive late from work. On the other hand, in a Dutch-based

65

Anujuo_binnenwerk_FINAL.indd 65 30/07/2018 21:19 Chapter 3

study that compares sleep habits of children from the USA and China, Dutch children had lower scores for bedtime resistance, sleep anxiety, night awakening and sleep disordered breathing.7 This indicate a favourable sleep patterns, which may differ from ethnic minority children, and may explain the observed longer sleep duration in the Dutch. Another possible explanation for the observed longer sleep duration in the Dutch can be attributed to the Dutch tradition of the three “R`s” concept: Rust (rest), Reinheid (cleanliness), and Regelmaat (regularity).36 These basic concepts lay much emphasis on rest and regularity and seem to drive a healthy sleep pattern in young Dutch children.

Short sleep has been shown to be related to overweight in children.20-24 The potential mechanisms linking short sleep and overweight has been proposed in several studies including energy balance,37 time available to eat/timing of meals,38 and increased intake of high-calorie foods,39 which could eventually induce overweight. In our present study, short sleep was associated with overweight, but only in Dutch and Moroccan children. The differences in our study, however, were mainly due to SES differences suggesting SES may influence the aforementioned mechanisms. By contrast, we observed no associations in Turkish, African Surinamese and Ghanaian children. The lack of association in other ethnic groups is unclear, but might be due to inadequate sample size to attain statistical power. Further studies should explore possible factors that may be responsible for lack of association between short sleep and overweight in ethnic minority groups.

With respect to potential relationship between sleep duration and blood pressure, in general, we did not find a significant association between sleep duration and blood pressure in any of the ethnic groups. Our finding is consistent with a previous study which showed no significant association between sleep duration and blood pressure in Gujarati Indian adolescents.26 On the other hand, our finding is in contrast with a previous study which focused on children aged 6-11 years.25 Our finding also contrasts well with another Japanese study, which demonstrated that short and long sleep duration were associated with higher levels of blood pressure.19 Though our study population`s age was similar to the Japanese study, demographic, environmental, genetic and other unmeasured factors may account for the differences in obtained results in both studies. The reason for a lack of association between sleep duration and BP in our study is not fully understood. It may be due to the fact that we observed no association between sleep duration and overweight. It may also be due to differences in

66

Anujuo_binnenwerk_FINAL.indd 66 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

study population (previous study focused on older children (6-11 years),25 whereas the children in our study were younger (5 years), probably it may be too early for the effect of high BMI and short sleep duration to clearly manifest on BP.

The results of our study need to be interpreted in light of a few limitations. Information on sleep duration was based on self-report, and therefore prone to reporting bias; hence parents may have underestimated or overestimated sleep duration. Also the measurement of BP in children is often difficult and may give unexpected false readings as children often find it difficult to relax during measurement. Due to small numbers, we were not able to stratify the analysis by gender. 3 In conclusion, like adults, also children from ethnic minority populations sleep less hours than children from Dutch origin. Measures to improve ethnic inequalities in sleep hygiene should therefore also include children at younger age. Associations as reported in adults with overweight and BP could not consistently be replicated in these young children, however.

67

Anujuo_binnenwerk_FINAL.indd 67 30/07/2018 21:19 Chapter 3

REFERENCES

1. Anujuo K, Stronks K, Snijder MB, Jean-Lous G, Ogedegbe G, Agyemang C. Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS study. Sleep Med 2014;15:1115-21. 2. Kripke DF, Brunner R, Freeman R, et al. Sleep Complaints of Postmenopausal Women. Clin J Womens Health 2001;1:244-252. 3. Park A. “How Well Do You Sleep? The Answer May Depend on Your Race”, Time Time Inc., 2012. 4. Spilsbury JC, Storfer-Isser A, Drotar D. et al. Sleep behavior in an urban U.S. sample of school-aged children. Arch Pediatr Adolesc Med 2004;158:988–994. 5. Biggs SN, Pizzorno VA, van den Heuvel CJ, Kennedy JD, Martin AJ, Lushington K. Differences in parental attitudes towards sleep and associations with sleep-wake patterns in Caucasian and Southeast Asian school-aged children in Australia. Behav Sleep Med 2010;8:207-18. 6. Liu X, Liu L, Owens JA, Kaplan DL. Sleep patterns and sleep problems among schoolchildren in the United States and China. Pediatrics 2005;115:241–249. 7. van Litsenburg RR, Waumanns RC, van den Berg G, Gemke RJ. Sleep habits and sleep disturbances in Dutch children: a population-based study. Eur J Pediatr 2010;169:1009-15. 8. Ng DK, Kwok KL, Cheung JM, et al. Prevalence of sleep problems in Hong Kong primary school children: A community-based telephone survey. Chest 2005;128:1315–1323. 9. Shinkoda H, Matsumoto K, Park YM, Nagashima H. Sleep–wake habits of schoolchildren according to grade. Psychiatry Clin Neurosci 2000;54:287–289. 10. Yang CK, Kim JK, Patel SR, Lee JH. Age-related changes in sleep/wake patterns among Korean teenagers. Pediatrics 2005;115:250–256. 11. Adam EK, Snell EK, Pendry P. Sleep timing and quantity in ecological and family context: A nationally representative time-diary study. J Fam Psychol 2007;21:4–19. 12. Gulliford MC, Price CE, Rona RJ, Chinn S. Sleep habits and height at ages 5 to 11. Arch Dis Child 1990;65:119–122. 13. Iglowstein I, Jenni OG, Molinari L, Largo RH. Sleep duration from infancy to adolescence: Reference values and generational trends. Pediatrics 2003;111:302–307. 14. Spruyt K, O’Brien LM, Cluydts R, Verleye GB, Ferri R. Odds, prevalence, and predictors of sleep problems in school-age normal children. J Sleep Research 2005;14:163–176. 15. Thorleifsdottir B, Bjornsson JK, Benediktsdottir B, Gislason T, Kristbjarnarson H. Sleep and sleep habits from childhood to young adulthood over a 10-year

68

Anujuo_binnenwerk_FINAL.indd 68 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

period. J Psychosom Res 2002;53:529–537. 16. Labree LJ, van de Mheen HD, Rutten FF, Rodenburg GG, Koopmans GT, Foets MM. Sleep duration differences between children of migrant and native origins. Z Gesundh Wiss., 2015, 23:149-156. 17. Cappuccio FP, Stranges S, Kandala NB, et al. Gender-specific associations of short sleep duration with prevalent and incident hypertension. The Whitehall II study. Hypertension 2007;50:694–701. 18. Cappucio FP, Taggart FM, Kandala NB, et al. Meta-analysis of short sleep duration and obesity in children and adults. Sleep 2008;31:619-26. 19. Sampei M, Dakeishi M, Wood DC, Murata K. Impact of total sleep duration on blood pressure in preschool children. Biomed Res 2006;27:111-5. 20. Beccuti G, Pannain S. Sleep and obesity. Curr Opin Clin Nutr Metab Care 2011;14:402-12. 3 21. Knutson KL. Does inadequate sleep play a role in vulnerability to obesity? Am J Hum Biol., 2012, 24:361-7. 22. Liu J, Zhang A, Li L. Sleep duration and overweight/obesity in children: review and implications for pediatric nursing. J Spec Pediatr Nurs 2012;17:193-204. 23. Rutters F, Gerver WJ, Nieuwenhuizen AG, Verhoef SP, Westerterp-Plantenga MS. Sleep duration and body-weight development during puberty in a Dutch children cohort. Int J Obes 2010;34:1508-14. 24. Labree W, van de Mheen D, Rutten F, Rodenburg G, Koopmans G, Foets M. Differences in overweight and obesity among children from migrant and native origin: the role of physical activity, dietary intake and sleep duration. PLoS One 2015, 10(6):e0123672. 25. Archbold KH, Vasquez MM, Goodwin JL, Quan SF. Effects of sleep patterns and obesity in blood pressure in a 5-year period: report from the Tucson Children`s Assessment of Sleep Apnea Study. J Pediatr 2012;161:26-30. 26. Shaikh WA, Patel M, Singh S. Association of Sleep Duration with Arterial Blood Pressure Profile of Gujarati Indian Adolescents.Indian J Community Med 2010;35:125-129. 27. van Eijsden M, Vrijkotte TGM, Gemke RJ, van der Wal MF. Cohort Profile: The Amsterdam Born Children and their Development (ABCD) Study.Int J Epidemiol 2011;40:1176–86. 28. “Sleep time recommendations: What`s changed?” Retrieved August 22, 2015 from http://sleepfoundation.org/how-sleep-works/how-much-sleep-do-we- really-need/page/0/1 29. National High Blood Pressure Education Programme Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation and treatment of high blood pressure in children and adolescents. Paediatrics 2004;114:555-576.

69

Anujuo_binnenwerk_FINAL.indd 69 30/07/2018 21:19 Chapter 3

30. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000;320:1240–1243. 31. Crabtree VM, Korhonen JB, Montgomery-Downs HE, Jones VF, O’Brien LM, Gozal D. Cultural influences on the bedtime behaviours of young children. Sleep Med 2005;6:319-324. 32. Sadeh A, Raviv A, Gruber R. Sleep patterns and sleep disruptions in school children. Dev Psychol 2000;36:291-301. 33. CultureGram 2012 World Editions – Republic of Surinamese. http://www. culturegrams.com/images/samples/SurinameWorld.pdf assessed on March 30 2015. 34. Luijk MP, Mileva-Seitz VR, Jansen PW, et al. Ethnic differences in prevalence and determinants of mother-child bed-sharing in early childhood. Sleep Med 2013;14:1092-99. 35. McDonald L, Wardle J, Llewellyn CH, van Jaarsveld CHM, Fisher A. Predictors of shorter sleep in early childhood. Sleep Med 2004;15:536-540. 36. Super CM, Harkness S, Tijen N, et al. The three R’s of Dutch childrearing and the socialization of infant arousal. In: Harkness S, Super CM et al, editors. Parents’ cultural belief systems: their origins, expressions, and consequences. New York: Guilford, 1996:558. 37. Hasler G, Buysse DJ, Klaghofer R, et al. The association between short sleep duration and obesity in young adults: a 13-year prospective study. Sleep 2004;27:661-6. 38. Zimberg IZ, Dâmaso A, Del Re M, et al. Short sleep duration and obesity: mechanisms and furture perspectives. Cell Biochem Funct 2012;30:524-9. 39. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med 2004;141:846-50.

70

Anujuo_binnenwerk_FINAL.indd 70 30/07/2018 21:19 Sleep duration, relationship with overweight and BP among preschoolers

3

71

Anujuo_binnenwerk_FINAL.indd 71 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 72 30/07/2018 21:19 CHAPTER 4

RELATIONSHIP BETWEEN SHORT SLEEP DURATION AND CARDIOVASCULAR RISK FACTORS IN A MULTI-ETHNIC COHORT – THE HELIUS STUDY

Published

Kenneth Anujuo, Karien Stronks, Marieke B. Snijder, Girardin Jean-Louis, Femke Rutters, Bert-Jan van den Born, Ron J. Peters, Charles Agyemang. Relationship between short sleep duration and cardiovascular risk factors in a multi-ethnic cohort – the HELIUS study. Sleep Med. 2015;16:1482-1488.

Anujuo_binnenwerk_FINAL.indd 73 30/07/2018 21:19 Chapter 4

ABSTRACT

Objective: To investigate the association between short sleep duration and CVD risk factors including hypertension, diabetes, obesity, and lipid profile among various ethnic groups (South-Asian Surinamese, African Surinamese, Ghanaians, Turks, Moroccans and Dutch) living in the Netherlands. We also examined the contribution of social economic status (SES) and lifestyle factors to this association.

Method: 12805 participants (aged 18-70 years) from the multi-ethnic HELIUS cohort. Short sleep duration was defined as <7 hours/night. The association between short sleep and CVD risk factors, and the contribution of SES and lifestyle factors were assessed using prevalence ratios (PRs).

Results: Short sleep was significantly associated with obesity in four out of six ethnic groups, with the socio-demographic adjusted PR of 1.45 (95% CI, 1.07-1.95) in Dutch, 1.21 (1.01-1.44) in South-Asian Surinamese, 1.25 (1.09-1.43) in African Surinamese and 1.16 (1.04-1.29) in Turks. Short sleep was significantly associated with diabetes in African Surinamese (1.45, 1.14-1.84), Turks (1.59, 1.26-2.02), and Moroccan (1.29, 1.02-1.63). By contrast, the associations between other cardiovascular risk factors and short sleep were not significant in most ethnic groups, with the exception of the association with hypertension in Dutch and Turks, and dyslipidaemia in South-Asians Surinamese (reduced HDL cholesterol and triglyceride) and Moroccans (raised total cholesterol). SES and lifestyle factors contributed little to the observed associations.

Conclusion: Our findings indicate that short sleep is associated with obesity and diabetes in most ethnic groups. The associations for other risk factors vary between ethnic groups. Further studies are needed to establish the potential factors that might lead to the observed differences across populations.

74

Anujuo_binnenwerk_FINAL.indd 74 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

INTRODUCTION

Cardiovascular disease (CVD) is the leading cause of global mortality. Several prospective and retrospective studies have shown that hypertension, diabetes, obesity and dyslipidaemia are important risk factors for CVD [1,2]. As the prevalence of these risk factors (especially obesity and diabetes) is increasing in adult populations in most countries [3], the necessity to identify potential modifiable risk factors increases, in order to reduce adverse CVD outcomes.

Recently, several studies have shown conflicting reports on the association between sleep duration and cardiovascular risk factors including hypertension [4,5], diabetes [6-9], obesity [10-12] and dyslipidaemia [13- 19]. The fact that two independent studies showed an association between short sleep and cardiovascular mortality is supportive of the hypothesis that short sleep is related to CVD risk factors [20,21]. Recent studies indicated that the relationship between sleep duration and CVD risk factors varies between study populations [22-23]. This suggests the potential importance 4 of contextual factors such as sociocultural and lifestyle factors which may influence the association. Previous studies suggested that the association between sleep duration and CVD risk may be mediated by SES, lifestyle factors and other covariates, they were therefore recommended to be included in future studies [24].

Studies investigating the relationship between sleep duration and cardiovascular risk factors have not considered differences among ethnic minority groups with different migration backgrounds. In addition, CVD and risk factors also differ between ethnic groups [25-29]. For instance, Surinamese people in Netherlands have higher incidence rates of stroke than the European Dutch. By contrast, Moroccans have lower incidence rates of stroke compared with Dutch [30]. Also, ethnic minority groups tend to experience shorter sleep duration compared to their host European majority populations [29]. Given the variation in the association between short sleep and CVD risk factors between populations, it is important to elucidate how short sleep is related to CVD risk factor in different ethnic groups. Furthermore, the data on the relationship between short sleep and CVD risk factors across ethnic minority groups are lacking particularly in Europe. Using data from the Healthy Life in an Urban Setting (HELIUS) study, the aim of this study was to investigate whether there are differences in the association of sleep duration with cardiovascular risk factors between

75

Anujuo_binnenwerk_FINAL.indd 75 30/07/2018 21:19 Chapter 4

various ethnic minority groups living in the Netherlands. In addition, the contribution of SES and lifestyle factors to these associations was examined.

STUDY POPULATION AND METHODS

Study population

This study was based on baseline data from the HELIUS study. The aims and design of the HELIUS study have been described elsewhere [30]. In brief, HELIUS is a large-scale prospective cohort study on health and health care among different ethnic groups living in Amsterdam. The study started in 2011 and it includes individuals aged between 18 and 70 years from the six major ethnic groups in Amsterdam (African-Surinamese, South-Asian Surinamese, Turks, Moroccan, Ghanaian, and Dutch origin). This study focuses on three major disease categories: cardiovascular disease, mental health and infectious diseases. Participants were randomly sampled from the municipal registers and stratified by ethnicity. The study protocols were approved by the Academic Medical Center (AMC) Ethical Review Board. All participants provided written informed consent.

This study used baseline data that were collected until June 2014. The study included 13,316 participants from whom data from both questionnaire and the physical examination were available. Participants with a Javanese Surinamese background (n=137), other/unknown Suriname background (n=141) or other/unknown background (n=26) were excluded because of the small sample sizes. In addition, individuals with no data on sleep duration (n=207) were also excluded from the analysis. This resulted in a dataset of 12,805 participants, including 2146 Dutch, 2158 African-Surinamese, 2262 South-Asian Surinamese, 1795 Ghanaians, 2242 Turks and 2202 Moroccans.

Measurements

Sleep duration Participants were asked to provide information on their average hours of sleep at night. Sleep duration was assessed using the item, `How many hours do you sleep on average per night?` Short sleep was defined as having <=6 hours of sleep per night according to National Sleep Foundation (NSF), American Academy of Sleep Medicine (AASM), and Sleep Research Society

76

Anujuo_binnenwerk_FINAL.indd 76 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

(SRS) which recommend 7 to 8 hours as the basal sleep need for healthy adults [31, 32]. We focused on short sleep only because in our previous study, we previously demonstrated that short sleep was the major problem for the ethnic minority groups [29], and because previous studies found that short sleep was more consistently related to CVD risk factors compared to long sleep [22-23], [33-35].

Ethnicity Participant’s ethnicity was defined according to the country of birth of the participant as well as that of their parents. Specifically, a participant is considered to be of ethnic minority origin if he/she fulfils either of the following criteria: 1) he or she was born abroad and has at least one of his/ her parents born abroad; or 2) he or she was born in the Netherlands but has both his/her parents born abroad [30]. The Surinamese population is made up of several ethnic subgroups including those of Africa Surinamese, South- Asian Surinamese, Javanese, Amerindian, and Chinese origin. Therefore, in Surinamese, self-identification was also used to further distinguish the subgroups of Surinamese origin. Surinamese are migrants from Suriname, 4 which is a former Dutch colony in South America. Surinamese people who have an African-Caribbean background are mostly of West African descent, while those with a South Asian background have North Indian origins. They all migrated from Suriname to the Netherlands due to the political instability, in Suriname in 1975 and 1980. Ghanaians migrated to the Netherlands between 1974 and 1983 predominantly for economic reasons, while others migrated due to drought, political instability and the deportation of Ghanaians from Nigeria in the early 1990s. The Turkish and Moroccans migrated to the Netherlands in the 1960s and early 1970s to fill labour shortages which were lacking in unskilled occupations. During 1970–1980, Turkish and Moroccans guest workers brought their spouses and children to the Netherlands.

Cardiovascular risk factors Weight was measured in light clothing with SECA877 to the nearest 0.1 kg. Height was measured without shoes with a portable stadiometer (SECA 217) to the nearest 0.1 cm. Blood pressure (BP) was measured using a validated automated digital BP device (Microlife WatchBP Home, Microlife AG, Heerbrugg, Switzerland) on the left arm in a seated position after the participant had seated for at least 5 minutes. All measurements were performed in duplicate; the mean of the two measurements was used in the analyses.

77

Anujuo_binnenwerk_FINAL.indd 77 30/07/2018 21:19 Chapter 4

Fasting blood samples were taken in the morning and used to determine the concentration of glucose by spectrophotometry, using hexokinase as primary enzyme (Roche Diagnostics, Japan). Total cholesterol, triglycerides and high-density lipoprotein (HDL) cholesterol were determined by colorimetric spectrophotometry (Roche Diagnostics, Japan). Low-density lipoprotein cholesterol (LDL) was calculated according to the Friedewald formula [36].

Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2). Participants were considered obese if the BMI was > 30 kg/ m2. Hypertension was defined as systolic BP ≥ 140 mmHg, or diastolic BP ≥ 90 mmHg, or being on antihypertensive medication, or self-reported hypertension. Type 2 diabetes was defined as increased fasting glucose ≥ 7mmol/L or current use of medication prescribed to treat diabetes as recorded at the physical examination, or self-reported diabetes. Individual components of dyslipidemia were determined separately using the recommended cut off as follows: Total serum cholesterol (TC) > 6.22mmol/L, high density lipoprotein cholesterol (HDL-C) < 1.04mmol/L, low density lipoprotein cholesterol (LDL- C) > 4.14mmol/L, and triglyceride (TG), > 1.69mmol/L [37-38].

Other measurements The level of educational level was determined using participant’s highest level of education (either in the Netherlands or in the country of origin). Participants were categorized into those who have never been to school or had elementary schooling only (1st category), those with lower vocational schooling or lower secondary schooling (2nd category), those with intermediate vocational schooling or intermediate/higher secondary education schooling (3rd category), and those with higher vocational schooling or university (4th category). The first two categories were combined because of small numbers during this analysis.

Marital status included married/ registered/ partnership, living together, unmarried/ never married, divorced/separated, or widowed. Using a questionnaire, the alcohol intake in the past 12 months (yes/no) and smoking status (yes/no/ex-smoker) were obtained.

Habitual physical activity was measured using the Short Questionnaire to Assess Health-Enhacing Physical Activity (SQUASH) questionnaire [39]. The SQUASH questions about multiple activities refer to a normal week

78

Anujuo_binnenwerk_FINAL.indd 78 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

in the past months. Participants were categorized according to the Dutch guideline for physical activity by summing up the number of days per week for each moderate- and high- intensity activity lasting at least 30 minutes. A total of 5 days resulted in participants being categorized as achieving the Dutch norm for physical activity.

Data Analysis

Baseline data were expressed as percentages or means with 95% CI. Chi- square tests were used to test the differences in categorical variables between ethnic groups. Comparisons among ethnic groups were made using Prevalence Ratios (PRs) with sleep duration categories as independent variables and cardiovascular risk factors as outcomes, with adjustment for potential confounders (age, gender) and possible mediators (education, marital status, BMI, smoking, alcohol consumption, physical activity). For all outcome variables, interaction between ethnicity and sleep duration was tested. All analyses were performed using STATA 11.0 (Stata Corp, Station, 4 TX). A p-value of <0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

Table 1 shows the characteristics of the study population by ethnic group. Moroccans and Turks were younger, had lower educational levels, consume less alcohol, less often achieved the physical activity norm, and had lower prevalence of hypertension compared to Dutch and other ethnic origin groups. South-Asian Surinamese, African Surinamese, and Ghanaian participants had a lower mean sleep duration and higher prevalence of short sleep than Dutch, Turks and Moroccans. Dutch and South-Asian Surinamese were less obese than other ethnic groups. The prevalence of diabetes was higher in all ethnic minority groups while lipid profile was mostly favourable for the ethnic minority groups as compared with Dutch.

79

Anujuo_binnenwerk_FINAL.indd 79 30/07/2018 21:19 Chapter 4

Interaction between ethnicity and sleep duration

A statistically significant interaction was observed between ethnicity and sleep duration for obesity in South-Asian Surinamese (p= 0.033), Ghanaians (p= 0.002), Turks (p= 0.030) and Moroccan (p= 0.008); as well as for hypertension in Ghanaians (p= 0.049) No statistically significant interaction was found between sleep duration and other CVD risk factors by ethnic background.

Sleep duration and obesity

Figure 1 (first part) shows the relationship between sleep duration and the prevalence of obesity by ethnic group. The prevalence of obesity was consistently higher in those reporting short sleep duration, with this association being statistically significant in people of Dutch origin, and South-Asian Surinamese, African Surinamese, and Turks. The significant association persisted after adjustment for covariates.

Sleep duration and diabetes

Also the prevalence of diabetes was consistently higher in those reporting short sleep duration (middle part of Figure 1). Table 2 shows that short sleep was significantly related to diabetes in three ethnic minority groups: African Surinamese, Turks, and Moroccan participants, after adjustment for age and sex. The significant associations persisted after adjustment for all covariates.

Sleep duration and hypertension

Figure 1 (last part) shows the relationship between sleep duration and prevalence of hypertension by ethnic group. It shows that the prevalence of hypertension was consistently higher in those reporting short sleep duration. After adjustment for age and gender, there was a significant positive relationship between short sleep and hypertension in Dutch and Turks only (Table 2). After further adjustment for education, marital status, BMI, smoking, alcohol consumption and physical activity, the observed significant association in Dutch disappeared but remained in Turks.

80

Anujuo_binnenwerk_FINAL.indd 80 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

Sleep duration and dyslipidaemia

Figures 2 show the relationship between sleep duration and lipid levels by ethnic group. In general, lipid levels did not vary much by sleep duration. A significant positive association of short sleep with high total cholesterol in Moroccans was observed, with low HDL cholesterol in South-Asian Surinamese, and with high triglycerides in South-Asian Surinamese, after adjusting for age and gender (Table 2). These observed associations persisted after adjustment for other factors.

4

81

Anujuo_binnenwerk_FINAL.indd 81 30/07/2018 21:19 Chapter 4 n = 2202 Moroccan 50.4 (48.3-52.5) 16.2 (14.8-17.9) 7.2 (7.1, 7.2) 33.3 (31.3-35.3) 26.9 (25.1-28.8) 40.2 (39.6-40.7) 24.9 (23.1-26.7) 11.4 (10.1-12.7) 11.4 30.1 (28.2-32.0) 5.05 (4.13-5.96) 17.8 (16.2-19.4) 6.09 (5.09-7.09) 9.46 (8.24-10.7) n = 2242 Turks 52.0 (49.9-54.1) 13.8 (12.3-15.2) 7.1 (7.0, 7.2) 34.2 (32.3-36.2) 28.7 (26.9-30.6) 40.4 (39.9-40.9) 30.3 (28.4-32.2) 10.5 (9.24-11.8) 36.1 (34.1-38.1) 8.8(7.58-9.92) 27.4 (25.5-29.2) 9.9 (8.63-11.1) 20.8 (19.1-22.4) n = 1795 Ghanaian 63.6 (61.4-65.9) 6.27 (5.14-7.39) 6.8 (6.7, 6.8) 30.1 (27.9-32.2) 43.2 (40.9-45.5) 45.0 (44.5-45.5) 57.8 (55.5-60.1) 11.9 (11.4-13.4) 11.9 33.3 (31.1-35.5) 10.8 (9.38-12.3) 6.80 (5.64-7.97) 10.9 (9.48-12.4) 3.4 (2.56-4.24) n = 2158 African Surinamese 69.8 (67.9-71.8) 23.4 (21.6-25.2) 6.7 (6.5, 6.7) 6.81 (5.74-7.88) 44.6 (42.5-46.7) 47.4 (46.9-47.9) 50.6 (48.5-52.7) 12.0 (10.7-13.4) 29.9 (28.1-31.9) 9.9 (8.62-11.1) 11.1 (9.76-12.4) 11.1 10.0 (8.70-11.2) 6.1 (5.11-7.13) n = 2262 62.2 (60.2- 64.2) South-Asian Surinamese 21.8 (20.0- 23.5) 6.9 (6.7, 6.9) 16.1 (14.6- 17.6) 38.5 (36.5-40.5) 45.9 (45.4-46.5) 43.8 (41.8-45.9) 19.7 (18.1-21.4) 19.3 (17.7-20.9) 12.2 (10.9-13.6) 21.8 (20.1-23.5) 14.5 (13.1-15.9) 16.9 (15.4-18.5) n = 2146 36.7 (34.6- 38.7) 60.4 (58.3- 62.4) Dutch 7.3 (7.2, 7.3) 3.00 (2.27- 3.72) 15.6 (14.1-17.2) 46.3 (45.8-46.9) 30.5 (28.6-32.5) 3.6 (2.77-4.35) 9.8 (8.53-11.1) 16.8 (15.2-18.4) 8.90 (7.69-10.1) 14.9 (13.5-16.5) 12.5 (11.1-13.8)

category (%) nd Characteristics of the study population by ethnicity category (%) category (%) rd th Sleep duration (h) 4 Short sleep (% yes) 3 Age Hypertension (% yes) Diabetes (% yes) Obese (% yes) Cholesterol (Total) Cholesterol (Total) >6.22mmol/l HDL cholesterol HDL <1.04 mmol/L LDL cholesterol LDL >4.14mmol/L Triglyceride (TG) Triglyceride >1.69 mmol/L Education 1st and 2 Table 1: Table

82

Anujuo_binnenwerk_FINAL.indd 82 30/07/2018 21:19 Association of sleep with CVD risk factors in adults n = 2202 Moroccan 57.9 (55.8-59.9) 2.36 (1.73-2.99) 27.2 (25.3-29.0) 10.4 (9.08-11.6) 1.77 (1.22-2.32) 74.1 (72.3-75.9) 12.5 (11.1-13.9) 13.3 (11.8-14.7) 47.6 (45.5-49.7) 7.56 (6.45-8.66) n = 2242 Turks 63.4 (61.4-65.4) 3.61 (2.84-4.39) 19.9 (18.3-22.6) 9.95 (8.71-11.2) 2.72 (2.05-3.39) 47.2 (45.1-49.3) 18.6 (16.9-20.2) 33.9 (31.9-35.9) 42.2 (40.2-44.3) 21.9 (20.2-23.6) n = 1795 Ghanaian 16.9 (15.2-18.6) 17.8 (16.1-19.6) 32.6 (30.4-34.8) 30.4 (28.2-32.5) 0.89 (0.46-1.33) 86.2 (84.6-87.8) 8.69 (7.39-9.99) 4.74 (3.75-5.72) 53.9 (51.7-56.3) 49.4 (47.0-51.7) 4 n = 2158 African Surinamese 19.0 (17.3-20.7) 10.6 (9.31-11.9) 52.6 (50.5-54.8) 15.4 (13.9-16.9) 1.48 (0.97-1.99) 49.8 (47.7-51.9) 19.5 (17.8-21.1) 30.5 (28.6-32.4) 57.8 (55.7-59.9) 67.1 (65.1-69.1) n = 2262 South-Asian Surinamese 35.5 (33.6-37.5) 10.0 (8.79-11.3) 30.9 (29.0-32.9) 19.7 (18.0-21.3) 3.40 (2.66-4.15) 58.0 (55.9-60.0) 13.8 (12.4-15.2) 28.0 (26.2-29.9) 52.9 (50.8-54.9) 56.3 (54.2-58.3) n = 2146 Dutch 37.9 (35.9-40.0) 18.3 (16.7-19.9) 32.9 (30.9-34.9) 8.76 (7.56-9.96) 1.82 (1.25-2.38) 36.4 (34.4-38.4) 38.0 (35.9-40.1) 25.4 (23.6-27.2) 75.1 (73.2-76.9) 91.8 (90.7-92.9) Marital status Married (% yes) Living together (% yes) Single (% yes) Divorced (% yes) Widowed (% yes ) Smoking status Never smoked (% yes) Ex-smoker (% yes) Current smoker (% yes) Physical activity (% yes) Alcohol intake (% yes) Table 1 - continued Table Data are presented as mean and percentages with 95% CI

83

Anujuo_binnenwerk_FINAL.indd 83 30/07/2018 21:19 Chapter 4 Hypertension Diabetes Obesity Association between sleep duration and obesity, diabetes, and hypertension among ethnic groups in Amsterdam diabetes, and hypertension among ethnic groups in Association between sleep duration and obesity, Fig. 1:

84

Anujuo_binnenwerk_FINAL.indd 84 30/07/2018 21:19 Association of sleep with CVD risk factors in adults Triglyceride

HDL Cholesterol HDL 4 LDL Cholesterol LDL High Cholesterol Association between sleep duration and dyslipidaemia among ethnic groups in Amsterdam Association between sleep duration and dyslipidaemia among ethnic groups in Fig. 2:

85

Anujuo_binnenwerk_FINAL.indd 85 30/07/2018 21:19 Chapter 4 1.53 (1.03-2.27)** 1.55 (1.04-2.29)** 1.57 (1.06-2.33)** 0.93 (0.80-1.08) 0.94 (0.81-1.09) 0.94 (0.81-1.09) 1.30 (1.03-1.64)** 1.29 (1.02-1.63)** 1.27 (1.01-1.59)** 1.10 (0.97-1.26) 1.10 (0.97-1.26) PR (95%CI) Moroccan n = 2202 1.09 (0.96-1.24) 0.96 (0.71-1.29) 0.95 (0.70-1.28) 0.95 (0.71-1.27) 1.17 (1.03-1.32)** 1.21 (1.07-1.37)** 1.21 (1.07-1.37)** 1.47 (1.16-1.87)** 1.59 (1.26-2.02)* 1.63 (1.28-2.07)* 1.14 (1.03-1.29)** 1.16 (1.04-1.29)** PR (95%CI) Turks Turks n = 2242 1.17 (1.05-1.31)** 0.89 (0.67-1.19) 0.89 (0.66-1.18) 0.92 (0.69-1.22) 0.98 (0.91-1.06) 0.98 (0.91-1.06) 0.98 (0.90-1.05) 1.00 (0.76-1.32) 0.99 (0.75-1.29) 1.04 (0.79-1.35) 1.05 (0.92-1.21) 1.04 (0.91-1.19) PR (95%CI) Ghanaian n = 1795 1.05 (0.92-1.20)

0.84 (0.64-1.09) 0.83 (0.63-1.08) 0.83 (0.64-1.08) 1.03 (0.96-1.12) 1.06 (0.98-1.15) 1.06 (0.98-1.15) 1.37 (1.09-1.74)** 1.45 (1.14-1.84)** 1.46 (1.15-1.85)** 1.25 (1.11-1.44)** 1.25 (1.09-1.43)** PR (95%CI) African Surinamese n = 2158 1.26 (1.10-1.43)** 0.91 (0.71-1.16) 0.88 (0.69-1.12) 0.89 (0.70-1.12) 1.03 (0.95-1.13) 1.06 (0.98-1.16) 1.06 (0.97-1.15) 0.98 (0.84-1.17) 1.02 (0.86-1.21) 1.01 (0.85-1.18) 1.20 (1.01-1.43)** 1.21 (1.01-1.44)** PR (95%CI) 1.19 (0.99-1.42) South-Asian 2262 = Surinamese n 0.93 (0.73-1.19) 0.97 (0.76-1.23) 0.97 (0.76-1.23) 1.08 (0.95-1.25) 1.15 (1.00-1.33)** 1.19 (1.04-1.37)** 1.05 (0.63-1.73) 1.18 (0.70-1.99) 1.31 (0.78-2.19) 1.41 (1.04-1.89)** 1.45 (1.07-1.95)** PR (95%CI) 1.53 (1.14-2.06)** Dutch n = 2146 Prevalence ratio(s) for the relationship between sleep duration (short vs normal sleep) and cardiovascular risk factors by ethnicity Model 3 Model 2 Raised Total Raised Total cholesterol Model 1 Model 3 Model 2 Hypertension Model 1 Model 3 Model 2 Diabetes Model 1 #Model 3 Model 2 Obesity Model 1 Table 2: Table

86

Anujuo_binnenwerk_FINAL.indd 86 30/07/2018 21:19 Association of sleep with CVD risk factors in adults 0.84 (0.63-1.12) 0.85 (0.64-1.15) 0.83 (0.62-1.11) 1.08 (0.75-1.54) 1.09 (0.76-1.57) 1.08 (0.75-1.56) 0.87 (0.71-1.05) 0.89 (0.73-1.09) PR (95%CI) Moroccan n = 2202 1.89 (0.73-1.09) 1.00 (0.84-1.19) 1.07 (0.90-1.28) 1.05 (0.89-1.25) 0.97 (0.74-1.29) 0.98 (0.74-1.29) 0.97 (0.74-1.28) 1.04 (0.91-1.19) 1.09 (0.95-1.26) PR (95%CI) Turks Turks n = 2242 1.09 (0.95-1.26) 0.96 (0.58-1.57) 0.94 (0.57-1.54) 0.87 (0.53-1.41) 0.75 (0.56-0.99)** 0.72 (0.55-0.96)** 0.75 (0.57-0.99)** 1.19 (0.83-1.71) 1.16 (0.79-1.67) PR (95%CI) Ghanaian n = 1795 1.10 (0.77-1.58)

4 . 0.81 (0.58-1.14) 0.85 (0.61-1.19) 0.85 (0.60-1.18) 0.79 (0.60-1.04) 0.83 (0.63-1.09) 0.83 (0.63-1.08) 0.91 (0.72-1.17) 0.98 (0.77-1.26) PR (95%CI) African Surinamese n = 2158 0.99 (0.78-1.27)

1.49 (1.24-1.81)* 1.58 (1.31-1.91)* 1.57 (1.30-1.89)* 0.90 (0.73-1.12) 0.90 (0.73-1.12) 0.90 (0.73-1.12) 1.29 (1.11-1.51)** 1.41 (1.21-1.65)* PR (95%CI) 1.40 (1.20-1.64)* South-Asian 2262 = Surinamese n 0.95 (0.72-1.25) 1.06 (0.80-1.40) 1.12 (0.84-1.48) 0.91 (0.69-1.19) 0.96 (0.73-1.25) 0.95 (0.73-1.23) 1.10 (0.79-1.55) 1.19 (0.85-1.68) PR (95%CI) 1.27 (0.91-1.79) Dutch n = 2146 Model 3 Model 2 Raised Triglyceride Model 1 Model 3 Model 2 Raised LDL Raised LDL cholesterol Model 1 Model 3 Model 2 Reduced HDL Reduced HDL cholesterol Model 1 Table 2 - continued Table Model 1: adjusted for Age and sex, model 2: 1 plus education marital status, 3: 2 BMI, smoking, alcohol Model 1: adjusted for physical inactivity (*: p < 0.001, **: 0.05) #Obesity model 3: 2 plus smoking, alcohol and physical inactivity

87

Anujuo_binnenwerk_FINAL.indd 87 30/07/2018 21:19 Chapter 4

4. DISCUSSION

Key finding

This study investigated whether there are differences in the association between short sleep duration and CVD risk factors across various ethnic minority groups living in the Netherlands. The result of this study demonstrated that sleep duration was association with CVD risk factors, but the association varies between ethnic groups and depend on CVD risk factor being considered. Short sleep was significantly associated with obesity and diabetes in most ethnic groups. By contrast, the associations between short sleep and hypertension and lipid levels were absent in most ethnic groups with the exception of association with hypertension in Dutch and Turks, reduced HDL cholesterol and triglyceride in South-Asian Surinamese, and raised total cholesterol in Moroccans. Additional adjustment for SES and lifestyle factors did not substantially change the observed associations.

Discussion of key findings

Previous studies have shown that short sleep was related to poor health outcomes, and the results of this study partly support this finding [40]. This study showed that short sleep was significantly associated with higher prevalence of some CVD risk factors in some ethnic minority groups, but this was not consistent for each risk factor or each ethnic group. The association between short sleep and CVD risk factors was more consistent for obesity and diabetes. This supports the biological mechanism linking short sleep and obesity and diabetes, as has been proposed in several studies [41-52]. For instance, short sleep has been associated with altered metabolic hormones (leptin and ghrelin) [41 - 43], energy balance [44, 45], time available to eat [44], timing of meals [46, 47], and increased intake of high-calorie foods [46-49] which could eventually induce obesity. It has been proposed that reduced sleep may lead to a reduction in brain glucose utilization, increase in growth hormone, elevation in evening cortisol concentrations and sympathetic nervous activity. These mechanisms may give rise to diabetes. There was no significant association (except in few ethnic groups) for other risk factors (hypertension and lipids). This seems to suggest that apart from known proposed biological mechanisms linking short sleep and hypertension and lipids, [15, 53-57], a different unidentified or non-biological mechanism may be involved.

88

Anujuo_binnenwerk_FINAL.indd 88 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

Although these mechanisms might drive the association between short sleep and obesity/diabetes, the association between short sleep and obesity/ diabetes are rather consistent across groups, though in some groups, there were no associations. The authors` finding is partly similar to a recent study that found ethnic-specific associations between self-reported sleep and cardiometabolic risk factors in various ethnic groups in the USA [22]. The reason for this observation is not fully understood, but may be partly related to differences in lifestyle factors and eating behaviour, particularly during the awake period, which may differ between ethnic groups, and may alter the association between short sleep and CVD risk factors. Ethnic differences may also be due to the quality of sleep during the remaining short sleep time. So, if there is no interruption of sleep during the short sleep time, and therefore a better sleep quality, this may also influence CVD risk factors. Unfortunately, data on sleep quality are not available in this study.

No significant association between short sleep and CVD risk factors in Ghanaians was observed. The lack of association between short sleep and CVD risk factors among Ghanaian is surprising, given that the prevalence 4 of short sleep is much higher in this group compared to the other groups. Smoking prevalence is indeed lower among Ghanaians, and therefore might be protective from increased CVD risk. However, smoking prevalence was lower in both short (5.7%) and normal sleep (3.7%) Ghanaian groups, and therefore could not explain the lack of association with CVD risk factors. Furthermore, if smoking was categorised into: Heavy smokers (15 or more cigarettes per day, light smokers (1-14 cigarettes per day), ex-smokers and never smokers were included in the model, the result remains essentially the same. Another reason for the lack of association between short sleep and CVD risk in the Ghanaian group may partly be attributed to various coping strategies employed by Ghanaians to forestall the manifestation of CVD risk factors even though they were short sleepers. It could be that the Ghanaians used adaptive coping strategies positively to better their health. Previous studies have shown that stress, which can increase prevalence of short sleep among ethnic groups [57-58], may or may not predispose to CVD risk factors depending on how the affected individuals cope with the stress to which they are exposed [59]. It has been documented that Ghanaians were able to cope with challenges related to migration stress and acculturation in a flexible way and were able to maintain health [60], thus, this may further explain why we did not found significant association between short sleep and CVD risk factors in Ghanaians. Further study is needed to examine to what extent coping strategies can influence or mediate

89

Anujuo_binnenwerk_FINAL.indd 89 30/07/2018 21:19 Chapter 4

the association between sleep duration and CVD risk factors among various ethnic minority groups.

The strength of this study lies in the large sample sizes, hence a more reliable estimations. In addition, investigation was conducted on multiple ethnic groups living together in one city in a similar manner. A limitation of the study is that only self-reported data on sleep were used, which is subject to recall bias; hence the participants may have under -or over- reported short sleep durations. However, because of such large numbers used in this study, this will probably level out. Being a cross-sectional study, causal associations between short sleep and CVD risk factors could not be established, although important associations of short sleep and CVD risk factors were observed in some ethnic groups. In addition, information on daytime sleeping which may affect sleep duration was lacking. The authors were unable to stratify the analysis by sex, or include very short sleep as a category because of small numbers.

In conclusion, the results of this study demonstrated that short sleep was significantly associated with obesity and diabetes in most ethnic groups. By contrast, the associations between short sleep and hypertension and dyslipidaemia were absent in most ethnic groups. SES and lifestyle factors contributed little to the observed association. This finding highlights that the potential factors linking short sleep and CVD risk factors and coping strategies may vary between ethnic groups. Future studies should examine the factors contributing to the different associations between short sleep and CVD risk factors among ethnic groups.

90

Anujuo_binnenwerk_FINAL.indd 90 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

4

91

Anujuo_binnenwerk_FINAL.indd 91 30/07/2018 21:19 Chapter 4

REFERENCES

1. Mottillo S, Filion KB, Genest J, Joseph L, Pilote L, Poirier P, Rinfret S, Schiffrin EL, Eisenberg MJ. The metabolic syndrome and cardiovascular risk a systematic review and meta-analysis. J Am CollCardiol. 2010 Sep 28;56(14):1113-32. 2. Galassi A, Reynolds K, He J. Metabolic syndrome and risk of cardiovascular disease: a meta-analysis. Am J Med. 2006 Oct;119(10):812-9. 3. Grundy SM. Metabolic syndrome pandemic. ArteriosclerThrombVasc Biol. 2008; 28: 629-636. 4. Gangwisch JE, Heymsfield SB, Boden-Albala B, Buijs RM, Kreier F, Pickering TG, Rundle AG, Zammit GK, Malaspina D. Short sleep duration as a risk factor for hypertension: analyses of the first National Health and Nutrition Examination Survey. Hypertension. 2006 May; 47(5):833-9. 5. Cappuccio FP, Stranges S, Kandala NB, Miller MA, Taggart FM, Kumari M, Ferrie JE, Shipley MJ, Brunner EJ, Marmot MG. Gender-specific associations of short sleep duration with prevalent and incident hypertension: the Whitehall II Study. Hypertension. 2007 Oct; 50(4):693-700 6. Spiegel K, Knutson K, Leproult R, Tasali E, Van Cauter E. Sleep loss: a novel risk factor for insulin resistance and Type 2 diabetes. J Appl Physiol 2005 Nov; 99(5):2008-19. 7. Ayas NT, White DP, Al-Delaimy WK, Manson JE, Stampfer MJ, Speizer FE, Patel S, Hu FB. A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care. 2003 Feb; 26(2):380-4. 8. Yaggi HK, AraujoAB, McKinlay JB. Sleep duration as a risk factor for the development of type 2 diabetes. Diabetes Care. 2006 Mar; 29(3):657-61. 9. Gottlieb DJ, Punjabi NM, Newman AB, Resnick HE, Redline S, Baldwin CM, Nieto FJ. Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med. 2005 Apr 25; 165(8):863-7. 10. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004 Dec; 1(3):e62. 11. Gangwisch JE, Malaspina D, Boden-Albala B, Heymsfield SB. Inadequate sleep as a risk factor for obesity: analyses of the NHANES I. Sleep. 2005 Oct; 28(10):1289-96. 12. Patel SR, Malhotra A, White DP, Gottlieb DJ, Hu FB. Association between reduced sleep and weight gain in women. Am J Epidemiol. 2006 Nov 15; 164(10):947-54.

92

Anujuo_binnenwerk_FINAL.indd 92 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

13. Schwartz J, Allison MA, Ancoli-Isreal S, Hovell MF, Patterson RE, Natarajan L, Marshall SJ, Grant I. Sleep, type 2 diabetes, dyslipidemia, and hypertension in elderly Alzheimers caregivers. Arch Gerontol Geriatr. 2013 Jul-Aug;57(1):70-7. 14. Kaneita Y, Uchiyama M, Yoshiike N, Ohida T. Associations of usual sleep duration with serum lipid levels. Sleep May 1, 2008;31(5):645-652. 15. Gangwisch JE; Malaspina D; Babiss LA; Opler MG; Posner K; Shen S; Turner JB; Zammit GK; Ginsberg HN. Short sleep duration as a risk factor for hypercholesterolemia: analyses of the National Longitudinal Study of Adolescent Health. Sleep. Jul 1, 2010; 33(7): 956–961. 16. Bjorvatn B, Sagen IM, Øyane N, Waage S, Fetveit A, Pallesen S, Ursin R. The association between sleep duration, body mass index and metabolic measures in the Hordaland Health Study. J Sleep Res. 2007 Mar; 16(1):66-76. 17. Choi KM, Lee JS, Park HS, Baik SH, Choi DS, Kim SM. Relationship between sleep duration and the metabolic syndrome: Korean National Health and Nutrition Survey 2001. Int J Obes (Lond). 2008 Jul; 32(7):1091-7. 18. Williams CJ, Hu FB, Patel SR, Mantzoros CS. Sleep duration and snoring in relation to biomarkers of cardiovascular disease risk among women with type 4 2 diabetes. Diabetes Care. 2007 May; 30(5):1233-40. 19. Kerkhofs M, Boudjeltia KZ, Stenuit P, Brohée D, Cauchie P, Vanhaeverbeek M. Sleep restriction increases blood neutrophils, total cholesterol and low density lipoprotein cholesterol in postmenopausal women: A preliminary study. Maturitas. 2007 Feb 20; 56(2):212-5. 20. Ferrie JE, Shipley MJ, Cappuccio FP, Brunner E, Miller MA, Kumari M, Marmot MG. A prospective study of change in sleep duration: associations with mortality in the Whitehall II cohort. Sleep. 2007 Dec; 30(12):1659-66. 21. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiatry 2002;59:131- 6. 22. Grandner MA, Chakravorty S, Perlis ML, Oliver L, Gurubhagavatula I. Habitual sleep duration is associated with self-reported and objectively determined cardiometabolic risk factors. Sleep Med. 2014 Jan;15(1):42-50. 23. Guo X, Zheng L, Wang J, Zhang X, Zhang X, Li J, Sun Y. Epidemiological evidence for the link between sleep duration and high blood pressure: A systematic review and meta-analysis. Sleep Med. 2013 Apr;14(4):324-32. 24. Grandner MA, Jackson N, Gerstner JR, Knutson KL. Dietry nutrients associated with short and long sleep duration. Data from a nationally representative sample. Appetite. May 2013;64:71-80.

93

Anujuo_binnenwerk_FINAL.indd 93 30/07/2018 21:19 Chapter 4

25. Agyemang C, van Valkengoed IG, van den Born BJ, Bhopal R, Stronks K. Heterogeneity in sex differences in the metabolic syndrome in Dutch white, Surinamese African and South Asian populations. Diabet Med. 2012 Sep;29(9):1159-64. 26. Van Leest LATM, van Dis SJ, Verschuren WMM. Hart- en vaatziekten bij allochtonen in Nederland Een cijfermatige verkenning naar leefstijl- en risicofactoren, ziekte en sterfte: RIVM Rapport. Bilthoven (The Netherlands); 2002. 27. Oosterberg EH, Devillé W, Brewster LM, Agyemang C, van den Muijsenbergh M. Chronic disease in ethnic minorities: tools for patient- centred care in diabetes, hypertension and COPD. Ned. Tijdschr Geneeskd. 2013;157(16):A5669. 28. Agyemang C, Kunst AE, Bhopal R, Zaninotto P, Nazroo J, Unwin N, van Valkengoed I, Redekop WK, Stronks K. A cross-national comparative study of metabolic syndrome among non-diabetic Dutch and English ethnic groups. Eur J Public Health. 2013 Jun;23(3):447-52. 29. Anujuo K, Stronks K, Snijder MB, Jean-Louis G, Ogedegbe G, Agyemang C. Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS study. Sleep Med. 2014 Sep;15(9):1115-21. 30. Agyemang C, van Oeffelen AA, Norredam M, Kappelle LJ, Klijn CJ, Bots ML, Stronks K, Vaartjes I. Ethnic disparities in ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage incidence in the Netherlands. Stroke. 2014 Nov;45(11):3236-42 31. How much sleep do we really need? Sleepfoundation.org/how-sleep-works/ how-much-sleep-do-we-really-need. Assessed on November 07 (2014-12-07). 32. Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, Dinges DF, Gangwisch J, Grandner MA, Kushida C, Malhotra RK, Martin JL, Patel SR, Quan SF, Tasali E. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. SLEEP 2015;38(6):843–844. 33. Meng L, Zheng Y, Hui R. The relationship of sleep duration and insomnia to risk of hypertension incidence: a meta-analysis of prospective cohort studies. Hypertens Res. 2013 Nov;36(11):985-95. 34. Mosca M, Aggarwal B. Sleep duration, snoring habits, and cardiovascular disease risk factors in an ethnically diverse population. J Cardiovasc Nurs. 2012 May-June;27(3):263-9 35. Cappucio FP, Taggart FM, Kandala NB, Currie A, Peile E, Stranges S, Miller MA. Meta-analysis of short sleep duration and obesity in children and adults. Sleep. 2008 May;31(5):619-26.

94

Anujuo_binnenwerk_FINAL.indd 94 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

36. Friedewald WT et al. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972, 18;499-502. 37. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2010 Jan;375:181-3. 38. http://www.merckmanuals.com/home/hormonal_and_metabolic_disorders/ cholesterol_disorders/dyslipidemia.html?qt=hormonal and metabolic disorders, cholesterol disorder dyslipidemia&alt=sh. Assessed on February 02, 2015. 39. Wendel-Vos GC, Schuit AJ, Saris WH, Kromhout D. Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity. J Clin Epidemiol. 2003 Dec;56(12):1163-9. 40. Altman NG. Izci-Balserak B. Schopfer E. Jackson N. Rattanaumpawan P. Patel NP. Grandner MA. Sleep duration versus sleep insufficiency as predictors of cardiometabolic health outcomes. Sleep Med. 2012 Dec;13(10):1261-70. 41. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med. 4 2004;141:846–850. 42. Chaput JP, Despres JP, Bouchard C, Tremblay A. Short sleep duration is associated with reduced leptin levels and increased adiposity: Results from the Quebec family study. Obesity (Silver Spring) 2007;15:253–61. 43. Morselli L, Leproult R, Balbo M, Spiegel K. Role of sleep duration in the regulation of glucose metabolism and appetite. Best Pract Res Clin Endocrinol Metab. 2010;24:687–702. 44. St-Onge MP, Roberts AL, Chen J, Kelleman M, O’Keeffe M, Roychoudhury A, et al. Short sleep duration increases energy intakes but does not change energy expenditure in normal-weight individuals. Am J Clin Nutr. 2011 Aug;(2):410-6. 45. Caron AM, Stephenson R. Energy expenditure is affected by rate of accumulation of sleep deficit in rats. Sleep. 2010;33:1226–35. 46. Baron KG, Reid KJ, Kern AS, Zee PC. Role of sleep timing in caloric intake and BMI. obesity (Silver Spring) 2011 Jul;19(7):1374-81. 47. Duffy JF, Kronauer RE, Czeisler CA. Phase-shifting human circadian rhythms: influence of sleep timing, social contact and light exposure.J Physiol. 1996 Aug;495 ( Pt 1):289–97. 48. Grandner MA, Kripke DF, Naidoo N, Langer RD. Relationships among dietary nutrients and subjective sleep, objective sleep, and napping in women. Sleep Med. 2010;11:180–4.

95

Anujuo_binnenwerk_FINAL.indd 95 30/07/2018 21:19 Chapter 4

49. Nedeltcheva AV, Kilkus JM, Imperial J, Kasza K, Schoeller DA, Penev PD. Sleep curtailment is accompanied by increased intake of calories from snacks. Am J Clin Nutr. 2009;89:126–33. 50. Thomas M, Sing H, Belenky G, Holcomb H, Mayberg H, Dannals R, Wagner H, Thorne D, Popp K, Rowland L, et al. Neural basis of alertness and cognitive performance impairments during sleepiness. I Effects of 24 h of sleep deprivation on waking human regional brain activity. J Sleep Res. 2000;9(4):335–352. 51. Spiegel K, Leproult R, Colecchia EF, L’Hermite-Baleriaux M, Nie Z, Copinschi G, Van Cauter E. Adaptation of the 24-h growth hormone profile to a state of sleep debt. Am J Physiol Regul Integr Comp Physiol. 2000;279(3):R874–R883. 52. Van Cauter E, Polonsky KS, Scheen AJ. Roles of circadian rhythmicity and sleep in human glucose regulation. Endocrine Reviews. 1997;18:716–738. 53. Knutson KL. Does inadequate sleep play a role in vulnerability to obesity? Am J Hum Biol. 2012 May-Jun;24(3):361-71. 54. Miller MA, Cappuccio FP. Inflammation, sleep, obesity and cardiovascular disease. Curr Vasc Pharmacol. 2007;5:93–102. 55. 55. Wolk R, Somers VK. Sleep and the metabolic syndrome. Exp Physiol. 2007;92:67–78 56. Goncharuk VD, van Heerikhuize J, Dai JP, Swaab DF, Buijs RM. Neuropeptide changes in the suprachiasmatic nucleus in primary hypertension indicate functional impairment of the biological clock. J Comp Neurol. 2001;431:320– 330. 57. Sadeh, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychol, 2004;23(5):542-545. 58. Amira S. External factors that influence sleep, Healthy Sleep Division of Sleep Medicine at Havard Medical School, 18 December 2007; Web 15 September 2013. Healthysleep.med.havard.edu/health/science/how/external-factors. 59. Lindquist TL, Beilin LJ, Knuiman NW. Influence of lifestyle, coping, and job stress on blood pressure in men and women. Hypertens 1997 Jan;29(1 pt 1):1-7. 60. Knipscheer JW, Kleber RJ. Acculturation and mental health among Ghanaians in the Netherlands. Int J Soc Psychiatry. 2007;53:353–369.

96

Anujuo_binnenwerk_FINAL.indd 96 30/07/2018 21:19 Association of sleep with CVD risk factors in adults

4

97

Anujuo_binnenwerk_FINAL.indd 97 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 98 30/07/2018 21:19 CHAPTER 5

RELATIONSHIP BETWEEN SLEEP DURATION AND ARTERIAL STIFFNESS IN A MULTI-ETHNIC POPULATION: THE HELIUS STUDY

Published

Kenneth Anujuo, Karien Stronks, Marieke B. Snijder, Girardin Jean- Louis, Bert-Jan van den Born, Ron J Peters, Charles Agyemang. Relationship between sleep duration and arterial stiffness in a multi-ethnic population: the HELIUS study. Chronobiol Int. 2016;33:543-552

Anujuo_binnenwerk_FINAL.indd 99 30/07/2018 21:19 Chapter 5

ABSTRACT

We examined the relationship between sleep duration and arterial stiffness among a multi-ethnic cohort, and whether the associations differed among ethnic minority groups in the Netherlands. Data were derived from 10994 participants (aged 18-71 years) of the Healthy Life in an Urban Setting (HELIUS) study. Self-reported sleep duration was categorised into: short (<7 hours/night), healthy (7-8 hours/night), and long (>=9 hours/night). Arterial stiffness was assessed by duplicate pulse-wave velocity (PWV in m/s) measurements using the Arteriograph system. The association of sleep duration with PWV was analysed using linear regression (β) with 95% confidence interval (CI). Results showed that neither short nor long sleep was related to PWV in all ethnic groups, except for long sleep in Dutch men which was associated with higher PWV (indicating stiffer arteries) after adjustment for potential confounders (β 0.67, 95%CI, 0.23-1.11). Our study showed no convincing evidence that sleep duration was related to arterial stiffness among various ethnic groups. The link between sleep duration and cardiovascular outcomes does not seem to operate through arterial stiffness. Further research is needed to consolidate these findings.

100

Anujuo_binnenwerk_FINAL.indd 100 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

INTRODUCTION

Evidence shows that sleep is involved in many pathophysiologic events in the body, including changes in heart rate (Elsenbruch et al., 1999), regulation of immunity (Krueger & Majde, 2003), changes in blood clotting activity (Heiser et al., 1997; Irokoawa et al., 1998; Muller et al., 1989), mental health (Bokjo et al., 2013), and cardio-metabolic diseases (Knutson, 2010), and related risk factors (Kahn et al., 2005). Several studies have shown an independent association between both short and long sleep duration and cardiovascular diseases (CVD) (Wingard & Berkman, 1983; Qureshi et al., 1997).

The underlying mechanism of the association between sleep duration and CVD is not clear, hence remains a subject of further investigation. Arterial stiffness predicts CVD events and all-cause mortality (Vlachopoulos et al., 2010; Mitchell et al., 2010; Meguro et al., 2009; Xiong et al., 2012). Arterial stiffness is regarded as a an early marker of CVD, and considered as an important etiological pathway leading to cardiovascular complications (O’Rourke et al., 2002). Relationship between sleep duration and arterial stiffness may exist, as both share common determinants, for instance, both are associated with inflammation, hormonal changes and metabolic risk factors; both share common cardiovascular risk factors, and both are associated with increased risk of CVD. Arterial stiffness could also play a role 5 in the link between sleep duration and cardiovascular outcomes. However, the few studies that assessed the relationship between sleep duration and arterial stiffness show some degree of inconsistency. For instance, one study found that long sleep, not short sleep duration, was associated with increased PWV but only in men (Yoshioka et al., 2011). By contrast, Sunbul et al., (2014), found that sleep deprivation was associated with increased PWV in healthy adults whereas Wolff and colleagues found that both short and long sleep were associated with increased carotid intima-media thickness (IMT) (Wolf et al., 2008).

Relationship between sleep duration and PWV has not been investigated among ethnic minority groups, who are known to have shorter sleep durations and more CVD risk compared to their host European populations (Anujuo et al., 2014; Agyemang et al., 2005). Therefore, investigation of the relationship between arterial stiffness and sleep duration among various ethnic groups could help to gain insight into the role of sleep in CVD and on ethnic differences in cardiovascular outcomes. The purpose of the present study, therefore, was to investigate the association between sleep duration

101

Anujuo_binnenwerk_FINAL.indd 101 30/07/2018 21:19 Chapter 5

and arterial stiffness in a multi-ethnic cohort, and whether the associations differed among ethnic minority groups in the Netherlands.

MATERIALS AND METHODS

Study population

The current study was based on baseline data from the HELIUS (Healthy Life in an Urban Setting) study. The aims and design of the HELIUS study have been described elsewhere (Stronks et al., 2013). In brief, HELIUS is a large-scale cohort study on health and health care among different ethnic groups living in Amsterdam. The study started in 2011 and it includes individuals aged 18-70 years from the six major ethnic groups in Amsterdam (African-Surinamese, South-Asian Surinamese, Turkish, Moroccan, Ghanaian, and Dutch origin), and focuses on three major disease categories: cardiovascular disease, mental health and infectious diseases. Participants were randomly sampled from the municipal registers, stratified by ethnicity. The study protocols were approved by the AMC Ethical Review Board. The study protocols conform to international ethical standards in line with the policy of Chronobiology Biology International (Portallupi et al., 2010). All participants provided written informed consent.

For the current study, we used baseline data that were collected until June 2014. Data from both questionnaire and the physical examination were available in 13316 participants. For current analyses, participants with a Javanese Surinamese origin (n=137), other/unknown Surinamese origin (n=141) or other/unknown origin (n=26) were excluded because of the small sample sizes. In addition, individuals with no data on sleep duration (n=207) and/or PWV (n=1811) were also excluded from the analysis. This resulted in a dataset of 10994 participants, including 1793 Dutch, 1815 African-Surinamese, 1828 South-Asian Surinamese, 1600 Ghanaians, 1968 Turkish and 1990 Moroccans.

Measurements

Sleep duration Participants were asked to provide information on the average number of hours they usually sleep at night. Sleep duration was assessed using the item,

102

Anujuo_binnenwerk_FINAL.indd 102 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

“How many hours do you sleep on average per night?”. Short sleep was defined as having less than 7 hours of sleep per night, healthy sleep (7-8 hours/night), and long sleep as having 9 or more hours of sleep per night This classification has also been used in previous studies (Wolfet al., 2008).

Pulse wave velocity Participants visited the research location in the morning after an overnight fast and were asked to refrain from smoking before the visit. Arterial stiffness measurements were performed in duplicate after 10 minutes of supine rest using the Arteriograph system (Tensiomed Kft., Budapest, Hungary), and the mean of these two PWV measurements was used for analyses. The details of the measurement has been described elsewhere (Snijder et al., 2015). For the calculation of PWV, the Arteriograph system assumes that the late systolic peak is generated by reflected waves from the bifurcation of the aorta. Wave travel distance, i.e. from the heart to the bifurcation and back, is estimated as the shortest distance between the suprasternal notch and pubic symphysis using a tape measure. PWV (m/s) can then be calculated as the distance travelled by the pressure wave and the time difference between the early and late systolic peak. Variability and reproducibility of PWV measurements appear better with the Arteriograph compared to the Complior and Sphychmocor system (Baulmann et al., 2008). PWV measured by the Arteriograph system generates similar PWV 5 values as obtained by MRI (Rezai et al., 2013).

Other measurements Weight was measured in light clothing only on a SECA877 to the nearest 0.1 kg. Height was measured without shoes with a portable stadiometer (SECA 217) to the nearest 0.1 cm. Blood pressure (BP) was measured using a validated automated digital BP device (Microlife WatchBP Home, Microlife AG, Heerbrugg, Switzerland) on the left arm in a seated position after the participant had seated for at least 5 minutes. All measurements were performed in duplicate; the mean of the two measurements was used in the analyses.

Fasting blood samples were taken to determine the concentration of glucose by spectrophotometry, using hexokinase as primary enzyme (Roche Diagnostics, Japan). Total cholesterol, triglycerides and high-density lipoprotein (HDL) cholesterol were determined by colorimetric spectrophotometry (Roche Diagnostics, Japan). Low-density lipoprotein cholesterol (LDL) was calculated according to the Friedewald formula (Friedewald et al., 1972).

103

Anujuo_binnenwerk_FINAL.indd 103 30/07/2018 21:19 Chapter 5

Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2). Participants were considered obese if the BMI was ≥ 30 kg/ m2.

Hypertension was defined as systolic BP ≥ 140 mmHg, or diastolic BP ≥ 90 mmHg, or being on BP lowering medication, or self-reported hypertension. Type 2 diabetes was defined as increased fasting glucose ≥ 7mmol/L or current use of glucose-lowering medication or self-reported diabetes. Dyslipidemia was defined as total cholesterol (TC) >6.22mmol/L, or high density lipoprotein cholesterol (HDL-C) <1.04mmol/L, or low density lipoprotein cholesterol (LDL-C) >4.14mmol/L, or triglyceride (TG) >1.69mmol/L (Eckel et al., 2010; Goldberg, 2015), or use of lipid- lowering medication.

Educational level was determined using participant’s highest level of education (either in the Netherlands or in the country of origin). Participants were categorized into those who have never been to school or had elementary schooling only (1st category), those with lower vocational schooling or lower secondary schooling (2nd category), those with intermediate vocational schooling or intermediate/higher secondary education schooling (3rd category), and those with higher vocational schooling or university (4th category). For the current analyses, the first two categories were combined because of small numbers.

Marital status included married/ registered/ partnership, living together, unmarried/ never married, divorced/separated, or widowed. Alcohol intake in the past 12 months (yes/no) and smoking status (yes/no/ex-smoker) were obtained by questionnaire.

Habitual physical activity was measured using the SQUASH questionnaire (Wendel-Vos et al., 2003). The SQUASH questions about multiple activities refer to a normal week in the past months. We categorized participants according to the Dutch guideline for physical activity by summing up the number of days per week for each moderate and high intensity activity lasting at least 30 minutes. A total of 5 days resulted in participants being categorized as achieving the Dutch norm for physical activity (Wendel-Vos et al., 2003).

104

Anujuo_binnenwerk_FINAL.indd 104 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

Ethnicity Participant’s ethnicity was defined according to the country of birth of the participant as well as that of his/her parents and self-report. Specifically, a participant is considered as of non-Dutch ethnic origin if he/she fulfils either of the following criteria: 1) he or she was born abroad and has at least one parent born abroad (first generation); or 2) he or she was born in the Netherlands but both his/her parents born abroad (second generation) (Stronks et al., 2009). Of the Surinamese immigrants in the Netherlands, approximately 80% are either African or South-Asian origin. Both subgroups were classified according to self-reported ethnic origin. Participants were considered as of Dutch origin if the person and both parents were born in the Netherlands.

Data Analysis

Baseline data were expressed as percentages, means or median with interquartile range and reported stratified for ethnicity. The association between sleep duration and arterial stiffness was analysed by linear regression adjustment for potential confounders. We used healthy sleep (i.e. 7-8 h/ night) as a reference category in the regression model. Independent factors were selected in the regression model based on the factors known to influence 5 arterial stiffness and sleep duration relationship (Tsai et al., 2014; Wolf et al., 2008; Yoshioka et al 2011). The influence of potential confounders was determined by the change in regression coefficient before and after inclusion in the regression models. We stratified the analysis by gender because of interaction between sleep duration and gender (p = 0.000). There was no significant interaction between sleep duration and antihypertensive agents in all the ethnic groups. All analysis were be performed using STATA 11.0 (Stata Corp, College Station, TX). A p-value of <0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

Table 1 and 2 show the characteristics of the study population by gender and ethnic group. In men, mean PWV ranges from 7.58m/s in Moroccans to 8.22m/s in South-Asian Surinamese. Among women, mean PWV ranges

105

Anujuo_binnenwerk_FINAL.indd 105 30/07/2018 21:19 Chapter 5

from 8.02m/s in Moroccans to 9.32m/s in South-Asian Surinamese. Both men and women of South-Asian Surinamese, African Surinamese, and Ghanaian origin had lower mean sleep duration and higher prevalence of short sleep than Dutch, Turkish and Moroccans. Prevalence of long sleep was higher in Moroccan, Turkish and South-Asian Surinamese men and women, compared with Dutch and other ethnic groups. Moroccan and Turkish men and women were younger, more often had lower educational levels, consume less alcohol, less often achieved the Dutch norm for physical activity and had lower prevalence of hypertension than Dutch and other ethnic groups. Dutch and South-Asian Surinamese men and women, had lower BMI and were less obese than other ethnic groups. While prevalence of diabetes was higher in South-Asian Surinamese, Ghanaians and Moroccan men and women, as well as in African Surinamese and Turkish women, compared with Dutch, the prevalence of dyslipidaemia were lower in African Surinamese, Ghanaians and Dutch men and women as well as in Moroccan women compared to other ethnic groups.

Association between sleep duration and arterial stiffness

Figures 1 and 2 show the crude associations between sleep duration and arterial stiffness in men and women, respectively.

Among men, mean PWV was higher in long sleepers in Dutch, South-Asian Surinamese and Ghanaians than short sleepers. By contrast, mean PWV was marginally higher among short sleepers in South-Asian Surinamese, African Surinamese, Dutch and Turkish and Moroccans than among long sleepers. In a multivariable model adjusting for age, marital status, mean arterial pressure, hypertension, diabetes, obesity and dyslipidaemia, smoking, alcohol consumption and physical activity, there was no significant difference in PWV between short sleepers and healthy sleepers in all ethnic groups except in Moroccans where PWV was lower in short sleepers (β= -0.21 m/s; 95% CI -0.40 to -0.02). (Table 3). There was no significant difference in PWV between long sleepers and healthy sleepers in all ethnic groups, except in Dutch where PWV was higher in long sleepers (β= 0.67; 95% CI, 0.23-1.11).

Among women, mean PWV was higher in short sleepers in South-Asian Surinamese, Ghanaian, and African Surinamese, than long sleepers, whereas, the mean PWV was marginally lower in both short and long sleepers in

106

Anujuo_binnenwerk_FINAL.indd 106 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

Moroccans, Turks and Dutch . In a multivariable model, however, there was no significant association between sleep duration and PWV (arterial stiffness) in all ethnic groups.

Fig. 1: Association between sleep duration and pulse wave velocity among ethnic groups in Amsterdam (Men) 5

Fig. 2: Association between sleep duration and pulse wave velocity among ethnic groups in Amsterdam (Women)

107

Anujuo_binnenwerk_FINAL.indd 107 30/07/2018 21:19 Chapter 5 50.2 (46.8, 53.6) 32.9 (29.7, 36.0) 16.9 (14.4, 19.5) 7.1 (6.98, 7.17) 30.5 (27.4, 33.6) 10.5 (8.43, 12.6) 7.58 (7.47, 7.70) 41.8 (40.9, 42.6) 30.1 (27.1, 33.2) (9.75, 14.1) 11.9 26.6 (26.3, 26.9) 19.2 (16.6, 21.9) 41.7 (38.4, 44.9) 91.2 (90.6, 91.9) Moroccan n = 829 57.4 (54.4, 60.4) 28.4 (25.6, 31.2) 14.2 (12.1, 16.3) 7.1 (7.02, 7.18) 30.5 (27.7, 33.3) 8.82 (7.09, 10.6) 7.77 (7.66, 7.88) 40.7 (39.9, 41.4) 33.7 (30.8, 36.6) 10.9 (8.94, 12.8) 27.9 (27.6, 28.1) 28.7 (25.9, 31.5) 57.5 (54.5, 60.5) 92.9 (92.2, 93.6) Turkish n = 972 105.4)

(103.3,

62.2 (58.6, 65.7) 29.1 (25.8, 32.5) 8.70 (6.65, 10.8) 6.5 (6.43, 6.62) 52.3 (48.6, 55.9) 4.24 (2.78, 5.70) 8.09 (7.96, 8.22) 46.9 (46.1, 47.8) 63.6 (60.1, 67.1) 14.8 (12.3, 17.4) 26.8 (26.5, 27.1) 18.4 (15.6, 21.2) 26.3 (23.1, 29.5) 104.4 Ghanaian n = 699 48.0 (44.5, 51.5) 34.3 (30.9, 37.6) 17.7 (15.0, 20.4) 6.5 (6.45, 6.64) 49.4 (45.9, 52.9) 3.81 (2.47, 5.14) 8.12 (7.97, 8.27) 47.4 (46.5, 48.4) 50.8 (47.3, 54.3) 10.8 (8.63, 12.9) 26.4 (26.1, 26.7) 17.5 (14.9, 20.2) 30.2 (27.0, 33.5) 99.6 (98.6, 100.6) African Surinamese n = 707 47.6 (44.5, 50.7) 29.9 (27.1, 32.7) 22.5 (19.9, 25.1) 6.8 (6.76, 6.92) 39.3 (36.3, 42.3) 5.53 (4.12, 6.94) 8.22 (8.08, 8.36) 45.2 (44.4, 46.1) 46.7 (43.7, 49.8) 21.5 (18.9, 24.1) 25.7 (25.5, 25.9) 15.2) 13.2 (11.1, 54.8 (51.8, 57.9) 96.0 (95.3, 96.8) South-Asian Surinamese n = 935 17.4 (14.9, 19.7) 23.9 (21.2, 26.5) 58.8 (55.7, 61.9) 7.2 (7.19, 7.29) 14.8 (12.6, 17.0) 4.46 (3.17, 5.75) 7.88 (7.76, 8.00) 47.3 (46.4, 48.2) 37.5 (34.4, 40.5) 4.99 (3.63, 6.35) 25.2 (24.9, 25.4) 9.23 (7.42, 11.0) 38.1 (35.1, 41.2) 95.2 (94.4, 95.9) Dutch n = 876 ) % category ( nd Characteristics of the study population by ethnicity and 2 st 3rd category (%) 4th category (%) Sleep duration (h) Short sleep (% yes) Long sleep (% yes) PWV (m/s) Age Hypertension (% yes) Diabetes (% yes) BMI (kg/m2) Obesity (% yes) Dyslipidaemia (% yes) Mean arterial pressure (mm Hg) Education 1 Men Table 1: Table

108

Anujuo_binnenwerk_FINAL.indd 108 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness 62.5 (59.2, 65.8) 3.18 (2.00, 4.37) 26.9 (23.9, 29.9) 6.37 (4.72, 8.01) 0.35 (0.05, 0.75) 25.4 (22.4, 28.3) 57.3 (53.9, 60.7) 12.7 (10.4, 14.9) Moroccan n = 829 64.9 (62.0, 67.8) 4.94 (3.62, 6.27) 21.6 (19.1, 24.1) 7.75 (6.12, 9.39) 0.49 (0.06, 0,91) 41.9 (38.9, 44.9) 50.2 (47.1, 53.3) 33.0 (30.2, 35.9) Turkish n = 972 22.6 (19.5, 25.6) 21.9 (18.9, 24.9) 29.9 (26.6, 33.3) 23.7 (20.6, 26.8) 0.41 (-0.05, 0.87) 8.07 (6.09, 10.1) 62.8 (59.3, 66.3) 55.7 (52.1, 53.4) Ghanaian n = 699 24.8 (21.7, 27.8) 13.2 (10.8, 15.6) 49.2 (45.7, 52.7) (9.43, 13.9) 11.7 0.38 (-0.05, 0.81) 42.8 (39.3, 46.2) 64.2 (60.8, 67.5) 79.3 (76.5, 82.2) African Surinamese n = 707 5 40.0 (36.9, 43.0) (9.86, 13.9) 11.9 32.1 (29.2, 34.9) 14.6 (12.5, 16.8) 0.99 (0.38, 1.59) 39.5 (36.5, 42.5) 56.3 (53.3, 59.4) 68.0 (65.1, 70.9) South-Asian Surinamese n = 935 41.2 (38.1, 44.3) 20.6 (18.1, 23.1) 30.2 (27.4, 33.1) 7.30 (5.68, 8.93) 0.71 (0.19, 1.23) 26.9 (24.2, 29.8) 73.5 (70.7, 76.3) 94.2 (92.8, 95.7) Dutch n = 876 Marital status Married (% yes) Living together (%yes) Single (% yes) Divorced (% yes) Widowed (% yes ) Smoking status Current smoker (%yes) Achieving Dutch norm for physical activity (% yes) Alcohol intake (%yes) Men Data are presented as mean and percentages with 95% CI Table 1 - continued Table

109

Anujuo_binnenwerk_FINAL.indd 109 30/07/2018 21:19 Chapter 5 52.0 (49.3, 54.7) 32.1 (29.6, 34.6) 15.9 (13.9, 17.9) 7.2 (7.14, 7.29) 24.7 (22.4, 26.9) 12.6 (10.9, 14.4) 28.2 (27.9, 28.5) 36.9 (34.3, 39.5) 17.3 (15.3, 19,3) 86.3 (85.7, 86.9) 8.02 (7.87, 8.16) 39.2 (38.5, 39.8) 21.7 (19.5, 23.9) (9.39, 12.8) 11.1 Moroccan n = 1161 60.6 (57.9, 63.4) 25.9 (23.5, 28.5) 15.3) 13.4 (11.5, 7.2 (7.07, 7.23) 27.2 (24.7, 29.7) 10.2 (8.46, 11.9) 29.4 (29.0, 29.8) 42.4 (39,6-45.2) 29.1 (26.5, 31.7) 88.7 (87.9, 89.5) 8.31 (8.16, 8.46) 40.3 (39.6, 40.1) 27.4 (24.9, 29.9) 10.2 (8.51, 11.9) Turkish n = 996 72.9 (70.2, 75.6) 22.5 (20.0, 25.1) 4.58 (3.32, 5.85) 6.9 (6.85, 7.02) 37.0 (34.1, 39.9) 9.49 (7.73, 11.3) 29.5 (29.2, 29.8) 43.6 (40.6, 46.6) 16.4 (14.1, 18.6) 102.3 (101.4, 103.3) 9.15 (8.98, 9.31) 43.7 (43.1, 44.3) 53.8 (50.8, 57.8) 9.93 (8.13, 11.7) Ghanaian n = 901 37.4 (34.8, 39.9) 36.0 (33.5, 38.6) 26.7 (24.3, 28.9) 6.7 (6.64, 6.79) 41.9 (39.3, 44.5) 5.69 (4.47, 6.92) 28.7 (28.4, 29.1) 37.2 (34.6, 39.7) 20.6 (18.4, 22.7) 98.9 (98.2, 99.7) 8.91 (8.77, 9.04) 47.4 (46.8, 48.1) 50.6 (47.9, 53.2) 12.7 (10.9, 14.5) African Surinamese n = 1108 51.6 (48.8, 54.4) 27.3 (24.8, 29.8) 21.1 (18.8, 23.4) 6.9 (6.79, 6.94) 37.8 (35.2, 40.5) 7.68 (6.20, 9.16) 26.8 (26.5, 27.1) 24.3 (21.9, 26.7) 31.5 (28.9, 34.1) 94.7 (93.8, 95.6) South-Asian Surinamese 9.32 (9.13, 9.51) 46.5 (45.8, 47.2) 41.5 (38.7, 44.2) 18.3 (16.1, 20.4) n = 893 17.6 (15.4, 19.8) 20.8 (18.4, 23.1) 61.7 (58.9, 64.5) Dutch 7.3 (7.22, 7.33) 16.3 (18.2, 18.4) 5.69 (4.36, 7.02) 24.3 (24.0, 24.6) 10.3 (8.52, 12.0) 25.0 (22.5, 27.5) 89.9 (88.9, 90.5) 8.29 (8.13, 8.45) 45.5 (44.7, 46.4) 24.7 (22.2, 27.1) 2.34 (1.47, 3.22) n = 917 ) ) 2 2 category (%) nd Characteristics of the study population by ethnicity and 2 st 3rd category (%) 4th category (%) Sleep duration (h) Short sleep Long sleep PWV (m/s Obesity (% yes) Dyslipidaemia (% yes) Mean arterial pressure (mm Hg) Education 1 Age Hypertension (% yes) Diabetes (% yes) BMI (kg/m Women Table 2: Table

110

Anujuo_binnenwerk_FINAL.indd 110 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness 54.9 (52.3, 57.6) 1.85 (1.13, 2.56) 27.3 (24.9, 29.7) 14.6) 12.9 (11.1, 2.66 (1.80, 3.52) 5.69 (4.45, 6.92) 41.5 (38.9, 44.1) 4.36 (3.27, 5.45) Moroccan n = 1161 62.2 (59.4, 64.9) 2.48 (1.60, 3.36) 18.6 (16.4, 20.8) (9.99, 13.6) 11.8 4.63 (3.44, 5.81) 27.2 (24.7, 29.7) 35.5 (32.8, 38.2) 12.4 (10.5, 14.2) Turkish n = 996 12.9 (10.9, 14.9) 15.0 (12.9, 17.2) 34.4 (51.9, 57.2) 34.9 (32.1, 37.8) 1.22 (0.56, 1.88) 2.44 (1.52, 3.37) 47.9 (44.9, 50.9) 44.9 (41.9, 47.9) Ghanaian n = 901 15.7 (13.8, 17.6) 9.12 (7.59, 10.7) 54.6 (51.9, 57.2) 17.6 (15.6, 19.6) 2.12 (1.35, 2.88) 23.4 (21.2, 25.7) 54.2 (51.5, 56.8) 60.1 (57.5, 62.7) African Surinamese n = 1108 5 31.9 (29.3, 34.5) 8.56 (7.01, 10.1) 30.0 (27.5, 32.5) 23.8 (21.4, 26.1) 6.61) 5.36 (4.11, 18.7 (16.6, 20.9) 50.1 (47.3, 52.9) 46.7 (43.9, 49.5) South-Asian Surinamese n = 893 Dutch 35.3 (32.5, 38.0) 16.4 (14.3, 18.5) 35.3 (32.5, 38.0) 10.0 (8.27, 11.7) 2.76 (1.82, 3.70) 25.1 (21.6, 26.5) 76.4 (73.9, 78.8) 89.8 (88.1, 91.6) n = 917 Marital status Married (% yes) Living together (%yes) Single (% yes) Divorced (% yes) Widowed (% yes ) Smoking status Current smoker (%yes) Achieving Dutch norm for physical activity (%yes) Alcohol intake (%yes) Women Table 2 - continued Table Data are presented as mean and percentages with 95% CI

111

Anujuo_binnenwerk_FINAL.indd 111 30/07/2018 21:19 Chapter 5 ** ** ** -0.19 (-0.42, 0.04) -0.19 (-0.43, 0.03) -0.21 (-0.46, 0.04) 0.44 (0.09, 0.79)** n = 1161 0.10 (-0.19, 0.39) -0.22 (-0.43, -0.01) 0.07 (-0.23, 0.37) β (95%CI) 0.02 (-0.24, 0.29) -0.03 (-0.35, 0.29) Moroccan n = 829 -0.24 (-0.64, 0.16) -0.22 (-0.42, -0.03) -0.21 (-0.40, -0.02) -0.12 (-0.36, 0.12) -0.09 (-0.33, 0.14) -0.09 (-0.34, 0.17) 0.51 (0.16, 0.87)** n = 996 -0.01 (-0.29, 0.27) 0.03 (-0.17, 0.23) -0.01 (-0.29, 0.27) β (95%CI) 0.29 (0.04, 0.04)** -0.05 (-0.37, 0.28) Turkish n = 972 -0.29 (-0.68, 0.11) -0.04 (-0.22, 0.13) -0.07 (-0.24, 0.11) 0.16 (-0.11, 0.42) 0.16 (-0.11, 0.17 (-0.09, 0.43) 0.05 (-0.25, 0.36) 0.34 (-0.01, 0.69) n = 901 0.05 (-0.49, 0.59) -0.08 (-0.32, 0.16) 0.16 (-0.38, 0.69) β (95%CI) 0.09 (-0.17, 0.36) 0.29 (-0.29, 0.87) Ghanaian n = 699 0.63 (-0.04, 1.29) -0.01 (-0.23, 0.21) -0.01 (-0.23, 0.20) 0.08 (-0.15, 0.31) 0.04 (-0.19, 0.27) 0.11 (-0.14, 0.36) 0.11 0.47 (0.18, 0.75)** n = 1108 0.01 (-0.58, 0.59) -0.02 (-0.28, 0.24) 0.06 (-0.53, 0.65) β (95%CI) 0.17 (-0.14, 0.47) 0.22 (-0.43 0.87) African Surinamese n = 707 0.31 (-0.47, 1.09) -0.04 (-0.27, 0.19) -0.03 (-0.27, 0.21) -0.13 (-0.42, 0.15) -0.08 (-0.37, 0.19) -0.09 (-0.41, 0.22) 0.65 (0.25, 1.05)** n = 893 0.29 (-0.12, 0.75) 0.01 (-0.23, 0.23) 0.39 (-0.06, 0.85) β (95%CI) 0.38 (0.08, 0.67)** 0.36 (-0.13, 0.85) South-Asian Surinamese n = 935 0.46 (-0.16, 1.07) -0.01 (-0.23, 0.22) -0.04 (-0.26, 0.18) 0.12 (-0.19, 0.43) 0.16 (-0.15, 0.46) 0.18 (-0.16, 0.51) 0.85 (0.42, 1.28)* n = 917 0.67 (0.23, 1.11)** 0.28 (-0.01, 0.56) 0.73 (0.28, 1.18)** β (95%CI) 0.51 (0.17, 0.86)** 0.83 (0.34, 1.33)** Dutch n = 876 1.09 (0.48, 1.69)* 0.20 (-0.06, 0.46) 0.17 (-0.08, 0.43) Regression coefficient (β) for the relationship between short sleep or long sleep (referenced to healthy sleep) and PWV (m/s) by ethnicity by (m/s) PWV and sleep) healthy to (referenced sleep long or sleep short between relationship the for (β) coefficient Regression Model 3 Model 2 Model 1 Unadjusted Model 3 Model 1 Model 2 Unadjusted Model 1 Unadjusted Model 2 Model 3 Short sleep Women Short sleep Men Long sleep Table 3: Table and gender

112

Anujuo_binnenwerk_FINAL.indd 112 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness 0.14 (-0.16, 0.43) 0.14 (-0.16, 0.43) β (95%CI) 0.18 (-0.14, 0.50) Moroccan -0.25 (-0.69, 0.19) 0.03 (-0.31, 0.37) 0.01 (-0.32, 0.35) β (95%CI) 0.09 (-0.27, 0.45) Turkish 0.06 (0.16, 0.87) -0.03 (-0.46, 0.40) 0.05 (-0.38, 0.48) β (95%CI) 0.12 (-0.38, 0.61) Ghanaian 0.17 (-0.41, 0.76) 5 0.24 (-0.23, 0.72) 0.18 (-0.29, 0.65) β (95%CI) 0.36 (-0.15, 0.86) African Surinamese 0.32 (-0.28, 0.91) 0.05 (-0.47, 0.58) 0.05 (-0.47, 0.57) β (95%CI) 0.18 (-0.39, 0.76) South-Asian Surinamese 0.33 (-0.42, 1.08) 0.35 (-0.12, 0.83) 0.33 (-0.15, 0.81) β (95%CI) 0.41 (-0.12, 0.93) Dutch 0.32 (-0.36, 1.00) Model 3 Model 2 Model 1 Unadjusted Long sleep Table 3 - continued Table Model 1: adjusted for age (years) and marital status (5 categories), 2: (y), mean arterial pressure (mmHg), Model 3: 2 plus hypertension (yes/no), diabetes obesity dyslipidaemia smoking alcohol (yes/no) and physical activity ( *: p < 0.001, **: 0.05).

113

Anujuo_binnenwerk_FINAL.indd 113 30/07/2018 21:19 Chapter 5

DISCUSSION

Key findings

We investigated the association between sleep duration and arterial stiffness in a multi-ethnic cohort, and whether the associations differed among ethnic minority groups in Netherlands. Our study showed that both long and short sleep duration was not associated with arterial stiffness as measured with PWV, except that PWV was higher in long sleepers as compared with healthy sleepers in Dutch men, and that PWV was lower in short sleepers in Moroccan men.

Discussion of key findings

The result of our study demonstrated that long sleep was not associated with arterial stiffness in all ethnic groups except in Dutch men. The positive association between long sleep and arterial stiffness in Dutch is consistent with previous studies among Japanese and Taiwanese populations. The mechanism for the association between long sleep and arterial stiffness is not clearly established. Previous studies suggest that the mechanism may be related to alterations in immunity (Grandner et al., 2007), and pro- inflammatory processes that result in dysregulation of collagen and elastin fibres of the vascular wall, leading to arterial stiffness (Ziemanet al., 2005). Although these proposed mechanisms may be involved in the association between long sleep and arterial stiffness, it is not fully understood how they may explain the relationship observed in Dutch men. It could be that selection mechanism may play a role, in which case, people with health problems sleep longer because of the health problems they are exposed to. Although we adjusted the association between sleep duration and arterial stiffness for physical activity, the degree of activity and the periods of inactivity may be different across ethnic groups and not fully be accounted for. One recent review, for example highlighted the positive changes in arterial properties resulting from higher intensity physical activity (Sacre et al., 2014). Age has also been shown to influence arterial stiffness (Mc Eniery et al., 2005). However, the difference persisted after adjusting for age. Furthermore, we found no interaction between sleep duration and PWV in the Dutch population (p = 0.371), suggesting that age is unlikely to explain the observed association. The association between long sleep and arterial stiffness in Dutch seems to be associated with differences in risk

114

Anujuo_binnenwerk_FINAL.indd 114 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

factors for CVD because the association is attenuated after correction for CVD risk factors. This suggests that differences in risk factors for CVD, at least in part, contribute to the association between long sleep and arterial stiffness.

For ethnic minority groups, we found no association. The reason for lack of association in ethnic minority groups is unclear. We have previously shown that the relationship between sleep duration and CVD risk factors such as obesity, diabetes, hypertension, dyslipidaemia, also differed between ethnic groups in this population (Anujuo et al., 2015). In this study, the relationship is also ethnic-specific. It has been suggested that the relationship between sleep duration and arterial stiffness may be explained by conventional CVD risk factors (Sunbul et al., 2014; Wolf et al., 2008). However, we did not observe a significant association between ethnic groups after adjusting for conventional CVD risk factors. Low SES has also been suggested to play a role in the association between sleep duration and arterial stiffness (Sunbul et al., 2014; Wolf et al., 2008). Similarly, adjustment for SES did not change the results (data not shown). On the other hand, it could be that other unmeasured factors, such as genetics may be involved in the differential associations of sleep duration with arterial stiffness among ethnic groups. Our finding partly contrasts those of Wolff and colleagues (Wolfet al., 5 2008), who found short and long sleep duration to be associated with increased carotid intima-media thickness in a general population – based study of men and women conducted in Germany. The difference in the result of our study, compared to that of Wolff et al. may be due to differences in outcome measure, for instance Wolff et al used intima-media thickness (IMT), whereas in our study, PWV velocity was used. Although, IMT has concordance with PWV, IMT is another marker of vascular aging and focuses on different stages in the atherosclerotic process, thus measuring more advanced structural changes of the vascular wall whereas PWV targets the dynamic property based on vascular function and structure. Alternatively, the observed difference in results of the two studies may be due to differences in the age of participants. Participants were younger and healthier in our study (18-71 years) than those in the previous study (45-81 years), as long sleep and arterial stiffness have been shown to be significantly associated with older age (Yoshioka et al., 2011; Woodard et al., 2011).

With respect to short sleep duration, our study demonstrated that in both men and women, short sleep was not positively associated with arterial

115

Anujuo_binnenwerk_FINAL.indd 115 30/07/2018 21:19 Chapter 5

stiffness in different ethnic groups. Our finding is consistent with previous studies which documented that short sleep was not associated with arterial stiffness in various studies conducted in Japan, and Taiwan (Yoshiokaet al., 2011; Tsai et al., 2014). Previous studies indicate that ethnic minority groups experience short sleep duration and increased prevalence of CVD risk factors and increased arterial stiffness (Anujuo et al., 2014; Agyemang et al., 2005; Snijder et al., 2015). Therefore it was expected that the association between short sleep and arterial stiffness might be a link through which short sleep may predispose to CVD in ethnic minority groups in addition to independent association that has been previously suggested (Wingard & Berkman, 1983). Contrary to our expectation, we found no association. The reason for lack of association between short sleep and arterial stiffness among ethnic groups is not fully understood. Further study is needed to understand the potential reasons behind the lack of association between short sleep and arterial stiffness among ethnic groups.

Another finding from our study indicates that neither short sleep, nor long sleep was associated with arterial stiffness in women. Our finding agrees with other studies which show gender- specific association of sleep duration with arterial stiffness (Tsai et al., 2014; Yoshioka et al., 2011). The reason for the observed non-significant association in women is unclear. Although, previous study indicated that increased hormonal and C-reactive protein activity in perimenopausal or postmenopausal women may confer protection for arterial stiffness (Woodard et al., 2011). Nonetheless, the lack of association between sleep and arterial stiffness in women are the same for men (except Dutch), suggesting that hormonal differences may not be the likely explanation. Further studies are needed to explore the role of hormonal activity, and how it may contribute to the relationship between sleep duration and arterial stiffness in women.

The strength of our study lies in a large sample size, hence more reliable estimations. Also multiple ethnic groups residing in one city were investigated together in a similar manner. However, there are limitations such as use of self-reported data for sleep, which is subject to recall bias, hence participants may have overestimated or underestimated sleep duration. Also information on daytime sleeping and sleep quality were lacking. The subjects included in this study were relatively younger than those in the earlier studies (Tsai et al., 2014; Yoshioka et al 2011). This might be associated with the different results in this study compared, compared with the earlier ones. Furthermore, the subjects with long sleep hours were fewer than those with short sleep

116

Anujuo_binnenwerk_FINAL.indd 116 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

duration. However, the number of long sleepers in each ethnic group was adequate for the analysis in the regression models. In addition, we did not exclude participants with obstructive sleep apnoea which has been shown to be associated with arterial stiffness due to lack of data on sleep apnoea in our study (Philips et al., 2013).

In conclusion, the result of our study demonstrate that both short sleep and long sleep duration were not associated with arterial stiffness in different ethnic groups, except in Dutch men in which long sleep duration was associated with increased arterial stiffness. By contrast, in Moroccan men short sleep was associated with decreased stiffness. The link between sleep duration and cardiovascular outcomes does not seem to operate through arterial stiffness. Further study is needed from other European countries to consolidate this findings.

5

117

Anujuo_binnenwerk_FINAL.indd 117 30/07/2018 21:19 Chapter 5

REFERENCES

1. Agyemang C, Bindraban N, Mairuhu G, Koopmans R, Stronks K.(2005). Prevalence, awareness, treatment and control of hypertension among black Surinamese, South-Asian Surinamese and White Dutch in Amsterdam, The Netherlands: the SUNSET study. J Hypertens. 23:197-7. 2. Anujuo K, Stronks K, Snijder MB, Jean-Lous G, Ogedegbe G, Agyemang C. (2014). Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS study. Sleep Med. 15:1115-21. 3. Anujuo K, Stronks K, Snijder MB. (2015). Relationship between short sleep duration and Cardiovascular risk factors in a multi-ethnic cohort – the HELIUS study. Sleep Med. doi:10.1016/j.sleep.2015.08.014. 4. Baulmann J, Schillings U, Rickert S. (2008). A new oscillometric method for assessment of arterial stiffness: comparison with tonometric and piezo- electronic methods. J Hypertens. 26:523-8. 5. Bokyo EJ, Seelig AD, Jacobson IG. (2013). Sleep characteristics, Mental Health, and Diabetes Risk: A prospective study of U.S. military members in the Millennium Cohort Study. Diabetes Care. 36:3154-61. 6. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. (2010). The metabolic syndrome. Lancet. 375:181-3. 7. Elsenbruch S, Harnish MJ, Orr WC. (1999). Heart rate variability during waking and sleep in healthy males and females. Sleep. 22:1067-71. 8. Friedewald WT, Levy RI, Fredrickson DS. (1972). Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 18:499-502. 9. Goldberg AC. Dyslipidemia (hyperlipidemia), (2015). http://www. merckmanuals.com/home/hormonal_and_metabolic_disorders/cholesterol_ disorders/dyslipidemia.html?qt=hormonal and metabolic disorders, cholesterol disorder dyslipidemia&alt=sh. Assessed on February 02, 2015. 10. Grandner MA, Drummond SP. (2007). Who are the long sleepers? Towards an understanding of the mortality relationship. Sleep Med Rev. 11:341-60. 11. Heiser P, Dickhaus B, Opper C. (1997). Platelet serotonin and interleukin-1 beta after sleep deprivation and recovery sleep in humans. J Neural Transm. 104:1049-58. 12. Irokoawa M, Nishinaga M, Funayama H. (1998). Effects of a change in the sleep/wake cycle on the diurnal variation of fibrinolytic parameters. J Thromb Thrombolysis. 5:165-8. 13. Kahn R, Buse J, Ferrannini E, Stem M. (2005). The metabolic syndrome: time for a critical appraisal. Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia. 48:1684-1699.

118

Anujuo_binnenwerk_FINAL.indd 118 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

14. Knutson KL. (2010). Sleep duration and cardiometabolic risk: a review of the epidemiologic evidence. Best Pract Res Clin Endocrinol Metab. 24:731-43. 15. Krueger JM, Majde JA. (2003). Humoral links between sleep and the immune system: research issues. Ann NY Acad Sci. 992:9-20. 16. Mc Eniery CM, Yasmin, Hall IR, Qasem A, Wilkinson IB, Cockcroft JR. Normal vascular aging: Differential effects on wave reflection and aortic pulse wave velocity: The Anglo-Cardiff Collaborative Trial (ACCT)JM Coll Cardiol. 2005;46:1753-60. 17. Meguro T, Nagatomo Y, Nagae A. (2009). Elevated arterial stiffness evaluated by brachial-ankle pulse wave velocity is deleterious for the prognosis of patients with heart failure. Circ J. 73:673-80. 18. Mitchell GF, Hwang SJ, Vasan RS. (2010). Arterial stiffness and cardiovascular events: The Framingham Heart Study. Circulation. 121:505-11. 19. Muller JE, Toffer GH, Stone PH. (1989). Circadian variation and triggers of onset of acute cardiovascular disease. Circulation. 79:733-4. 20. O’Rourke MF, Staessen JA, Vlachopoulos C, Duprez D, Plante GE. (2002). Clinical applications of arterial stiffness; definitions and reference values.Am J Hypertens. 15:426-44. 21. Phillips CL, Butlin M, Wong KK, Avolio AP. (2013). Is obstructive sleep apnoea causally related to arterial stiffness? A critical review of the experimental evidence. Sleep Med Rev. 17:7-18. 22. Portaluppi F, Smolensky MH, Touitou Y. Ethics and methods for biological 5 rhythm research on animals and human beings. Chronobiol Int. 2010;27:1911- 29. 23. Qureshi AI, Giles WH, Croft JB, Bliwise DL. (1997). Habitual sleep patterns and risk for stroke and coronary heart disease: a 10-year follow-up from NHANES I. Neurology. 48:904–11. 24. Rezai MR, Cowan BR, Sherratt N. (2013). A magnetic resonance perspective of the pulse wave transit time by the Arteriograph device and potential for improving aortic length estimation for central pulse wave velocity. Blood Press Monit. 18:111-8. 25. Sacre JW, Jennings GJ, Kingwell BA. (2014). Exercise and dietry influences on arterial stiffness in cardiometabolic disease. Hypertension. 63:888-93. 26. Snijder MB, Stronks K, Agyemang C, Busschers WB, Peters RJ, van den Born BJ. (2015). Ethnic differences in arterial stiffness the HELIUS study.Int J Cardiol. 191:28-33. 27. Stronks K, Kulu-Glasgow I, Agyemang C. (2009). The utility of ‘country of birth’ for the classification of ethnic groups in health research: the Dutch experience. Ethn. Health. 14:255-269.

119

Anujuo_binnenwerk_FINAL.indd 119 30/07/2018 21:19 Chapter 5

28. Stronks K, Snijder MB, Peters RJ, Prins M, Schene AH, Zwinderman AH. (2013). Unravelling the impact of ethnicity on health in Europe: the HELIUS study. BMC Public Health. 13:402. 29. Sunbul M, Kanar BG, Durmus E, Kivrak T, Sari I. (2014). Acute sleep deprivation is associated with increased arterial stiffness in healthy young adults: Sleep Breath. 18:215-20. 30. Tsai TC, Wu JS, Yang YC, Huang YH, Lu FH, Chang CJ. (2014). Long sleep duration associated with a higher risk of increased arterial stiffness in males. Sleep. 37:1315-20. 31. Vlachopoulos C, Aznaouridis K, Stefanadis C. (2010). Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol. 55:1318-27. 32. Wendel-Vos GC, Schuit AJ, Saris WH, Kromhout D. (2003). Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity. J Clin Epidemiol. 56:1163-9. 33. Wingard DL, Berkman LF. (1983). Mortality risk associated with sleeping patterns among adults. Sleep. 6:102–7. 34. Wolff B, Volzke H, Schwahn C, Robinson D, Kessler C, John U. (2008). Relation of self-reported sleep duration with carotid intima-media thickness in a general population sample. Atherosclerosis. 196:727–32. 35. Woodard GA, Mehta VG, Mackey RH. (2011). C-reactive protein is associated with in a cohort of African American and white women transitioning through menopause. Menopause. 18:1291–7. 36. Xiong Z, Zhu C, Zheng Z. (2012). Relationship between arterial stiffness assessed by brachial-ankle pulse wave velocity and coronary artery disease severity assessed by the SYNTAS score. J Atheroscler Thromb. 19:970-6. 37. Yoshioka E, Saijo Y, Kita T. (2011). Relation between self-reported sleep duration and arterial stiffness: a cross-sectional study of middle-aged Japanese civil servants. Sleep. 34:1681-1686. 38. Zieman SJ, Melenovsky V, Kass DA. (2005). Mechanisms, pathophysiology, and therapy of arterial stiffness.Arterioscler Thromb Vasc Biol. 25:932-43.

120

Anujuo_binnenwerk_FINAL.indd 120 30/07/2018 21:19 Sleep duration, relationship with arterial stiffness

5

121

Anujuo_binnenwerk_FINAL.indd 121 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 122 30/07/2018 21:19 CHAPTER 6

CONTRIBUTION OF SHORT SLEEP DURATION TO ETHNIC DIFFERENCES IN CARDIOVASCULAR DISEASE: RESULTS FROM A COHORT STUDY IN THE NETHERLANDS

Published

Kenneth Anujuo, Charles Agyemang, Marieke B. Snijder, Girardin Jean-Louis, Bert-Jan van den Born, Ron J Peters, Karien Stronks. Contribution of short sleep duration to ethnic differences in cardiovascular disease: results from a cohort study in the Netherlands. BMJ Open. 2017;7: :e017645. doi: 10.1136/ bmjopen-2017-017645.

Anujuo_binnenwerk_FINAL.indd 123 30/07/2018 21:19 Chapter 6

ABSTRACT

Objectives: We analysed association between short sleep duration and prevalence of cardiovascular disease (CVD) in a multi-ethnic population living in the Netherlands, and the contribution of short sleep to the observed ethnic differences in the prevalence of CVD, independent of CVD risk factors.

Methods: 20730 participants (aged 18-71 years) of the HELIUS study were investigated. Self-reported sleep duration was classified as: short (<7 hours/ night), and healthy (7-9 hours/night). Prevalence of CVD was assessed using the Rose questionnaire on angina pectoris, intermittent claudication and possible myocardial infarction. Association of short sleep duration with prevalent CVD, and the contribution of short sleep to the observed ethnic differences in the prevalence of CVD were analysed using adjusted prevalence ratio(s) (PR) with 95% confidence interval (CI).

Results: Results indicate that Short sleep was associated with CVD among all ethnic groups with PRs ranging from 1.41 (95%CI 1.21-1.65) in Moroccans to 1.62 (95%CI 1.20-2.18) in Dutch after adjustment for age, sex and conventional CVD risk factors. The independent contributions of short sleep (in percentage) to ethnic differences in CVD compared to Dutch were 10%, 15%, 15%, 5%, and 5% in South-Asian Surinamese, African Surinamese, Ghanaian, Turkish and Moroccan respectively.

Conclusion: Short sleep contributed to ethnic differences in CVD independent of well-known CVD risk factors particularly in Surinamese and Ghanaian groups. Reducing sleep deprivation may be a relevant entry point for reducing increased CVD risks among the various ethnic minority groups.

124

Anujuo_binnenwerk_FINAL.indd 124 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

INTRODUCTION

Cardiovascular disease (CVD) ranks first as the leading cause of mortality across the globe. Previous studies have shown that CVD is higher in ethnic minority groups compared to host populations. 1-3 It is evident that known conventional risk factors of CVD do not entirely explain ethnic differences in prevalence CVD. 4 Therefore, it becomes necessary to identify novel modifiable risk factors, which may play a mediatory role in ethnic differences in CVD, and which may reduce the likely surge in prevalence of CVD among various ethnic groups.

Several studies have shown that sleep may play a significant role in the pathogenesis and progression of cardiac and vascular diseases. 5 Studies conducted in Europe and the United States have demonstrated that aberrant sleep duration was associated with increased risk of CVD risk factors including obesity, 6 diabetes, 7 hypertension, 8 and dyslipidaemia. 9,10 Sleep may influence CVD through these risk factors and other factors. 11

However, results from existing studies on the association between short sleep duration and CVD are contradictory. 12-19 For instance, while previous studies found that short sleep was associated with CVD such as stroke, myocardial infarction, and coronary heart disease (CHD); 12-16 other studies indicate that short sleep was not associated with CHD, 17 and stroke. 18 Only one of these studies extended their investigation by including several ethnic groups, and the results suggested that short sleep was associated with CVD in Hispanic and African Americans and other ethnicities. 19 6 In Europe, data on sleep duration and CVD are lacking, and have not yet been explored among ethnic minority groups. Literature has shown that CVD risk factors are more common in these groups. 20 We have previously shown that in the Netherlands, several ethnic minority groups also experience shorter sleep duration compared to their host majority populations, 21 and the association between short sleep duration and CVD risk factors differed among the various ethnic groups. 22 Ethnic minority groups experience short sleep duration because they more frequently have a low socioeconomic status (education and occupation), are more frequently engaged in shift work, 21 or because of adverse living conditions such as crowding and stressful neighbourhood conditions. 23 Therefore, investigating the relationship between short sleep duration and CVD among various ethnic groups is relevant to provide additional insight on

125

Anujuo_binnenwerk_FINAL.indd 125 30/07/2018 21:19 Chapter 6

how short sleep is related to ethnic differences in cardiovascular disease. This might be first, through well-known CVD risk factors, second, short sleep may also independently contribute to CVD, and as such could be a novel target for prevention of CVD among various ethnic groups (see research model presented in Figure 1).

METHODS

Study population

The current study was based on baseline data from the HELIUS (Healthy Life in an Urban Setting) study. The aims and design of the HELIUS study have been described elsewhere. 24 In brief, HELIUS is a large-scale cohort study on health and health care among different ethnic groups living in Amsterdam. The study includes individuals aged 18-70 years from the six major ethnic groups in Amsterdam (African-Surinamese, South-Asian Surinamese, Turkish, Moroccan, Ghanaian, and Dutch origin), and focuses on three major disease categories: cardiovascular disease, mental health and infectious diseases. Participants were randomly sampled from the municipal registers, stratified by ethnicity. The study protocols were approved by the AMC Ethical Review Board. All participants provided written informed consent.

Baseline data collection took place in 2011-2015. Data from both questionnaire and the physical examination were available in 22165 participants. For current analyses, individual with no data on sleep duration (n=345), as well as those sleeping more than 9 hours per night (n=560) were excluded from the analysis. This resulted in a dataset of 21260 participants, including 4495 Dutch, 2933 South-Asian Surinamese, 4039 African Surinamese, 2181 Ghanaians, 3395 Turks, 3687 Moroccans, 228 Javanese Surinamese origin, 255 other/unknown Surinamese origin and 47 other/ unknown origin. Because of small sample size, the last three groups were excluded resulting in a final dataset of 20730 participants.

Ethnicity

Participant’s ethnicity was defined according to the country of birth of the participant as well as that of his/her parents and self-report. Specifically, a

126

Anujuo_binnenwerk_FINAL.indd 126 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

participant was considered as of non-Dutch ethnic origin if he/she fulfils either of the following criteria: 1) he or she was born abroad and has at least one parent born abroad (first generation); or 2) he or she was born in the Netherlands but both his/her parents born abroad (second generation). 25 Of the Surinamese immigrants in the Netherlands, approximately 80% are either African or South-Asian origin. Both subgroups were classified according to self-reported ethnic origin. Participants were considered as of Dutch origin if the person and both parents were born in the Netherlands.

Sleep duration

Participants were asked to provide information on the average number of hours they usually sleep at night. Sleep duration was assessed using the item, “How many hours do you sleep on average per night?”. Short sleep was defined as having less than 7 hours of sleep per night, according to National Sleep Foundation (NSF), American Academy of Sleep Medicine (AASM), and Sleep Research Society (SRS), which recommend 7 to 9 hours as the basal sleep need for healthy adults. 26 We focused on short sleep only because in our previous study, we demonstrated that short sleep was the major problem for the ethnic minority groups, 21 and because previous studies found that short sleep was more consistently related to CVD risk factors compared to long sleep. 27-29

Cardiovascular disease 6 The prevalence of cardiovascular disease was assessed using the Rose questionnaire on angina pectoris, intermittent claudication and possible myocardial infarction. The Rose questionnaire has three parts. Part A: include questions of experience of pain or discomfort in the chest during exercise, or walking fast or climbing stairs, and whether the pain stops or as the exercise or walking/running stops; and how long (less or more than 10 minutes), and the part of the body where pain was experienced. Part B: include questions of experience of severe chest pain lasting for half an hour or more, and Part C: include questions of experience of pains on either legs while walking uphill, or at ordinary pace, while standing still or sitting, and exact location of the pain (calf/calves), and whether the pain disappear when stopped walking, and how long (less or more than 10 minutes). Participants were classified as having angina pectoris, or possible myocardial infarction,

127

Anujuo_binnenwerk_FINAL.indd 127 30/07/2018 21:19 Chapter 6

or intermittent claudication based on their responses to these questions according to Rose et al 1977. 30

Although Rose questionnaire was not specifically validated for the ethnic groups in this study, it has been shown to work well in other validated studies with similar ethnic group as in our study. 31

Other measurements

Weight (kg) and height (cm) were measured in duplicate in barefoot subjects wearing light clothes only. Waist circumference (cm) was measured twice using a tape measure at the level midway between the lowest rib margin and the iliac crest, and hip circumference (cm) was measured twice at the widest level over the trochanter major. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2), and waist-to-hip ratio (WHR) was calculated as waist circumference divided by hip circumference.

Blood pressure (BP) was measured in duplicate using a semi-automatic sphygmomanometer (Microlife WatchBP Home, Microlife AG, Switzerland) on the left arm in a seated position after the participant had seated for at least 5 minutes. Hypertension was defined as systolic BP ≥ 140 mmHg, or diastolic BP ≥ 90 mmHg, or being on BP lowering medication, or self- reported hypertension.

Fasting blood samples were taken to determine the concentration of glucose by spectrophotometry, using hexokinase as primary enzyme (Roche Diagnostics, Japan). Total cholesterol, triglycerides and high- density lipoprotein (HDL) cholesterol were determined by colorimetric spectrophotometry (Roche Diagnostics, Japan). Low-density lipoprotein cholesterol (LDL) was calculated according to the Friedewald formula. 32 Type 2 diabetes was defined as increased fasting glucose ≥7mmol/L or current use of glucose-lowering medication or self-reported diabetes. Dyslipidemia was defined as total cholesterol (TC) >6.22mmol/L, or high density lipoprotein cholesterol (HDL-C) <1.04mmol/L, or low density lipoprotein cholesterol (LDL-C) >4.14mmol/L, or triglyceride (TG) >1.69mmol/L, 33 or use of lipid-lowering medication.

Educational level was determined using participant’s highest level of education obtained (either in the Netherlands or in the country of

128

Anujuo_binnenwerk_FINAL.indd 128 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

origin). Participants were categorized into those who have never been to school or had elementary schooling only (1st category), those with lower vocational schooling or lower secondary schooling (2nd category), those with intermediate vocational schooling or intermediate/higher secondary education schooling (3rd category), and those with higher vocational schooling or university (4th category). For the current analyses, the first two categories were combined because of small numbers. Occupation was categorised into: elementary, lower, medium, higher and scientific levels depending on the type of work. Shift work was assessed with the item “do you work irregular hours including services during night hours” (yes/no).

Alcohol intake in the past 12 months (yes/no) and smoking status (yes/no/ ex-smoker) were obtained by questionnaire. Habitual physical activity was measured using the SQUASH questionnaire. 34 The SQUASH questions about multiple activities referring to a normal week in the past months. We categorized participants according to the Dutch guideline for physical activity by summing up the number of days per week for each moderate and high intensity activity lasting at least 30 minutes. A total of 5 days or more resulted in participants being categorized as achieving the Dutch norm for physical activity. 34

Data analysis

Baseline data were expressed as percentages, means or median. For the association of short sleep with CVD, we assessed interaction between short sleep and ethnicity. There was no significant interaction between short sleep 6 and ethnicity. The analyses were performed in two steps. In the first step, the association of short sleep duration with prevalent CVD within each ethnic group was analysed using prevalence ratio(s) (PR) with 95% confidence interval (CI), with adjustments for potential confounders (age, gender) and, additionally, conventional CVD risk factors. Although we considered the inclusion of other potential confounders such as depressive symptoms and socio-economic status, we decided not to include these in the multivariate model, because of the risk of overadjustment. For depressive symptoms could also be considered as an intermediary factor in the causal pathway between sleep and CVD, whereas factors such as socioeconomic status and shift work are factors that drive the pattern of short sleep among ethnic groups. In the second step, comparisons of prevalent CVD between ethnic groups was performed with adjustments for age and gender (model 1). In

129

Anujuo_binnenwerk_FINAL.indd 129 30/07/2018 21:19 Chapter 6

order to assess the contribution of short sleep and conventional CVD risk factors to the ethnic differences in CVD, we subsequently added short sleep (model 2), CVD risk factor variables (hypertension, diabetes, BMI, WHR, dyslipidaemia, smoking, alcohol, and physical activity model 3), or both short sleep and CVD risk factors (model 4) to the regression models. The change in PR (percentage) before and after inclusion was used to assess the relative total contribution of short sleep and CVD risk factors to ethnic differences in CVD. By subtracting the changes in PR (percentage) of model 4 and 3, the contribution of short sleep independent of CVD risk factors was calculated for each ethnic minority group. This method of calculation has been used in previous study. 35 All analyses were performed using STATA 11.0 (Stata Corp, College Station, TX). A p-value of <0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

Table 1 shows the characteristics of the study population by ethnic group. Moroccans and Turks were younger, had higher prevalence of CVD, lower educational levels, consumed less alcohol, less often achieved the physical activity norm, and had a lower prevalence of hypertension as compared with the other ethnic origin groups. Similar to Turkish participants, South- Asian Surinamese also had higher prevalence of CVD than the other groups. Ghanaians also had lower educational levels than Dutch. All ethnic minority groups have lower occupational levels compared with Dutch. African Surinamese and South-Asian Surinamese have higher prevalence of shift work compared with other ethnic groups. South-Asian Surinamese, African Surinamese, and Ghanaian participants had a lower mean sleep duration and higher prevalence of short sleep than Dutch, Turks and Moroccans. Dutch and South-Asian Surinamese had a lower mean BMI than the other ethnic groups, whereas mean WHR was higher in South- Asian Surinamese, Ghanaian and Turkish participants as compared with the other ethnic groups. The prevalence of depressive symptoms was higher in Turks, Moroccans and South-Asian Surinamese compared with other ethnic groups. The prevalence of hypertension and diabetes was higher in all ethnic minority groups, while lipid profile was more favourable for the ethnic minority groups, as compared with Dutch.

130

Anujuo_binnenwerk_FINAL.indd 130 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

Association between short sleep duration and cardiovascular disease

Figure 2 shows the crude association between short sleep duration and the prevalence of CVD by ethnicity. The prevalence of CVD was consistently higher in short sleepers than among healthy sleepers in all ethnic groups. This association remained significant after adjustment for potential confounders and conventional CVD risk factors (Table 2).

Contributions of short sleep duration to ethnic differences in the prevalence of CVD

Table 3 shows the ethnic differences in CVD, adjusting for short sleep and CVD risk factors separately and simultaneously. Model 1 shows the PR for each of the ethnic groups adjusting for confounders (age and sex) only. For instance, South-Asian Surinamese were 3.69 times more likely than Dutch to have a CVD. Adjusting for short sleep as can be seen in model 2, the PR reduced to 3.3. This implies that 14 % of the increased prevalence can be explained by short sleep (model 2 compared to model 1: 3.69-3.31/3.69-1 * 100%). The contribution of short sleep (model 2) was lower than that of the all conventional CVD risk factors (24%, model 3). This contribution of short sleep is the sum of the contribution through conventional risk factors and its independent contribution (figure 1). The contribution of short sleep independent of CVD risk factors is indicated by the difference in PR reduction between model 3 and model 4. That is, 10% (calculated as 34% - 24%) is contributed by sleep independently of CVD risk factors, 6 while in total, this was 14% (model 2). This implies that two-third of the total contribution of short sleep was independent of conventional CVD risk factors, and one-third through these risk factors. Similarly, this applies to the other ethnic minority groups as shown in table 3. The independent contribution of short sleep was higher in African Surinamese (15%) and Ghanaians (15%) compared to other ethnic groups ranging from 5% in Turkish to 10% in South-Asian Surinamese.

131

Anujuo_binnenwerk_FINAL.indd 131 30/07/2018 21:19 Chapter 6 Moroccans n = 3687 40.4 (40.0-40.8) 27.6 (27.4-27.7) 38.9 (37.4-40.5) 0.89 (0.89-0.90) 7.0 (6.99, 7.09) 27.6 (26.1-29.0) 27.0 (25.6-28.4) 4.83 (4.14-5.52) 7.35 (6.51-8.19) 18.7 (17.4-19.9) 10.1 (9.14-11.1) 0.87 (0.57-1.17) 15.9 (14.7-17.0) 24.4 (22.9-25.7) 11.4 (10.3-12.4) 11.4 Turks n = 3395 40.4 (39.9-40.8) 28.5 (28.3-28.7) 45.4 (43.7-47.0) 0.90 (0.90-0.91) 6.9 (6.92, 7.04) 40.8 (39.1-42.4) 29.5 (27.9-30.9) 8.84 (7.89-9.79) 10.1 (9.09-11.1) 26.5 (24.9-27.9) 12.2 (11.1-13.3) 1.24 (0.86-1.61) 19.9 (18.6-21.3) 29.1 (27.6-30.4) 10.2 (9.14-11.2) Ghanaians n = 2181 44.7 (44.3-45.2) 28.4 (28.2-28.6) 39.4 (37.3-41.4) 0.90 (0.90-0.91) 6.5 (6.38, 6.55) 20.1 (18.4-21.7) 44.2 (42.2-46.3) 10.5 (9.21-11.8) 5.14 (4.21-6.06) 6.75 (5.69-7.79) 6.69 (5.64-7.74) 1.51 (1.00-2.02) 11.6 (10.3-12.9) 11.6 55.8 (53.4-57.9) 11.4 (10.1-12.8) 11.4 African Surinamese n = 4039 47.9 (47.6-48.3) 27.8 (27.6-28.0) 39.0 (37.5-40.5) 0.89 (0.89-0.90) 6.5 (6.49, 6.58) 24.0 (22.7-25.3) 45.6 (44.0-47.1) 9.73 (8.82-10.6) 5.45 (4.75-6.15) 11.0 (10.0-11.9) 11.0 8.44 (7.58-9.30) 1.02 (0.70-1.32) 12.7 (11.7-13.7) 50.2 (48.7-51.7) 12.0 (11.0-13.0) 45.5 (45.0-46.0) South-Asian Surinamese n = 2933 26.3 (26.1-26.5) 45.3 (43.5-47.1) 0.92 (0.92-0.93) 6.8 (6.70, 6.81) 41.7 (39.9-43.4) 39.4 (37.6-41.1) 11.7 (10.5-12.8) 11.7 8.11 (7.12-9.10) 8.11 22.9 (21.4-24.5) 11.4 (10.3-12.6) 11.4 0.89 (0.55-1.22) 17.0 (15.6-18.3) 42.4 (40.6-44.2) 19.4 (17.9-20.8) 46.2 (45.7-46.6) 24.7 (24.6-24.9) Dutch n = 4495 45.8 (44.4-47.3) 0.88 (0.87-0.88) 7.2 (7.21, 7.27) 29.9 (28.5-31.2) 16.2 (15.1-17.2) 15.3 (14.2-16.3) 1.62 (1.25-1.99) 9.77 (8.90-10.6) 3.20 (2.69-3.72) 0.24 (0.10-0.39) 4.63 (4.01-5.24) 29.5 (28.2-30.9) 3.58 (3.03-4.12)

) 2 Characteristics of the study population by ethnicity Age Men (%) WHR Sleep duration (h) Dyslipidaemia (% yes) Short sleep (% yes) Cholesterol (Total) Cholesterol (Total) >6.22mmol/l Angina (% yes) HDL cholesterol HDL <1.04 mmol/L Myocardial infarction (% yes) Intermittent claudication (% yes) Prevalence CVD (%) Hypertension (% yes) Diabetes (% yes) BMI (kg/m Table 1: Table

132

Anujuo_binnenwerk_FINAL.indd 132 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD Moroccans n = 3687 5.86 (5.10-6.62) 20.3 (18.9-21.6) 48.7 (47.1-50.3) 10.6 (9,59-11.6) 33.3 (31.8-34.9) 17.9 (16.7-19.2) 12.9 (11.9-14.0) 24.8 (23.4-26.2) 21.1 (19.8-22.3) 11.5 (10.5-12.5) 11.5 2.36 (1.87-2.85) 14.7 (13.6-15.9) Turks n = 3395 10.1 (9.09-11.1) 22.8 (21.3-24.2) 55.9 (54.3-57.6) 20.1 (18.8-21.5) 28.7 (27.1-30.2) 15.4 (14.2-16.6) 14.9 (13.7-16.1) 30.5 (28.9-32.1) 18.8 (17.5-20.1) 8.45 (7.52-9.39) 3.21 (2.62-3.80) 14.0 (12.9-15.2) Ghanaians n = 2181 10.6 (9.35-11.9) 9.08 (7.87-10.3) 68.0 (66.1-70.0) 3.35 (2.59-4.10) 25.6 (23.7-27.4) 6.37 (5.34-7.40) 53.7 (51.6-55.8) 20.2 (18.5-21.8) 7.79 (6.67-8.92) 2.57 (1.90-3.23) 0.87 (0.48-1.26) 17.1 (15.5-18.7) African Surinamese n = 4039 10.5 (9.58-11.5) 10.6 (9.67-11.6) 40.8 (39.3-42.3) 5.67 (4.96-6.38) 35.9 (34.4-37.4) 23.2 (21.9-24.5) 6.29 (5.54-7.04) 31.9 (30.4-33.3) 32.3 (30.9-33.8) 17.8 (16.6-19.0) 2.62 (2.13-3.12) 26.8 (25.4-28.1) 14.3 (13.0-15.5) South-Asian Surinamese n = 2933 18.4 (16.9-19.8) 47.8 (46.0- 49.6) 16.9 (15.5-18.2) 29.0 (27.4- 30.7) 23.2 (21.6- 24.7) 9.48 (8.42-10.5) 30.7 (29.0-32.3) 27.4 (25.8-29.0) 16.1 (14.8-17.5) 4.67 (3.91-5.43) 21.9 (20.4-23.4) 6 13.7 (12.7-14.7) Dutch n = 4495 6.99 (6.24-7.73) 17.3 (16.2-18.4) 11.9 (10.9-12.8) 11.9 21.7 (20.5- 22.9) 60.9 (59.5- 62.4) 1.65 (1.27-2.02) 14.1 (13.1-15.1) 21.9 (20.7-23.2) 36.4 (35.0-37.8) 19.9 (18.7-21.0) 19.4 (18.2-20.5) LDL cholesterol LDL >4.14mmol/L Depressive symptoms (%) 1st and 2nd category (%) Triglyceride (TG) Triglyceride <1.69 mmol/L Education 3rd category (%) 4th category (%) Occupation Elementary (%) Lower (%) Medium (%) Higher (%) Scientific (%) Shift work (%) Table 1 - continued Table

133

Anujuo_binnenwerk_FINAL.indd 133 30/07/2018 21:19 Chapter 6 TG, Moroccans n = 3687 73.9 (72.5-75.3) 47.2 (45.6-48.8) 12.7 (11.6-13.8) 7.18 (6.35-8.02) 13.0 (11.9-14.1) Turks n = 3395 47.4 (45.7-49.1) 42.5 (40.8-44.2) 18.3 (17.0-19.6) 22.9 (21.5-24.3) 33.6 (32.0-35.2) Ghanaians n = 2181 86.7 (85.3-88.1) 54.2 (52.1-56.3) 8.30 (7.14-9.46) 48.1 (45.9-50.2) 4.49 (3.62-5.36) African Surinamese n = 4039 48.9 (47.4-50.5) 61.2 (59.7-62.7) 19.3 (18.1-20.5) 68.8 (67.4-70.2) 31.2 (29.8-32.7)

57.9 (56.2-59.7) South-Asian Surinamese n = 2933 53.2 (51.4-54.9) 13.7 (12.5-14.9) 56.7 (54.9-58.5) 27.9 (26.3-29.6) 37.1 (35.7-38.5) Dutch n = 4495 75.5 (74.3-76.8) 38.1 (36.7-39.5) 91.2 (90.4-92.0) 24.6 (23.4-25.9) Smoking status Never smoked (% yes) Achieving Dutch norm for physical activity (%yes) Ex-smoker (% yes) Alcohol intake (%yes) Current smoker (% yes) Table 1 - continued Table Data are presented as means and percentages with 95% CI CVD, cardiovascular disease; BMI, body mass index; WHR, waist to hip ratio; HDL, high density lipoprotein; LDL, low lipoproptein; triglyceride; CI, confidence interval.

134

Anujuo_binnenwerk_FINAL.indd 134 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD Moroccans n = 3687 PR (95%CI) 1.44 (1.24-1.68)* 1.39 (1.19-1.63)* 1.41 (1.21-1.65)* Turks n = 3395 PR (95%CI) 1.56 (1.36-1.79)* 1.51 (1.32-1.73)* 1.46 (1.27-1.68)* Ghanaians n = 2181 PR (95%CI) 1.46 (1.16-1.85)** 1.54 (1.22-1.95)* 1.55 (1.22-1.97)* African Surinamese n = 4039 PR (95%CI) 1.52 (1.29-1.79)* 1.56 (1.32-1.84)* 1.51 (1.28-1.78)*

South-Asian Surinamese n = 2933 PR (95%CI) 1.67 (1.42-1.96)* 1.59 (1.35-1.87)* 1.53 (1.31-1.79)* 6 Dutch n = 4495 PR (95%CI) 2.05 (1.54-2.75)* 1.84 (1.38-2.46)* 1.62 (1.20-2.18)** Prevalence ratio(s) for the relationship between sleep duration (short sleep versus healthy sleep) and the prevalence of cardiovascular relationship the between sleep duration (short versus healthy sleep) and the prevalence of Prevalencefor ratio(s) Short sleep Crude Model 1 Model 2 Table 2: Table disease by ethnicity Model 1: adjusted for age and sex 2: model 1 plus BMI, WHR, hypertension, diabetes, dyslipidaemia, smoking, alcohol consumption and physical activity (*: p < 0.001, **: 0.05) PR, prevalence ratio; CI, confidence interval.

135

Anujuo_binnenwerk_FINAL.indd 135 30/07/2018 21:19 Chapter 6 Z Difference in Difference reduction (%) 10 15 15 5 5 Difference in (%) reduction between model 4 and 3 Y

34 39 34 29 15 Reduction PR total short sleep, total CVD risk factors (%) 1.00 2.78 (2.36-3.28)* 2.00 (1.70-2.36)* 2.02 (1.67-2.43)* 3.57 (3.02-4.23)* 3.23 (2.72-3.84)* PR (95% CI) Confounders + short sleep, CVD risk factors (model 4)

J

24 24 19 24 10 Reduction PR total CVD risk factors (%) 1.00 3.05 (2.59-3.59)* 2.24 (1.91-2.63)* 2.24 (1.85-2.69)* 3.76 (3.18-4.45)* 3.37 (2.84-4.00)* PR (95% CI) Confounders + CVD risk factors (model 3)

X

Reduction PR total short sleep (%) 14 19 20 9 8 Confounders + short sleep (model 2) PR (95% CI) 1.00 3.31 (2.83-3.87)* 2.32 (1.98-2.72)* 2.23 (1.87-2.67)* 4.31 (3.71-5.00)* 3.40 (2.92-3.96)* PR (95% CI) 1.00 3.69 (3.16-4.31)* 2.64 (2.26-3.08)* 2.54 (2.13-3.02)* 4.64 (4.00-5.39)* 3.63 (3.12-4.22)* Prevalence ratio(s) for ethnic differences in prevalence CVD, adjusting for short sleep and CVD risk factors separately, and simultaneously. separately, short sleep and CVD risk factors adjusting for in prevalence CVD, ethnic differences for Prevalence ratio(s) Confounders (model 1) Ethnic group Dutch South-Asian Surinamese African Surinamese Ghanaians Turks Moroccans Table 3: Table Model 1: Confounders: Adjusted for age and gender. Model 2: Adjusted for age, gender and short sleep. Model 3: Adjusted for age, gender and Adjusted for age, gender and short sleep. Model 3: Model 2: Adjusted for age and gender. Model 1: Confounders: AND short sleep. (*: p < 0.001) CVD risk factors Adjusted for age, gender, CVD risk factors. Model 4: CVD, cardiovascular disease; CI, confidence interval. X: Contribution of short sleep: Reduction in PR (%) between model 1 and 2 J: Contribution of CVD risk factors: Reduction in PR (%) between model 1 and 3 Contribution of both CVD risk factors and short sleep: Reduction in PR (%) between model 1and 4 Y: in PR reduction (%) between model 3 and 4 Z: contribution of short sleep independent CVD risk factors; Difference

136

Anujuo_binnenwerk_FINAL.indd 136 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

Fig. 1: Research model by which short sleep and CVD risk factors may lead to CVD in various ethnic groups

Fig. 2: Crude association between sleep duration and prevalence of cardiovascular disease among ethnic groups in Amsterdam 6

137

Anujuo_binnenwerk_FINAL.indd 137 30/07/2018 21:19 Chapter 6

DISCUSSION

In this study, we investigated the contribution of short sleep to ethnic inequalities in CVD prevalence, and to what extent short sleep affects these inequalities, independent of conventional CVD risk factors. Our study findings indicate that, compared to healthy sleep, short sleep duration was consistently associated with an increased prevalence of CVD in all ethnic groups. Short sleep contributed substantially to ethnic differences in prevalence of CVD. In African Surinamese and Ghanaians, the contribution of short sleep was almost comparable to the contribution of all conventional CVD risk factors combined. In addition, short sleep contributed mostly independent of these risk factors to ethnic inequalities in CVD. The independent contribution and total contribution of sleep was higher in African Surinamese and Ghanaians than in South-Asian Surinamese, Turks and Moroccans.

Our study indicated that short sleep was consistently associated with CVD in all ethnic groups after adjustment for age, gender, and also after conventional CVD risk factors had been controlled for. Amagi et al. (2010) 12 and Meisinger et al. (2007) 13 also demonstrated a higher risk of myocardial infarction in short sleepers compared to healthy sleepers. The MORGEN study, investigating sleep duration and sleep quality in relation to 12- year CVD incidence, 14 alongside other studies, 15, 16 also found that short sleep was associated with CVD. Our cross-sectional study results are thus consistent with these findings. In addition, our result concurs with a recent multi-ethnicity study, which found that short sleep was associated with prevalent CVD (angina, MI and stroke) in Hispanics, African Americans, and other ethnicities in the U.S. 19 The influence of sleep on CVD is partly mediated through conventional risk factors, such as obesity. In particular, biological mechanisms involving endocrinological and metabolic functions have been proposed for the association between short sleep and CVD. Sleep deprivation may trigger inflammatory processes and activation of the sympathetic nervous system, increasing BP and cortisol levels, and impaired glucose tolerance, 36-40 and therefore may lead to an increased risk of CVD. It has also been suggested that short sleep promotes increased levels of ghrelin and reduced leptin levels, which are hormones involved in appetite and satiety regulation. Imbalance in these hormones increases food ingestion, leading to obesity. 41

In the second part of our study, that examined the contribution of short

138

Anujuo_binnenwerk_FINAL.indd 138 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

sleep to the ethnic differences in prevalence of CVD, we found that short sleep significantly accounted for ethnic differences in CVD prevalence, with most of the contribution being independent of known CVD risk factors. The independent contribution of short sleep was particularly marked in African Surinamese, Ghanaian, and South-Asian Surinamese (15%, 15% and 10%) compared with Turks and Moroccans (5% and 5%). The reason that short sleep contributed more in African Surinamese, Ghanaian, and South-Asian Surinamese than in Turks and Moroccans is likely to be due to higher prevalence of short sleep in these groups and its accompanying negative consequences compared to Turks and Moroccans. We noticed that large proportions of the difference in CVD between Dutch and ethnic minority groups were not explained by conventional CVD risk factors and short sleep, suggesting that other unmeasured factors may be at play.

The mechanism underlying the contribution of short sleep to differences in CVD prevalence independently of conventional CVD risk factors is not clear. However, the mechanism may be through allostatic load, involving the stress response system, 11 or through depressive symptoms. Short sleep results in the activation of the neuroendocrine stress response leading to increased sympathetic activity. 42 that involves sympathoadrenal system and the hypothalamic-pituitary-adrenal (HPA) axis, which might also directly influence the risk of CVD. Although this is partly through CVD risk factors, it may also be independent of these risk factors. This activation results in increased levels of glucocorticoids, notably cortisol and catecholamine which play an important role in regulation of energy balance and in cardiovascular function. Catecholamines can accelerate disease process through allostatic load when the activation lasts over a long period of time. According to 6 this mechanism, disease manifest as a price, which the body pays for being forced to adapt to adverse psychosocial or physical situations. Allostatic load, represents either the presence of too much stress, or the inefficient operation of the stress hormone response system to adjust properly after the occurrence of the stressful event. 11 Future studies should further investigate other factors that may influence the mechanism of allostatic load in the contribution of short sleep to ethnic differences in CVD.

The strength of our study lies in utilization of large sample sizes, permitting more reliable estimations. Also, multiple ethnic groups living in one city were investigated together using the same methodology. In addition, three CVD endpoints were combined ensuring more reliable results. A limitation of the study is that we only used self-reported data on CVD and sleep, which may

139

Anujuo_binnenwerk_FINAL.indd 139 30/07/2018 21:19 Chapter 6

be subject to recall bias; hence the participants may have under -or over- reported short sleep durations and CVD. Also, the conventional CVD risk factors (health behavior) used in the study were measured only once, and therefore may not have been measured accurately. Stringhini et al. (2010) 43 have demonstrated a larger contribution of health behavior in explaining inequalities when health behaviors were measured longitudinally. Hence, once the measurements are followed up, a greater percentage of CVD may be explained. Being a cross-sectional study, causal associations between short sleep and CVD could not be established. Because important associations of short sleep and CVD were observed among the ethnic groups, further longitudinal studies are required among these populations. Furthermore, information on sleep quality, sleep hours during weekends and on daytime sleepiness, use of sleep hypnotics, antidepressant medications and insomnia, which may affect sleep duration, were lacking and should be considered in future studies. Also, sleep apnea has been shown to be associated with CVD and cardiorespiratory problems. 44,45 and causes hemodynamic changes and significant sleep disturbances. 46,47 Sleep apnea may confound the association between sleep duration and CVD. However, we did not investigate sleep apnea in this study as we do not have the information in our dataset. We also note that inaccurate measures of conventional behaviour risk factors of CVD may affect the obtained results. In other words, residual confounding cannot be excluded and might have led to overestimating the independent contribution of sleep. Another important factor which may play a role in the association of sleep duration with CVD include genetics. 48 However, information on genetics was not available in our dataset, and was not investigated. Despite the lack of data on these factors, we were still able to answer our key research question.

In conclusion, our study showed that short sleep duration was related to the prevalence of CVD across all ethnic groups, and short sleep, independent of conventional CVD risk factors contributed significantly to ethnic differences in CVD. The findings might suggest that reducing sleep deprivation may be a relevant entry point for reducing increased CVD risks among the various ethnic minority groups.

140

Anujuo_binnenwerk_FINAL.indd 140 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

6

141

Anujuo_binnenwerk_FINAL.indd 141 30/07/2018 21:19 Chapter 6

REFERENCES

1. Mensah GA, Mokdad AH, Ford ES, et al. State of disparities in cardiovascular health in the United States. Circulation. 2005;111:123-41. 2. Scarborough P, Bhatnagar P, Kaur A, et al. Ethnic differences in cardiovascular disease. http://www.bhf.org.uk/~/media/files/research/heart-statistics/hs2010fc ethnic differences in cardiovascular disease-full-copy.pdf. Assessed on December 9, 2015. 3. Agyemang C, Van Oeffelen AA, Norredam M,et al. Ethnic disparities in ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage incidence in the Netherlands. Stroke. 2014;11:3236-42. 4. Chaturvedi N. Ethnic differences in cardiovascular disease. Heart. 2003;6:681-6. 5. Wolk R, Gami AS, Garcia-Touchard A, et al. Sleep and cardiovascular disease. Curr Probl Cardiol. 2005;30:625–62. 6. Taheri S, Lin L, Austin D, et al. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004;3: e62. 7. Spiegel K, Knutson K, Leproult R, et al. Sleep loss: a novel risk factor for insulin resistance and Type 2 diabetes. J Appl Physiol. 2008;5:2008-19. 8. Cappuccio FP, Stranges S, Kandala NB, et al. Gender-specific associations of short sleep duration with prevalent and incident hypertension: the Whitehall II Study. Hypertension. 2007;4:693-700. 9. Schwartz J, Allison MA, Ancoli-Isreal S, et al. Sleep, type 2 diabetes, dyslipidemia, and hypertension in elderly Alzheimers caregivers. Arch Gerontol Geriatr. 2013;1:70-77. 10. Gangwisch JE, Malaspina D, Babiss LA, et al. Short sleep duration as a risk factor for hypercholesterolemia: analyses of the National Longitudinal Study of Adolescent Health. Sleep. 2010;7: 956–961. 11. McEwens BS. Allostasis and allostatic load: implications for neuropsychopharmacology. Neuropsychopharmacology. 2000;2:108-24. 12. Amagai Y, Ishikawa S, Gotoh T, et al. Sleep duration and incidence of cardiovascular events in a Japanese population: the Jichi Medical School cohort study. J Epidemiol. 2010;2:106-10. 13. Meisinger C, Heier M, Löwel H, et al. Sleep duration and sleep complaints and risk of myocardial infarction in middle-aged men and women from the general population: the MONICA/KORA Augsburg cohort. Sleep. 2007;9:1121-7. 14. Hoevenaar-Blom MP, Spijkerman AM, Kromhout D, et al. Sleep duration and sleep quality in relation to 12-year cardiovascular disease incidence: the MORGEN study. Sleep. 2011;11:1487-92.

142

Anujuo_binnenwerk_FINAL.indd 142 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

15. Von Ruesten A, Weikert C, Fietze I, et al. Association of sleep duration with chronic diseases in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Postdam study. PLoS One. 2012;1: e30972. 16. Aggarwal S, Loomba RS, Arora RR, et al. Association between sleep duration and prevalence of cardiovascular events. Clin Cardiol. 2013;11: 671-6. 17. Qureshi AI, Giles WH, Croft JB, et al. Habitual sleep patterns and risk for stroke and coronary heart disease: a 10-year follow-up from NHANES I. Neurology. 1997;48:904–11. 18. Chen JC, Brunner RL, Ren H, et al. Sleep duration and risk of ischemic stroke in postmenopausal women. Stroke. 2008;39:3185–92. 19. Sabanayagam C, Anoop Shankar. Sleep duration and cardiovascular disease: results from the National Health Interview Survey. Sleep. 2010;8:1037–42. 20. Agyemang C, Kieft S, Snijder MB, et al. Hypertension control in a large multi-ethnic cohort in Amsterdam, The Netherlands: the HELIUS study.Int J Cardiol. 2015;183:180-9. 21. Anujuo K, Stronks K, Snijder MB, et al. Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS study. Sleep Med. 2014;9:1115- 21. 22. Anujuo K, Stronks K, Snijder MB, et al. Relationship between short sleep duration and cardiovascular risk factors in a multi-ethnic cohort – the HELIUS study. Sleep Med. 2015;12:1482-8. 23. Chambers EC, Pichardo MS, Rosenbaum E. Sleep and the housing and neighborhood environment of urban Latino adults living in low-income housing: The AHOME Study. Behav Sleep Med. 2016;14:169-8. 24. Stronks K, Snijder MB, Peters RJ, et al. Unravelling the impact of ethnicity on health in Europe: the HELIUS study. BMC Public Health. 2013;13: 402. 25. Stronks K, Kulu-Glasgow I, Agyemang C. The utility of ‘country of birth’ for 6 the classification of ethnic groups in health research: the Dutch experience. Ethn Health. 2009;14: 255-69. 26. Watson NF, Badr MS, Belenky G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;6: 843–4. 27. Grandner MA, Chakravorty S, Perlis ML, et al. Habitual sleep duration is associated with self-reported and objectively determined cardiometabolic risk factors. Sleep Med. 2014;1: 42-50. 28. Guo X, Zheng L, Wang J, et al. Epidemiological evidence for the link between sleep duration and high blood pressure: A systematic review and meta-analysis. Sleep Med. 2013;4: 324-32. 29. Cappucio FP, Taggart FM, Kandala NB, et al. Meta-analysis of short sleep duration and obesity in children and adults. Sleep. 2008;5: 619-26.

143

Anujuo_binnenwerk_FINAL.indd 143 30/07/2018 21:19 Chapter 6

30. Rose G, McCartney P, Reid DD. Self-administration of a questionnaire on chest pain and intermittent claudication. Br J Prev Soc Med. 1977;31: 42–48. 31. Fischbacher CM, Bhopal R, Unwin N, et al. The performance of the Rose angina questionnaire in South-Asian and European origin populations: a comparative study in Newcastle, UK. Int J Epidemiol. 2001;30:1009-16. 32. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18: 499-502. 33. Eckel RH, Alberti KG, Grundy SM, et al. The metabolic syndrome. Lancet. 2010;375: 181-3. 34. Wendel-Vos GC, Schuit AJ, Saris WH, et al. Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity. J Clin Epidemiol. 2003;12: 1163-9. 35. Moor I, Rathmann K, Stronks K, et al. Psychosocial and behavioural factors in the explanation of socioeconomic inequalities in adolescent health: a multilevel analysis in 28 European and North American countries. J Epidemiol Community Health. 2014;68:912-21. 36. Gottlieb DJ, Redline S, Nieto FJ, et al. Association of usual sleep duration with hypertension: the Sleep Heart Health Study. Sleep. 2006;29:1009–14. 37. Mullington JM, Haack M, Toth M, et al. Cardiovascular, inflammatory, and metabolic consequences of sleep deprivation. Prog Cardiovasc Dis. 2009;51: 294–302. 38. Meier-Ewert HK, Ridker PM, Rifai N, et al. Effect of sleep loss on C-reactive protein, an inflammatory marker of cardiovascular risk.J Am Coll Cardiol. 2004;43: 678–83. 39. Leproult R, Copinschi G, Buxton O, et al. Sleep loss results in an elevation of cortisol levels the next evening. Sleep. 1997;20: 865–70. 40. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet. 1999;354:1435–9. 41. Knutson KL, Van Cauter E. Associations between sleep loss and increased risk of obesity and diabetes. Ann N Y Acad Sci. 2008;1129: 287–304. 42. Cappuccio FP, Cooper D, D’Elia L, et al. Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. Eur Heart J. 2011;32:1484–92. 43. Stringhini S, Sabia S, Shipley M, et al. Association of socioeconomic position with health behaviors and mortality. JAMA 2010; 12:1159-66. 44. Jackson CL, Redline S, Emmons KM. Sleep as a potential fundamental contribution to cardiovascular health disparieties. Annu Rev Public Health. 2015;36:417-440.

144

Anujuo_binnenwerk_FINAL.indd 144 30/07/2018 21:19 Contribution of short sleep to ethnic differences in CVD

45. Netzer NC, Hoegel JJ, Loube D, et al. Prevalence of symptoms and risk of sleep apnea in primary care. Sleep in Primary Care International Study Group. Chest. 2003;124:1406-14. 46. Pinto JM, Garpestad E, Weiss JW, et al. Hemodynamic changes associated with obstructive sleep apnea followed by arousal in a porcine model. J Appl Physiol. 1993; 75:1439-43. 47. Findley LJ, Weiss JW, Jabour ER. Drivers with untreated sleep apnea. A cause of death and serious injury. Arch Intern Med. 1991; 51:1451-2. 48. Grandner MA, Sands-Lincoln MR, Pak VM, et al. Sleep duration, cardiovascular disease, and proinflammatory biomarkers. Nat Sci Sleep. 2013; 5: 93–107.

6

145

Anujuo_binnenwerk_FINAL.indd 145 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 146 30/07/2018 21:19 CHAPTER 7

ASSOCIATION BETWEEN DEPRESSED MOOD AND SLEEP DURATION AMONG VARIOUS ETHNIC GROUPS – THE HELIUS STUDY

Submitted

Kenneth Anujuo, Karien Stronks, Marieke B. Snijder, Anja Lok, Girardin Jean-Louis, Charles Agyemang. Association between depressed mood and sleep duration among various ethnic groups – the HELIUS study

Anujuo_binnenwerk_FINAL.indd 147 30/07/2018 21:19 Chapter 7

ABSTRACT

Background: Evidence has shown increased prevalence of both depression and short sleep among ethnic minority groups compared to the European host populations. However, it is unclear to what extent depression accounts for the increased prevalence of short or long sleep duration among different ethnic groups.

Objective: We analysed the association between depressed mood and short or long sleep duration in a multi-ethnic population living in Amsterdam, the Netherlands. In addition, we quantified the extent to which depressed mood accounts for the short and long sleep duration among six ethnic groups.

Methods: 21072 participants (aged 18-71 years) of the HELIUS study of Dutch, Ghanaian, Turkish, Moroccan, South-Asian Surinamese and African Surinamese ethnic origin were analyzed. Self-reported sleep duration was classified as: short (<7 hours/night), healthy (7-8 hours/night), and long (>=9 hours/night). We used the Patient Health Questionnaire (PHQ-9) to measure depressed mood (PHQ-9 sumscore ≥10). The association between depressed mood and sleep duration was assessed using logistic regression analysis. The extent to which depressed mood accounted for short and long sleep duration was assessed using population attributable fraction (PAF).

Results: Depressed mood was associated with short sleep duration in all ethnic groups with age and sex adjusted odd ratios (AOR) ranging from 1.9 (95% CI 1.4-2.6) in Ghanaians to 2.8 (2.2-3.7) in Dutch. Depressed mood was not associated with long sleep duration except in Dutch (AOR 2.5; 1.9-3.6) and Moroccans (AOR 1.3; 1.0-1.6), respectively. Depressed mood accounted for part of the prevalence of short sleep: 7.7% (Dutch), 9.7% (South-Asian Surinamese), 4.8%, 3.4%, 16.1% and 14.3% in African Surinamese, Ghanaian, Turkish and Moroccan participants, respectively. For long sleep, these percentages were 9.1% and 4.9% in Dutch and Moroccans, respectively.

Conclusion: Depressed mood was associated with short sleep in all ethnic groups, but not with long sleep, except in Dutch and Moroccans. In ethnic minority groups with a high prevalence of depressed mood, this accounted for up to 16% of the prevalence of short sleep. If confirmed in longitudinal analyses, these results may suggest that strategies aimed to reduce depression may reduce the prevalence of short sleep, especially in those ethnic minority groups with a high prevalence of depression, such as the Turkish and Moroccan groups in the Netherlands.

148

Anujuo_binnenwerk_FINAL.indd 148 30/07/2018 21:19 Association between depressed mood and sleep duration

INTRODUCTION

Depression is a complex mental disorder that may cause disability and morbidity in the general population. Depression cause more disability than any other disorder, and it is estimated to affect about 350 million people across the globe [1, 2]. Major depressive disorder (MDD) is one of the most prevalent psychiatric disorders and is estimated to be the leading cause of disease burden in high-income countries by 2030 [3].

A previous longitudinal study conducted in the United States of America (USA) has shown that adolescents with symptoms of depression at baseline, had higher rates of sleep problems at 4-year follow-up than did those who had not experienced depressive symptoms at baseline [4]. This suggests that depressive symptoms may influence sleep disruption. On the other hand, studies have also shown that aberrant sleep duration was associated with depressive symptoms [5, 6], and yet other studies have demonstrated a bi- directional relationship between sleep duration and depression [7, 8, 9]. In addition, sleep problems and depression may also share common risk factors and biological features and the two conditions may respond to some of the same treatment strategies [10].

Thus, the association between depression and sleep at least partly reflects an effect of depression on sleep, for instance rumination due to depression may precipitate sleep problems. This paper focuses on the association between depression and sleep duration. Among ethnic minority groups in the Netherlands, there is high prevalence of both depression and short sleep compared to the Dutch host population [11, 12, 13]. However, the extent to which depression influences sleep duration across different ethnic groups has not yet been elucidated. In the current study, we examined the association between depressed mood 7 and sleep duration among six ethnic groups. In addition, we quantified to what extent depressed mood accounted for the prevalence of short and long sleep duration in each ethnic group.

149

Anujuo_binnenwerk_FINAL.indd 149 30/07/2018 21:19 Chapter 7

STUDY POPULATION AND METHODS

Study population

The current study was based on baseline data from the HELIUS (Healthy Life in an Urban Setting) study. The aims and design of the HELIUS study have been described elsewhere [14, 15]. In brief, HELIUS is a large-scale cohort study on health and health care among different ethnic groups living in Amsterdam. The study includes individuals aged 18-70 years from the six major ethnic groups in Amsterdam (African-Surinamese, South-Asian Surinamese, Turkish, Moroccan, Ghanaian, and Dutch origin), and focuses on three major disease categories: cardiovascular disease, mental health and infectious diseases. Participants were randomly sampled from the municipal registers, stratified by ethnicity. The study protocols were approved by the AMC Ethical Review Board. All participants provided written informed consent.

Baseline data collection took place in 2011-2015. Data from both questionnaire and the physical examination were available in 22165 participants. For current analyses, participants with Javanese Surinamese origin (n=233), other/unknown Surinamese origin (n= 267) and other/ unknown origin (n=48) were excluded from the analysis because of small sample sizes, resulting in 21617 participants. Subsequently, individuals with missing data on sleep duration (n=334) or depressed mood (n=211) were excluded from the analysis. This resulted in a dataset of 21072 participants, including 4540 Dutch, 2992 South-Asian Surinamese, 4052 African Surinamese, 2192 Ghanaians, 3494 Turks, and 3802 Moroccans.

Ethnicity

Ethnicity was defined according to the country of birth of the participant as well as that of parents. Specifically, a participant was considered as of non-Dutch ethnic origin if he/she fulfils either of the following criteria: 1) he or she was born abroad and has at least one parent born abroad (first generation); or 2) he or she was born in the Netherlands but both his/her parents born abroad (second generation) [16]. Of the Surinamese migrants in the Netherlands, approximately 80% were either African or South-Asian origin. Both subgroups were classified according to self-reported ethnic origin. Participants were considered as of Dutch origin if the person and both parents were born in the Netherlands.

150

Anujuo_binnenwerk_FINAL.indd 150 30/07/2018 21:19 Association between depressed mood and sleep duration

Depressive symptoms

Depressive symptoms were measured using the Patient Health Questionnaire-9 (PHQ -9) [17]. PHQ-9 determines depressive symptoms over the preceding 2 weeks. A validity study within the HELIUS study indicated that the PHQ-9 measures the same concept in among ethnic groups and that there are no systematic differences in reporting between the ethnic groups [18]. The PHQ-9 consists of nine items, with a response scale varying from 0 (never) to 3 (nearly every day). If one of the items was missing, the mean score of the other eight items was used to replace the missing item. If more than one item was missing, the variable was considered missing. A cut-off point for the sumscore (ranging from 0 to 27) of ≥10 was used to determine depressed mood [12].

Sleep duration

Participants were asked to provide information on the average number of hours they usually sleep at night. Sleep duration was assessed using the item, “How many hours do you sleep on average per night?”. Short sleep was defined as having less than 7 hours of sleep per night, healthy sleep was defined as having 7-8 hours of sleep per night and long sleep defined as having 9 or more hours of sleep per night, in line with National Sleep Foundation (NSF), American Academy of Sleep Medicine (AASM), and Sleep Research Society (SRS) recommendations [19]. Sleep duration was measured in full hour increments.

Other Measurements

Educational level was determined using participant’s highest level of education obtained (either in the Netherlands or in the country of 7 origin). Participants were categorized into those who have never been to school or had elementary schooling only (1st category), those with lower vocational schooling or lower secondary schooling (2nd category), those with intermediate vocational schooling or intermediate/higher secondary education schooling (3rd category), and those with higher vocational schooling or university (4th category). For the current analyses, the first two categories were combined because of small numbers of these categories among Dutch. Alcohol intake in the past 12 months (yes/no) and current smoking status (yes/no) were also obtained by questionnaire.

151

Anujuo_binnenwerk_FINAL.indd 151 30/07/2018 21:19 Chapter 7

Data Analysis

Characteristics of the study population, by ethnicity, were expressed as percentages or means with 95% confidence intervals. The relationship between depressed mood (independent variable) and short sleep duration (outcome variable) was analysed by ethnicity by the use of logistic regression with adjustments for potential confounders (age and sex), and expressed as odds ratios (ORs) with 95% confidence intervals (CI). Effect modification by ethnicity was tested by adding interaction terms (ethnic groups*depression) to these regression models. The contribution of depression to sleep duration was calculated using population attributable fraction (PAF) for each ethnic group as proposed by Williamson 2010 and Rockhill 1998 [20, 21]. The PAF was interpreted as the proportional reduction in the prevalence of short or long sleep duration that would occur, if ideally no one has depression. All analysis were performed using STATA 14.2 (Stata Corp, College Station, TX).

RESULTS

Characteristics of the study population

Table 1 shows the characteristics of the study population by ethnicity. Moroccan and Turkish participants were younger, less often consumed alcohol, had lower educational levels, and had a higher prevalence of long sleep and depressed mood, compared with the other ethnic groups. Similar to Turkish and Moroccan participants, South-Asian Surinamese also had a higher prevalence of depressed mood compared with the other ethnic groups. The prevalence of depressed mood was about two times higher in South-Asian Surinamese, and three times higher in Turks and Moroccans compared to Dutch. Ghanaians also had lower educational levels than Dutch. South-Asian Surinamese, African Surinamese and Ghanaian participants had a lower mean sleep duration and higher prevalence of short sleep than Dutch, Turks and Moroccans. Ghanaian and Moroccan groups had the lowest percentage of smokers, whereas the Turkish group had the highest.

152

Anujuo_binnenwerk_FINAL.indd 152 30/07/2018 21:19 Association between depressed mood and sleep duration

Association between depressed mood and short sleep duration.

Figure 1 shows the prevalence of short sleep duration by ethnicity and the presence of depressed mood. There was significant interaction between ethnic groups and depression (p<0.000). The prevalence of short sleep was consistently higher in participants who had depressed mood than those who did not have depressed mood, in all ethnic groups. The observed association remained significant after adjustment for age and sex, with ORs ranging from 1.9 to 2.8 (Table 2).

Association between depressed mood and long sleep duration.

Figure 2 shows the prevalence of long sleep duration by ethnicity and depressed mood. The prevalence of long sleep was higher in participants who had depressed mood than among those who did not have depressed mood, in all ethnic groups. However, the observed association was no longer significant after adjustment for age and sex, except in Dutch (AOR 2.5; 1.8-3.6) and in Moroccans (AOR 1.3; 1.0-1.7 ) (Table 2).

The extent to which depressed mood accounts for prevalence of short and long sleep duration among ethnic groups

Using PAF, we calculated to what extent depressed mood accounted for short and long sleep duration in each ethnic group (Table 2). For short sleep, the PAF in Dutch was 7.7% (95% CI 5.4-9.9), suggesting that the prevalence of short sleep would be reduced by 7.7%, if ideally no one among them 7 had depressed mood. Similarly, among the ethnic minority groups, the PAF ranged from 3.4% in Ghanaians to 16.1% in Turkish people, respectively. For long sleep, the PAFs for Dutch and Moroccans were 9.1% and 4.9% respectively.

153

Anujuo_binnenwerk_FINAL.indd 153 30/07/2018 21:19 Chapter 7 48.9 (47.3-50.5) 40.4 (40.0-40.8) 33.6 (32.1-35.1) 38.9 (37.4-40.5) Moroccans 17.5 (16.4-18.8) 7.2 (7.12, 7.21) n = 3802 13.5 (12.4-14.6) 25.9 (24.5-27.3) 7.33 (6.55-8.20) 10.7 (9.79-11.7) 20.8 (19.5-22.1) 56.4 (54.8-58.0) 40.4 (39.9-40.8) Turks 28.5 (27.1-30.0) 45.4 (43.7-47.0) 15.0 (13.9-16.3) 7.1 (7.10, 7.19) n = 3494 34.2 (32.6-35.8) 28.2 (26.8-29.7) 22.6 (21.3-24.0) 10.2 (9.29-11.3) 23.1 (21.7-24.5) 68.5 (66.5-70.3) 44.7 (44.3-45.2) Ghanaians 25.3 (23.5-27.1) 39.4 (37.3-41.4) 6.27 (5.33-7.36) 6.8 (6.71, 6.83) n = 2192 4.54 (3.75-5.48) 42.9 (40.9-45.0) 47.7 (46.1-50.2) 7.25 (6.25-8.40) 9.0 (7.9-10.2) 41.3 (39.7-42.8) 47.9 (47.6-48.3) African Surinamese 35.8 (34.3-37.2) 39.0 (37.5-40.5) 22.9 (21.7-23.4) 6.6 (6.6, 6.7) n = 4052 31.4 (29.9-32.8) 45.0 (43.5-46.5) 68.1 (66.7-69.5) 4.60 (3.99-5.28) 10.7 (9.79-11.7) 47.8 (45.9- 49.5) 45.5 (45.0-46.0) South-Asian Surinamese 29.2 (27.6- 30.9) 45.3 (43.5-47.1) n = 2992 22.9 (21.5- 24.5) 6.8 (6.79, 6.89) 28.4 (26.8-30.0) 38.3 (36.6-40.1) 56.4 (54.6-58.2) 6.61 (5.77-7.55) 18.5 (17.1-19.9) Dutch 17.5 (16.4-18.6) 46.2 (45.7-46.6) 21.4 (20.6- 23.1) 45.8 (44.4-47.3) n = 4540 60.7 (59.2- 62.1) 7.3 (7.24, 7.29) 24.7 (23.4-25.9) 16.0 (14.9-17.1) 90.9 (90.1-91.8) 5.7 (5.1-6.4) 7.2 (6.5-8.0) Characteristics of the study population by ethnicity Age (years) 1st and 2nd category (%) 3rd category (%) Men (%) 4th category (%) Sleep duration (hours) Current smoker (% yes) Short sleep (% yes) Alcohol intake (% yes) Long sleep (% yes) Depressed mood (%yes) Educational level Table 1: Table Data presented as means or percentages with 95% confidence interval (CI)

154

Anujuo_binnenwerk_FINAL.indd 154 30/07/2018 21:19 Association between depressed mood and sleep duration 4.9 (-0.1, 9.9) OR (95%CI) 1.3 (1.0-1.6) n = 3802 14.3 (11.3-17.2) Moroccans 2.4 (2.0-2.8) 2.6 (-2.9, 8.0) 1.1 (0.9-1.5) OR (95%CI) n = 3494 16.1 (12.9-19.2) Turks 2.5 (2.1-2.9) 2.1 (-3.3, 7.3) 1.2 (0.7-2.1) OR (95%CI) n = 2192 3.4 (1.9-5.0) Ghanaians 1.9 (1.4-2.6) 0.6 (-4.7, 5.6) 1.0 (0.7-1.7) OR (95%CI) 4.8 (3.6-6.0) n = 4052 African Surinamese 2.3 (1.9-2.8) 1.1 (-5.9, 7.7) 1.1 (0.7-1.5) OR (95%CI) 9.7 (7.5-11.9) n = 2992 South-Asian Surinamese 2.3 (1.9-2.8) 7 9.1 (4.5-13.5) 2.5 (1.9-3.6) OR (95%CI) 7.7 (5.4-9.9) n = 4540 2.8 (2.2-3.7) Dutch The association (expressed as Odd ratio (OR)) of depressed mood with the prevalence of short sleep and with the prevalence of long PAF Long sleep (>=9 h/night) Age-sex adjusted PAF Short sleep (<7 h/night) Age-sex adjusted Table 2: Table sleep duration, by ethnicity

155

Anujuo_binnenwerk_FINAL.indd 155 30/07/2018 21:19 Chapter 7

Fig. 1: Association between depressed mood and short sleep duration among six ethnic groups living in Amsterdam

Fig. 2: Association between depressed mood and long sleep duration among six ethnic groups living in Amsterdam

156

Anujuo_binnenwerk_FINAL.indd 156 30/07/2018 21:19 Association between depressed mood and sleep duration

DISCUSSION

In this study, we investigated whether the higher prevalence of depressed mood in ethnic minority groups may account for the higher prevalence of short sleep duration or long sleep duration among these groups. Depressed mood was consistently associated with short sleep in all ethnic groups accounting for a moderate part (16.1% and 14.3%) of the prevalence of short sleep in Turks and Moroccans, but little in other ethnic minority groups. Depressed mood was not associated with long sleep, except in Dutch and Moroccans in whom depressed mood accounted for a moderate part of the prevalence of long sleep (9.1% in Dutch and 4.9% in Moroccans).

To our knowledge, this is the first study that examined the association between depressed mood and sleep duration, and the role of depressed mood in the prevalence of short or long sleep among various ethnic groups. Patten et al (2000) [4], in a previous population-based longitudinal study conducted in the USA, demonstrated that adolescents with symptoms of depression at baseline, had increased rate of sleep problems at the 4-year follow-up than did those who had not experienced depressive symptoms at baseline. In a Norwegian-based study, Silvertsen et al (2014) [9] also reported that depression was cross-sectionally associated with sleep duration among adolescents. Our cross-sectional results partly concur with these findings.

Potential underlying mechanisms for the association between depression and sleep duration have been proposed in a previous study which suggests that both environmental and biological factors may be involved in the association between depression and sleep duration [22]. For instance, it has been shown that people with depression have reduced self-regulatory skills necessary to maintain/enforce bedtime routines, which may result in variable sleeping time at night [22, 23]. Also, the dysregulation of the hypothalamic pituitary adrenal (HPA) axis, which is associated with 7 depression [23] may have negative impact on the timing and patterns of sleep due to increased levels of cortisol (which is a hormone released in response to stress or anxiety) during pre-sleep period [24]. Also, pre-sleep worry and cognitive arousal commonly observed in adults has been shown to have similar negative impact on the timing and patterns of sleep [25, 26]. Also, insufficient/excessive sleep, as well as dysfunctions of sleep rhythm, are likely to occur during depression [27]. The sleep-wake cycle is regulated by two interacting processes, the circadian process and the homeostatic (or recovery) process.

157

Anujuo_binnenwerk_FINAL.indd 157 30/07/2018 21:19 Chapter 7

The former regulates the daily rhythms of the body and the brain; this is mainly due to the suprachiasmatic nucleus of the hypothalamus which provides an oscillatory pattern of activity regulating fundamental mechanisms, e.g., sleep-wake activity, hormone release, and liver function [27]. The resulting effect of these mechanisms may lead to variations/ disruptions in sleep duration which manifests as short or long sleep duration.

One important observation from our study is the larger contribution of depressed mood to the prevalence of short sleep in Turks and Moroccan compared to other ethnic groups. This could be attributed to the higher prevalence of depressive symptoms in Turks and Moroccans compared to the other groups, as has been widely reported in previous studies [12, 18].

Also, our study results indicate that the higher prevalence of short sleep in Ghanaian and Surinamese group was not accounted for by depression, suggesting that other factors such as work-related factors [28] may account for the increased prevalence of short sleep rather than depression. Further studies should explore other potential factors contributing to the higher prevalence of short sleep in these groups.

Another observation from our study is the low prevalence of long sleep and depressed mood in Dutch compared to other groups (except long sleep in African Surinamese), yet, depressed mood accounted for a higher percentage of the prevalence of long sleep in Dutch (9.1%) than in other groups (0.6 to 4.9%). A possible explanation for the association being absent in the other ethnic minority groups could be that the disadvantaged position of the ethnic minority groups such as job insecurity. For example, because many people rely on multiple insecure jobs, they may not get the time off to address their depression problem. In general, the higher prevalence of long sleep is not accounted for by depressed mood in all ethnic groups due to weak or absence of association.

The strength of our study lies on large sample size, thereby, providing us the possibility to stratify for ethnicity and compare ethnic groups. Some limitations of this study include the use of self-reported data on depressive symptoms and sleep, which may be subject to recall bias, hence participants may have under- or over-reported depressive symptoms and sleep durations. However, in the HELIUS study it was recently shown that the PHQ-9 measured the same underlying construct in all the ethnic groups [18],

158

Anujuo_binnenwerk_FINAL.indd 158 30/07/2018 21:19 Association between depressed mood and sleep duration

which implies that the ethnic differences in depressed mood are not caused by measurement variance of the questionnaire, but reflect real differences between the ethnic groups. Although the use of cross-sectional data cannot allow us to infer causal association between depression and short or long sleep, our finding concurs with a previous longitudinal study conducted on similar subject. However, since a bi-directional association between depression and sleep duration does exist, we cannot rule out the possibility of an effect of sleep on depression, and therefore we acknowledge our inability to unravel both mechanisms. However, discussion of the reverse direction is beyond the scope of this paper.

Previous studies [4, 7, 9] suggest an association between depressed mood and sleep duration. Our study corroborates with these previous studies, suggesting that improving psychosocial health among ethnic minority groups may be an important entry point for reducing ethnic inequalities in sleep.

In conclusion, the results of our study indicate that depressed mood is associated with short sleep in all ethnic groups, but not with long sleep (except in Dutch and Moroccans). Depressed mood accounted for a moderate part of the increased prevalence of short sleep in groups where its prevalence was high, in particular Turks and Moroccans. These findings suggest that strategies aimed to promote healthy sleep may also include management of depression, especially in Turks and Moroccans.

7

159

Anujuo_binnenwerk_FINAL.indd 159 30/07/2018 21:19 Chapter 7

REFERENCES

1. De Torres C. Mental health: The great depression.Nature . 2014;515:179 doi:10.1038/515179a 2. Depression: Fact sheet http://www.who.int/mediacentre/factsheets/fs369/ en/. Assessed on July 28, 2016. 3. Lépine J-P, Briley M. The increasing burden of depression.Neuropsychiatr Dis Treat. 2011;7(Suppl 1):3-7. doi:10.2147/NDT.S19617. 4. Patten CA, Choi WS, Jillin JC, Pierce JP, Depressive symptoms and cigarette smoking predict development of sleep problems in US adolescents. Paediatrics. 2000;106:E23. 5. Kaneita Y, Ohida T, Uchiyama M, Takemura S, Kawahara K, Yokoyama E, Miyake T, Harano S, Suzuki K, Fujita T. The relationship between depression and sleep disturbances: a Japanese nationwide general population survey. J Clin Psychiatry. 2006;67(2):196-203. 6. Matushita M, Koyama A, Ushijima H, et al ( 2014) Sleep duration and its association with sleepiness and depression in “ronin-sei” preparatory school students. Asian J Psychiatr. 2014 9:61-6. doi: 10.1016/j.ajp.2014.01.006. Epub 2014 Jan 29. 7. Tsuno N, Besset A, Ritchie K. Sleep and depression. J Clin Psychiatry. 2005; 66(10):1254-69. 8. Benca RM, Peterson MJ. Insomnia and depression. Sleep Med. 2008 Sep; 9 Suppl 1():S3-9. 9. Sivertsen B, Harvey AG, Lundervold AJ, Hysing M. Sleep problems and depression in adolescence: results from a large population-based study of Norwegian adolescents aged 16-18 years. Eur Child Adolesc Psychiatry. 2014;23(8):681-9. 10. Carney CE, Segal ZV, Edinger JD, Krystal AD (2007) A comparison of rates of residual insomnia symptoms following pharmacotherapy or cognitive- behavioral therapy for major depressive disorder. J Clin Psychiatry 68:254– 260. 11. Van der Wurff FB, Beekman AT, Dijkshoorn H, et al. Prevalence and risk- factors of depression in elderly migrants in the Netherlands. J Affect Disord. 2004;83:33-41. 12. Ikram UZ, Snijder MB, Fasseart TJ, et al. The contribution of perceived ethnic discrimination to the prevalence of depression. Eur J Public Health 2015;25:243-248. 13. Anujuo K Stronks K, Snijder MB, et al. Ethnic differences in self-reported sleep duration in the Netherlands – the HELIUS Study Sleep Med. 2015;15:1115- 1121.

160

Anujuo_binnenwerk_FINAL.indd 160 30/07/2018 21:19 Association between depressed mood and sleep duration

14. Stronks K, Snijder MB, Peters RJ, Prins M, Schene AH, Zwinderman AH. Unravelling the impact of ethnicity on health in Europe: the HELIUS study. BMC Public Health. 2013;13: 402. 15. Snijder MB, Galenkamp H, Prins M, Derks EM, Peters RJG, Zwinderman AH, Stronks K. Cohort profile: the Healthy Life in an Urban Setting (HELIUS) study in Amsterdam, the Netherlands. BMJ Open. 2017;7(12):ed 017873. doi: 10.1136/bmjopen-2017-017873. 16. Stronks K, Kulu-Glasgow I, Agyemang C. The utility of ‘country of birth’ for the classification of ethnic groups in health research: the Dutch experience. Ethn Health. 2009;14: 255-269. 17. Kroenke K, Spitzer RL, Williams JB, Lowe B et al. The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: a systematic review, Gen Hosp Psychiatry. 2010;32:345-59. 18. Galenkamp H, Stronks K, Snijder M, Derks EM. The PHQ-9 was measurement invariant across six ethnic groups: the HELIUS study. BMC Psychiatry. 2017;17(1):349. 19. Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, Dinges DF, Gangswisch J, Grandner MA, Kushida C, Malhotra RK, Martin JL, Patel SR, Quan SF, Tasali E. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;6: 843–844. 20. Williamson DF. The population attributable fraction and confounding: buyer beware. Int J Clin Pract. 2010;64:1019-23. 21. Rockhill B, Newman B, Weinber C. Use and misuse of population attributable fractions. Am J Public Health. 1998;88:15-9. 22. Dahl RE. The regulation of sleep and arousal: Development and Psychopathology. Development and psychopathology. 1996;8:3–27. 23. Lok A, Mocking RJ, Ruhé HG, Visser I, Koeter MW, Assies J, Bockting CL, Olff M, Schene AH. Longitudinal hypothalamic-pituitary-adrenal axis trait and state effects in recurrent depression.Psychoneuroendocrinology . 2012;37(7):892-902. 7 24. Gold PW. The organization of the stress system and its dysregulation in depressive illness. Mol Psychiatry. 2015;20(1):32-47. 25. Antypa N, Verkuil B, Molendijk M, Schoevers R, Penninx BW, Van der Does W. Association between chronotypes and psychological vulnerability factors of depression. Chronobiology International. 2017;34(8):1125-1135. 26. Levenson JC, Benca RM, Rumble ME. Sleep related cognitions in individuals with symptoms of insomnia and depression. J Clin Sleep Med. 2015;11(8):847- 854.

161

Anujuo_binnenwerk_FINAL.indd 161 30/07/2018 21:19 Chapter 7

27. Luca A, Luca M, Calandra C. Sleep disorders and depression: brief review of the literature, case report, and nonpharmacologic interventions for depression. Clin Interv Aging. 2013;8:1033-9. 28. Sadeh, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychol. 2004;23:542-545.

162

Anujuo_binnenwerk_FINAL.indd 162 30/07/2018 21:19 Association between depressed mood and sleep duration

7

163

Anujuo_binnenwerk_FINAL.indd 163 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 164 30/07/2018 21:19 CHAPTER 8

GENERAL DISCUSSION

Anujuo_binnenwerk_FINAL.indd 165 30/07/2018 21:19 Chapter 8

GENERAL DISCUSSION

This chapter highlights the key findings of the studies in this thesis, and discusses the methodological issues pertaining to the design of HELIUS study and ABCD study whose databases were used for the studies in this thesis, including strengths and limitations. We also discuss the methodological issues for assessing sleep duration in adults and children as recommended by the joint consensus of the National Sleep Foundation (NSF), American Academy of Sleep Medicine (AASM), and Sleep Research Society (SRS). Finally, we reflect on the key findings, relating it to the formulated model, and ends with implications for future research, with recommendations for further studies.

KEY FINDINGS

Differences in sleep duration/patterns between ethnic groups in Netherlands

The prevalence of short sleep was higher in all ethnic minority groups than the Dutch in both adults (chapter 2) and children (chapter 3). Among adults the prevalence of short sleep was 2 to 3 times higher in ethnic minority groups than in Dutch people both in men and women. Similarly, the prevalence of short sleep was 3 to 4 times higher in ethnic minority children than in Dutch children. The prevalence of long sleep was 2 times higher in all ethnic minority women (except in African Surinamese and South-Asian Surinamese) than in Dutch women. The prevalence of long sleep was also higher in ethnic minority men (except in African Surinamese and Ghanaians) than in Dutch.

Factors associated with the sleep patterns among ethnic minority groups

SES and shift work are factors found to be associated with sleep in all ethnic groups. As a consequence, these factors accounted for a small part of ethnic differences in short sleep. This implies that ethnic inequalities in sleep are not simply a reflection of differences in socio-economic position of various ethnic groups. Ethnic specific factors might play a role as well, such as the higher prevalence of depressed mood in ethnic minority groups. Depressed

166

Anujuo_binnenwerk_FINAL.indd 166 30/07/2018 21:19 General discussion

mood accounted for 3-16% of the prevalence of short sleep depending on the ethnic group involved, and moderate part of the prevalence of long sleep in Dutch and Moroccans (chapter 7).

Consequences of the observed sleep patterns among adults and preschool children of ethnic minority groups

Short sleep was associated with obesity and diabetes, and this association was partly consistent across ethnic groups (chapter 4), whereas the association between short sleep and dyslipidaemia differs between ethnic groups and depends on the particular CVD risk factors involved.

Short sleep was associated with overweight in Dutch and Moroccan preschool children in the unadjusted analysis, but this association was mediated by SES (education). Short sleep was not associated with raised BP in all ethnic minority preschool children except in African Surinamese.

Short and long sleep duration was not associated with arterial stiffness as measured with pulse wave velocity (PWV) in all ethnic minority groups (chapter 5), suggesting that the link between sleep duration and CVD outcomes does not seem to operate through arterial stiffness.

Short sleep was consistently associated with increased prevalence of CVD in all ethnic groups (chapter 6), and accounted for 10-15%, to the increased risk of CVD in South-Asian Surinamese, African Surinamese, Ghanaians, and 5% to the increased risk of the Turks and Moroccans, independent of conventional CVD risk factors. Conventional CVD risk factors jointly accounted for 10-24% to the increased risk of CVD in ethnic minority populations.

Distinction in results between adults and preschool children of ethnic groups. 8 There are some similarities in results between ethnic minority adults and preschool children in this thesis. For instance, both ethnic minority adults and their preschool children sleep shorter than ethnic Dutch (chapter 2 and 3).This highlights similar trends in sleep patterns between adults and children of ethnic minority groups. Additional similar finding between

167

Anujuo_binnenwerk_FINAL.indd 167 30/07/2018 21:19 Chapter 8

adults and preschool children was that short sleep duration was associated with obesity in adults, (chapter 4) and overweight in preschool children (chapter 3). This indicates that short sleep may be a common risk factor for obesity/overweight in both adult and children of ethnic minority groups. On the other hand, there are differences in results between ethnic minority adults and preschool children noticed in this thesis, including: Lack of significant association between short sleep and blood pressure in preschool children (except in African Surinamese in which association of short sleep with raised BP was observed), whereas an association of short sleep with BP(hypertension) was found for some ethnic minority adults. This study also found additional similarities between adults and preschool children, for example, among ethnic minority groups, the prevalence of short sleep was 2-3 times as high in adults and preschool children compared with ethnic Dutch.

METHODOLOGICAL CONSIDERATIONS

Cross-sectional data

The dataset used for this thesis came from the Healthy Life in an Urban Setting (HELIUS) and Amsterdam Born Children and their Development (ABCD) studies. The data used from both studies were cross-sectional, although the studies have been designed prospectively. Methodological issues regarding the methods and designs, strengths and limitations of the studies have been discussed in various chapters of this thesis. Additional detailed explanation is given for the strengths and limitations.

Strengths

Healthy Life in an Urban Setting (HELIUS) study and Amsterdam Born Child and their Development (ABCD) study were differently designed to conform and meet specific purposes. HELIUS study is one of the largest population-based cohort studies in Europe with focus on major ethnic minority groups resident in Amsterdam, the Netherlands [1]. The Amsterdam Born Children and their Development study (ABCD) is a large prospective population-based study established in 2003 to examine the association between maternal lifestyle, medical, psychosocial and environmental conditions during pregnancy and children`s health at birth, as well as later in life in Amsterdam, the Netherlands [2]. Both studies share

168

Anujuo_binnenwerk_FINAL.indd 168 30/07/2018 21:19 General discussion

certain common features by focusing on same ethnic groups (except for South-Asian Surinamese in ABCD study). This provides a homogenous population which makes it possible to make comparisons in results. Also both studies contain information on important variables such as sleep duration and other health related variables such as obesity/overweight and blood pressure/hypertension, which makes it possible to see similarities and differences in results between adults and children.

Limitations

The major limitation of studies in this thesis is the use of self-report data for sleep duration. Whereas in HELIUS study, information on sleep duration was provided by adult participants through the use of a questionnaire, for the children in ABCD study, information on sleep duration was provided by the parents of the children. As both studies are cross-sectional, it has the implication of introducing bias as well as making causal inference impossible for the obtained results.

In the HELIUS study, the sleep variable was obtained subjectively through self-reported questionnaire with an item which requested participants to indicate “on average , the number of hours of sleep per night”. For ABCD study, participants were asked to indicate “on average the number of hours the child sleep per night on school days and during weekend”. The use of self-reported data is prone to recall bias, as participants may under- or over-report sleep duration. Such bias could be minimized or eliminated by the use of objective sleep measures such as polysomnography (PSG), electroencephalograph (EEG), and actigraphy which provide accurate and more reliable measures. However, despite the self-reported nature of the sleep variable, subjective sleep measures (questionnaires, diary) are still widely used, hence use of self-reported data seem justified, given what we know on validity. For instance, it has been shown that perceived inadequate sleep data obtained subjectively was similar to objective sleep measures provided in a primary care setting [3]. In addition, while some other research has indicated that self-reported data on sleep duration typically overestimates 8 the true duration of sleep, it has been demonstrated that the reporting bias is non-differential across sex and ethnic groups [4].

Another methodological issue that relates to sleep studies generally, and which also relate to studies in this thesis is the inconsistency in the definition/

169

Anujuo_binnenwerk_FINAL.indd 169 30/07/2018 21:19 Chapter 8

classification of short and long sleep duration. In various studies involving sleep duration, several authors have used different definitions for short and long sleep duration. There is discordance in the definition of short and long sleep duration due to non-harmonization of standard definition for healthy sleep by the National Sleep Foundation (NSF). For instance, in population- based study investigating ethnic differences in self-reported sleep duration, Hale et al 2007, classified short sleep as sleeping <=6 hours /night [5]; whereas in another study, Stamatakis et al 2007 [6] defined short sleep as sleeping <7 hours/night. In another instance, Whinnery et al (2014) [7] used a more extreme definition of <=5 hours/night commonly referred to as “very short ” sleep duration. It was not until 2015, that the NSF, American Academy of Sleep Medicine (AASM) and Sleep Research Society (SRS) jointly agreed on a more universally accepted definition for healthy sleep duration for different age categories/groups as can be seen in the table 1 [8]. For the sole purpose of this thesis, sleep duration generally recommended as healthy for adults (7-8 hours per night) and preschool children (10-11 hours/night) were used [9]. Accordingly, short sleep was defined as <7 hours of sleep/night for adults, and <10 hours/night for preschool children; and this definition was used throughout the entire study. Whereas, long sleep was defined as either >9 hours, or >=9 hours per night depending on the period (year) the study was carried out and also on the number of participants within the long sleep range (sample size). The new recommendations for defining sleep duration for adults and children came into effect in the midst of this research, when some studies have already been concluded and when there was no harmonization of sleep guidelines. For instance, in chapters 2, 5, and 7, long sleep duration was defined as >=9 hours/night, whereas the new guidelines emphasized >9 hours/night. In addition, although, the new guidelines clearly differentiated between young adults (18-25 years), adults (26-64 years) and older adults (>=65 years), and recommended 7-9 hours/night as healthy for young adults and adults, and 7-8 hours/night for older adults. However, throughout the entire study, participants were not stratified by age according to the new classifications, and thus, the results may not apply to all age groups, and should be interpreted with caution. The new classification also included the number of hours which `may be appropriate` for different age groups (Table 1). On one hand, this may suggest that if ethnic minority groups sleep less than the recommended hours but had a good quality sleep, that it may not be much a problem as sleep quality is important too. On the other hand, that people from either of the ethnic groups may sleep longer than the recommended hours could be an indication of health status (ill-health) and not necessarily a problem of

170

Anujuo_binnenwerk_FINAL.indd 170 30/07/2018 21:19 General discussion

long sleeping. The differences in the definition of sleep had no implication for the comparability of our results with other studies, as this also applies to other studies carried out within the same period. However, the definition issue is a general concern to overall studies involving sleep duration, hence it is expected that expert bodies should further improve on the definition of sleep to capture both quantity and quality of sleep.

Table 1: the recommended hours of sleep per night for preschool children and adults

Age Recommended May be appropriate Preschoolers 3-5 years 10-13 hours 8-9 hours/14 hours Young adults 18-25 years 7-9 hours 6 hours/10-11 hours Adults 26-64 years 7-9 hours 6 hours/10 hours Older adults >65 years 7-8 hours 5-6 hours/9 hours

Derived from the website of National Sleep Foundation: NSF sleep duration recommendation [9]

Why did we focus on short sleep and less on long sleep?

The studies in this thesis focused more on short sleep rather than long sleep because ethnic differences in short sleep were most apparent. Whereas ethnic differences in short sleep persisted even after adjusting for covariates (chapter 2), ethnic differences in long sleep disappeared after adjusting for SES, indicating that SES explains ethnic differences in long sleep but, not in short sleep. This suggests that the major problem specific for ethnic minority groups is short sleep duration. However, the fact that SES explains ethnic differences in long sleep does not necessarily mean that long sleep is not a problem among ethnic minority groups, hence it was also investigated alongside short sleep in chapters 5 and 7. Also, emphasis was laid on short sleep because, although long sleep may be associated with adverse health outcomes, previous studies found that short sleep was more consistently related to CVD risk factors compared to long sleep [10, 11]. 8 How generalizable are the results to other ethnic groups?

One important question emerging from the assessment of results contained in this thesis is how generalizable the results are to other ethnic minority groups, i.e. other than those considered in this study? Considering the design

171

Anujuo_binnenwerk_FINAL.indd 171 30/07/2018 21:19 Chapter 8

and the overall strengths of the studies in this thesis which included, (but not limited to) large sample size, which allows for more reliable estimations of population characteristics being studied, use of same methodology to investigate multiple ethnic groups that reside in one city, conducting appropriate statistical analysis, etc., which affirms internal validity of our study. At the same time, this study identified factors which are determinants of sleep duration, such as SES (education and employment), shift work, etc. which were also basic characteristics of other ethnic minority groups identified in other studies [12, 13, 14]. As the findings of this thesis is similar to those of previous studies conducted in cities based on common characteristics and determinants, our results are likely to be generalizable to other ethnic minority groups living in comparable urban centres. It may however, differ for specific groups with different living circumstances.

REFLECTIONS ON THE KEY FINDINGS IN THIS THESIS

This thesis succinctly demonstrates significant role of sleep in ethnic inequalities in health specifically on CVD and risk factors. This section further expatiates on the key findings by reflecting on three basic themes as shown below.

Why is short sleep more prevalent among ethnic minority groups in the Netherlands?

According to the findings of this thesis, SES and shift work explained a little part of the ethnic differences in sleep duration. Ethnic minority groups have low educational levels, low employment, more often do shift work. These factors are recognized drivers of short sleep duration among ethnic minority groups, and have been discussed in the various chapters of this thesis. Similarly, other studies have shown that low SES and shift work are common among ethnic minority groups, and are among the reasons for increased prevalence of short sleep [15-19]. Our findings therefore partly agree with these studies. Additional factors, found as potential reasons for increased prevalence of short sleep among ethnic minority groups include cultural factors such as co-sleeping, disregard for napping, and the belief that hardworking and successful people require short sleep [20, 21], environmental factors such as noise, light, poor air quality/pollution, temperature [22,23,24], unsafe residential quarters/disadvantaged neighborhoods [25], acculturation [7], immigration (leaving one`s country to reside in a different country) [26] ,

172

Anujuo_binnenwerk_FINAL.indd 172 30/07/2018 21:19 General discussion

stress [27], use of electronic devices in the hour before bedtime, household size[28,29, 20,30], genetics and perceived discrimination [31]. Although these have not been studied in this thesis, they might be hypothesized part of the observed inequalities in sleep patterns.

A potential explanatory factor that has been studied in this thesis is depression (Chapter 7). Depressed mood was consistently associated with the prevalence of short sleep in all ethnic groups, and accounted substantially for the prevalence of short sleep in all ethnic groups; and long sleep especially in Moroccans and Dutch groups. This finding highlights depression, as relevant in ethnic inequalities in sleep patterns.

Also migration itself could be a factor to reckon with for increased prevalence of short sleep among ethnic minority groups [26], but this has not been clearly explained in literature. Does it mean that exposure to different environment as a result of migration may potentially alter the circadian rhythm? For instance, a recent study conducted in South Africa showed that European South Africans reported shorter sleep than other population groups consisting of native Africans, Indians/Asian Africans, and multi- ancestral groups [32]. These findings may suggest that the notion that non- European groups sleep shorter than Europeans may not always hold true. The question arises as to whether migration has a significant role to play in differences in sleep patterns among migrant/ethnic groups, for European South-Africans have at some point migrated to South Africa. Or could it be that the stress associated with migration, as reported in earlier studies [27], may have influenced the sleep patterns of migrants, irrespective of their ethnic background? Or could it be that changes in circadian rhythm is influenced when people migrate to another country? Yet, previous study has shown that American Indians sleep shorter than [33], indicating that some unmeasured factors may play a role. Further studies should unravel the potential role of migration and stress on differences in sleep patterns among various ethnic groups.

How important is sleep for ethnic inequalities in health? 8 This thesis showed a clear difference in sleep patterns among various ethnic groups, such that all ethnic minority groups sleep shorter than their Dutch counterparts in both adults and preschool children in both males and females. This finding is consistent with previous studies in the U.S.

173

Anujuo_binnenwerk_FINAL.indd 173 30/07/2018 21:19 Chapter 8

which also reported shorter sleep duration among ethnic minority groups than White Americans [34]. The question arises as to what this findings actually mean for ethnic inequalities in health. This thesis attempted to address this questions in the findings of chapter 6 and 5 respectively, with results showing that short sleep was consistently associated with CVD in all the ethnic groups, and partly associated with CVD risk factors in specific ethnic groups respectively. These novel findings seem to suggest that ethnic minority groups with increased prevalence of short sleep may be more prone to develop CVD. In addition, we studied to what extent short sleep contributed to the observed ethnic inequalities in CVD. We have shown that sleep alone statistically accounted for 5-15%, to the ethnic differences in CVD, suggesting that the prevalence of CVD could be reduced by these amounts if short sleep was eliminated. This finding that sleep, independent of other CVD risk factors accounted for ethnic differences in CVD is very crucial, judging from the general beliefs of people that sleep affects CVD through these conventional CVD risk factors. Again, considering the many intervention studies focusing on obesity, diabetes, hypertension, dyslipidaemia, and certain health behaviours for the management or prevention of CVD in various populations, does it mean that giving a similar attention to sleep could result in a better management/prevention of CVD compared to the former? These are potential questions which future research may address.

Is sleep an important entry point for reducing health inequalities?

A thorough examination of the results of studies presented in this thesis showed that the results offer supportive evidence that sleep is a novel risk factor of CVD among various ethnic minority groups. Evidence from other studies have shown that by improving sleep, cardiovascular health might improve. For instance, a preliminary data which suggested that sleep extension may have the potential to improve cardiovascular risk factors has been reported [35]. For example, in a small pilot study in which 22 subjects with prehypertension and sleep duration of <7 hours per night were randomized to sleep extension or maintenance groups, with the aim of increasing sleep duration by I hour over 6 weeks. The result showed that though, there was a reduction in blood pressure in both groups, it was greater in the group randomized to sleep extension although the difference was not statistically significant [35]. Also, results from a recent study has shown

174

Anujuo_binnenwerk_FINAL.indd 174 30/07/2018 21:19 General discussion

that improving sleep by treating sleep related disorder such as insomnia led to a decrease in participant’s blood pressure [36]. For example, in a study in which patients with insomnia and coronary heart disease (CHD) were randomized to 6 weeks of web-based cognitive behavioral therapy – wCBT- 1(You go to sleep) group compared with 6 weeks of a sleep education website, followed by wCBT-1 access (control group), the researchers found a 2.9mm/Hg ± 19.2 reduction in systolic blood pressure in the treatment group compared to 0.6mm/Hg ± 6.6 increase in the control group. This suggests that improving sleep via wCBT improved both insomnia and blood pressure in patients with comorbid CHD, indicating how improved sleep may lead to improving adverse health outcomes. From the results of this thesis, promoting a healthy sleep pattern in ethnic minority groups could be expected to help in reducing ethnic inequalities in CVD and risk factors. This, in combination with the intervention studies showing an effect of sleep in improving health outcomes, clearly indicate that sleep may serve as entry point to reducing health inequalities. In other words, measures aiming to reduce ethnic inequalities in health outcomes, including CVD and risk factors therefore should include improvement in healthy sleep , as supported by the results in this thesis.

When promoting healthy sleep patterns in these groups it is of course important to consider the living condition of the people. Situations where ethnic minority groups live in crowded homes [30], poor neighbourhood/ environments [25], and poor lifestyles should be improved. Improvement of these factor may have positive impact on sleep health and subsequent effect on health outcomes.

IMPLICATIONS FOR PUBLIC HEALTH AND RECOMMENDATIONS FOR FUTURE RESEARCH

Implications for public health (policy and practice)

The public health burden of sleep short sleep is immense [37, 38, 39, 40, 41], as can be partly attested from the findings in this thesis which revealed 8 impending adverse health outcome among ethnic minority groups that may result from short sleep. The findings that all ethnic minority groups reported shorter sleep duration than ethnic Dutch in both men and women, and in preschool children revealed a relevant aspect of unhealthy behavior among ethnic minority groups. As a result of the high prevalence of short sleep

175

Anujuo_binnenwerk_FINAL.indd 175 30/07/2018 21:19 Chapter 8

among ethnic minority groups, they are likely to become more exposed to potential adverse health outcomes associated with short sleep duration. To avoid this unwanted development, there is the need for an urgent call for action so as to avert (if not eliminate) eventual consequences in the future.

When compared with other risk factors such as smoking, excess consumption of alcohol, physical activity and obesity which have been giving much priority, sleep received less or no attention in the public health sphere and clinical settings. The reason is not far-fetched, and is primarily due to lack of awareness as sleep problems and sleep disorders are usually undiagnosed [42,43,44]. Lack of awareness among the general public, and ethnic minority groups in particular, concerning sleep health creates a lacuna of knowledge about adverse health impacts of short sleep, and the required amount of sleep needed for optimal health [39]. Also health care providers and health care professionals often do not give counselling to their patients about healthy sleep habits [45]. In addition, sleep is not usually included in health screenings as opposed to other health behaviour such as smoking, alcohol consumption, physical activity and healthy eating [40]. Exclusion of sleep in screening programmes may be related with clinician`s and public health professionals lack of knowledge of the importance of sleep [46]. This lack of knowledge often exposes ethnic minority groups to potential risks of diseases attributed to short sleep, and this has the overriding effects of further worsening their health, as ethnic minority groups already have increased prevalence of CVD and risk factors. It is therefore necessary to create public awareness effective enough to inform and inculcate the need for a change to a healthy behavior towards sleep health. I would suggest the use of strategies involving public education and enlightenment campaigns. This can be carried out at national, local and community levels (including churches); and also in colleges, universities and work places. Parents can also transfer the knowledge to their children by ensuring that they achieve sufficient healthy sleep which is very important for mental development and improved academic performance. Also, health care providers should not focus only on patients known medical conditions during consultation, but also they should inquire about sleep problems as patients themselves, do not seem to recognize and discuss the potential of their poor sleep habits. In addition, there is also the need to further identify specific determinants of poor sleep in low-income populations/minority groups for instance, by focusing on factors operating in households and work places, and behaviours that influence parents and their children which might make them vulnerable to having short sleep. Moreover, there is the need for

176

Anujuo_binnenwerk_FINAL.indd 176 30/07/2018 21:19 General discussion

more standardized sleep health interventional studies to demonstrate the modifiable role of sleep in health disparities among ethnic groups, especially at grassroots levels, as exemplified by Jean-Louis and colleagues [47] whose community-based intervention tagged “culturally-tailored education/ approach” create awareness about sleep health through interactive campaign, for example, in a church congregation or in schools, barber shops, beauty salons, etc. whereby, factors that interfere with healthy sleep were discussed, accompanied by counselling and suggesting ways for improvement. People could also be recruited for focus group discussion concerning sleep health. Other ways of creating awareness is via educational by means of website where relevant information about sleep shall be disseminated. Awareness could also be provided by distributing relevant information about sleep through various media, such as TV, radios, and other social networks.

Another recommendation is to embark on sleep screening exercise so as to identify individuals with symptoms associated with sleep-related problems, and who may be at risk of possible adverse health outcomes related with sleep. Finally, stakeholders such as general practitioners can ask questions relating to sleep habits of patients and giving the necessary counsel where applicable.

Implications for future research

With few exceptions, most research involving ethnic disparities in sleep and its association with CVD and risk factors were conducted with the use of cross-sectional data. Studies in this thesis also used cross-sectional data which makes it difficult to present a proof of causality. So far, the data on previous longitudinal studies were based on European populations. Therefore, there is a need to conduct further similar studies with longitudinal data among ethnic minority groups especially in Europe since the determinants may differ due to cultural factors. The studies shall also offer more supportive evidence for our results in demonstrating causality for the relationship between sleep and CVD and risk factors. Most population-based studies on sleep, including this thesis relied on self-reported data that may introduce 8 recall bias, in which case, number of hours of sleep may have been inaccurately reported. Future studies should consider the use of objective sleep measurement obtained with actigraphy, polysomnography (PSG) and electroencephalography (EEG) which are not influenced by recall bias. Additionally, PSG equally provide information about quality of sleep, and

177

Anujuo_binnenwerk_FINAL.indd 177 30/07/2018 21:19 Chapter 8

other aspects such as sleep timing and variability. If possible, future studies can combine both subjective and objective measurements in order to make room for easy comparisons of results. Also there is the need to standardize the measurement of sleep duration as well as other sleep characteristics used for the studies.

Other factors such as, environmental factors, acculturation, and generation type (first or second generation) of ethnic groups, all of which may affect sleep duration [7, 48], but which were not taken into account in this thesis, should be considered in future studies. Environmental factors such as place of residence, noise, pollution, temperature may differ among ethnic minority groups, and may have different impact on sleep duration, and could mediate the observed associations of sleep with CVD and risk factors. It is equally important to identify other unmeasured confounders and introduce them during analysis so as to further comprehend the extent sleep duration is related with CVD and risk factors, and its usefulness as a possible target for public health intervention. It is therefore recommended that future studies include these factors in their analysis.

Interventional studies are also necessary to examine the extent by which the prevalence of CVD and risk factors could be reduced with improved sleep duration in ethnic minority populations in particular. Such intervention studies should consider in its inclusion, all other factors that improve lifestyle such as diet and physical activity which, alongside improved sleep, would assist to realize the overall health improvement. In addition, the factors that might be particularly important in shaping sleep patterns in ethnic minority populations, such as crowded housing conditions, shift work and depression, should be taken into account. A recent intervention study has been successfully conducted for promoting awareness about sleep and sleep- related disorders in African American community-based settings [49]. The study, which made use of tailored website for promoting awareness about sleep and sleep-related disorders among African American in a community- based settings included the use of visuals, key messages, video narratives, in-depth interviews, usability procedures and brief surveys; reported that this tailored approach instilled in participants, knowledge on improving sleep, and it also has potential for serving as a tool for advancing health equity [49]. This method could also be employed to address the problem of awareness among ethnic minority groups discussed in Europe. In addition, since there exists some similarities in sleep patterns, and association of short sleep with obesity/overweight between adults and preschool children of

178

Anujuo_binnenwerk_FINAL.indd 178 30/07/2018 21:19 General discussion

some ethnic groups as demonstrated in our study, it may be a good idea to develop a family intervention strategy that would integrate both adults and preschool children, taking into account these common determinants of short sleep among ethnic groups.

Overall conclusion

This thesis has ushered in new information/knowledge about sleep health and its related adverse health outcomes observed among various ethnic groups resident in the Netherlands. Based on the results in this thesis, it can be concluded that all ethnic minority groups, on average, reported shorter sleep duration than ethnic Dutch in both men and women, and among preschool children. Short sleep was associated with CVD and risk factors (obesity, diabetes, hypertension and dyslipidaemia), and contributed significantly to ethnic inequalities in CVD independently of other well- known CVD risk factors. The observed high prevalence of short sleep among ethnic minority groups, and its association/relationship with CVD and risk factors revealed an important consequence pattern of health behaviour with likely concomitant adverse consequences, and which may predispose to unwanted health outcomes. There is a need to provide the necessary information to the public, to guard and ensure healthy sleep patterns. Applying proper and effective strategy with major attention, paid to ethnic minority groups could help to disseminate this knowledge. This could be followed by carrying out intervention studies designed to address the impending problems of short sleep. If successfully executed, part of the increased prevalence of CVD and risk factors, attribute to sleep would be reduced, hence leading to better health. It is also important to consider the inclusion of sleep when designing intervention studies for CVD and risk factors, taking adverse living conditions that shape sleep patterns into account. 8

179

Anujuo_binnenwerk_FINAL.indd 179 30/07/2018 21:19 Chapter 8

REFERENCES

1. Stronks K, Snijder MB, Peters RJ, Prins M, Schene AH, Zwinderman AH. Unravelling the impact of ethnicity on health in Europe: the HELIUS study. BMC Public Health 2013; 13:402. 2. van Eijsden M, Vrijkotte TGM, Gemke RJBJ, van der Wal MF. Cohort Profile: The Amsterdam Born Children and their Development (ABCD) Study.Int J Epidemiol. 2011;40:1176–86. 3. Lauderdale DS, Knutson KL, Yan LL, et al. Objectively measured sleep characteristics among early-middle-aged adults: the CARDIA study. Am J Epidemiol. 2006;164(1):5–16. 4. Voderholzer U, Al-Shajlawi A, Weske G, Feige B, Riemann D. Are there gender differences in objective and subjective sleep measures? A study of insomniacs and healthy controls. Depression and anxiety. 2003; 17(3):162–172. 5. Hale L, Do DP: Racial differences in self-reported sleep duration in a population-based study, Sleep. 2007;30(9):1096-103. 6. 6. Stamatakis KA, Kaplan GA, Roberts RE. Short sleep duration across income, education, and race/ethnic groups: population prevalence and growing disparities during 34 years of follow-up. Ann Epidemiol. 2007;17(12):948-55. 7. Whinnery J, Jackson N, Rattanaumpawan P, Grandner MA. Short and long sleep duration associated with race/ethnicity, sociodemographics, and socioeconomic position. Sleep. 2014;37(3):601-11. 8. Watson NF, Badr MS, Beleny G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;38(6):843-844. 9. National Sleep Foundation recommends new sleep times. Retrieved on November 07, 2017 from https://sleepfoundation.org/press-release/national- sleep-foundation-recommends-new-sleep-times/page/0/1. 10. Grandner MA, Chakravorty S, Perlis ML, Oliver L, Gurubhagavatula I. Habitual sleep duration is associated with self-reported and objectively determined cardiometabolic risk factors. Sleep Med. 2014 Jan;15(1):42-50. 11. Guo X, Zheng L, Wang J, Zhang X, Zhang X, Li J, Sun Y. Epidemiological evidence for the link between sleep duration and high blood pressure: A systematic review and meta-analysis. Sleep Med. 2013 Apr;14(4):324-32. 12. Hall M.H., Matthews K.A., Kravitz H.M., Gold E.B., Buysse D.J., Brom- berger J.T., et al.Race and financial strain are independent correlates of sleep in midlife women: The SWAN Study. Sleep. 2009;32(1):73-8. 13. Luckhaupt S.E.Short sleep duration among workers – United States, 2010 Weekly. 2012;61 (16):281-285.

180

Anujuo_binnenwerk_FINAL.indd 180 30/07/2018 21:19 General discussion

14. US Department of Labour Bureau of Statistic. Workers on flexible and shift schedules in May 2004. Available from http://www.bls.gov/news.release/flex. toc.htm (2005) [accessed 13.12.22.]. 15. Jean-Lous G, Magai CM, Cohen CI, Zizi F, von Gizycki H, DiPalma J, Casimir GJ. Ethnic differences in self-reported sleep problems in older adults. Sleep. 2001;24(8):926-933. 16. Gellis LA, Lichstein KL, Scarinci IC, et al. Socioeconomic status and insomnia, J Abnorm Psychol. 2005;114(1):111–8. 17. Patel NP, Gooneratne N, Xie D, Branas CC. Poverty effects on population sleep. Sleep. 2008;31:A353 . 18. Grandner MA, Patel NP, Gehrman PR, Xie D, Sha D, Weaver T, Gooneratne N. Who gets the best sleep? Ethnic and socioeconomic factors related to sleep complaints. Sleep Med. 2010 May;11(5):470-478. 19. Mezick EJ, Matthews KA, Hall M, Strollo PJ, Buysse DJ, Kamarck TW, Owens JF, Reis SE. Influence of race and socioeconomic status: Pittsburgh Sleep SCORE Project. Psychosom Med. 2008;70(4):410-416. 20. CultureGram 2012 World Editions – Republic of Surinamese. http://www. culturegrams.com/images/samples/SurinameWorld.pdf accessed on March 21 2016. 21. Hollan D. Sleeping, dreaming, and health in rural Indonesia and the U.S. urban: A cultural and experiential approach. Soc Sci Med. 2013;79:23-30. 22. Halparin D. Environmental noise and sleep disturbances: A threat to health? Sleep Science. 2014;7(4):209-12. 23. Stansfeld SA, Matheson MP. Noise pollution: non-auditory effects on health. Br Med Bull. 2003;68():243-257. 24. Basner M, Muller u, Elmenhorst EM. Single and combined effects of air, road and rail traffic noise on road and recuperation. Sleep. 2011;34(1):11-23. 25. Pampel FC, Krueger PM, Denney JT. Socioeconomic disparities in health behaviors Annu Rev Sociol. 2010 Aug;36:349-370. 26. Jackson Cl, Hu FB, Redline S, Williams DR, Mattei J, Kawachi I. Racial/ ethnic disparities in short sleep duration by occupation: the contribution of immigrant status. Soc Sci Med. 201;118:71-9. 27. Sadeh A, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychology.2004;23(5):542–545. 28. Shochat T. Impact of lifestyle and technology development on sleep. Nat Sci Sleep. 2012;4():19-31. 8 29. Calamaro CJ, Mason TB, Ratcliffe SJ. Adolescents living the 24/7 lifestyle: effects of caffeine and technology on sleep duration and daytime functioning. Pediatrics. 2009;123(6):e1005-10.

181

Anujuo_binnenwerk_FINAL.indd 181 30/07/2018 21:19 Chapter 8

30. Harker L. Chances of a lifetime: the impact of a bad housing on children`s life. Assessed online on 31/01/2018 from https://england.shelter.org.uk/__ data/assets/pdf_file/0016/39202/Chance_of_a_Lifetime.pdf. 31. Grandner MA, Hale L, Jackson N, Patel NP, Gooneratne NS, Troxel WM. Perceived racial discrimination as an independent predictor of sleep disturbance and daytime fatigue. Behav Sleep Med. 2012;10(4):235-49. 32. Peltz K. Differences in sleep duration among four different population groups of older adults in South Africa. Int J Environ Res Public Health. 2017;14(5). Pii: E502. doi: 10.3390/ijerph14050502. 33. Rosenberg E. American sleep patterns vary by age, race and place, study finds. Accessed on 10/02/2018 from http://www.nytimes.com/2016/02/21. us.american-sleep-patterns-vary-by-race-and-place-study-finds.html. 34. Nunes J, Jean-Louis G, Zizi F, Casimir GJ, von Gizycki H, Brown CD, McFarlane SI. Sleep duration among black and White Americans: results of the National Health Interview Survey. J Natl Med Assoc. 2008;100(3):317- 322. 35. Haack M, Serrador J, Cohen D, Simpson N, Meier-Ewert H, Mullington JM. Increasing sleep duration to lower beat-to-beat blood pressure: a pilot study. J Sleep Res. 2013;22(3):253-304. 36. Javaheri S, Mehra R, Drerup D, Redline S. A planning study: Insomnia intervention for cardiovascular disease reduction. Sleep. 2017;40(1):A132. 37. Hashmi AM, Khawaja IS. (2017) Sleep-Disordered Breathing and Mental Illness. In: Sharafkhaneh A, Yohannes A, Hanania N, Kunik M. (eds) Depression and Anxiety in Patients with Chronic Respiratory Diseases. Springer, New York, NY. 38. Sorscher AJ. How is your sleep: a neglected topic for health care screening. J Am Board Fam Med. 2008;21(2):141-8 39. Colton HR, Altevogt BM. Sleep disorders and sleep deprivation: an unmet public health problem. Institue of Medicine (US) Committee on Sleep Medicine Research. National Academies Press 2006. 40. Heffron T. Sleep well, be well: A National Health Priority. 2014. Assessed on Dec 12, 2017 from: http://sleepeducation.org/news/2014/05/16/sleep-well- be-well-a-national-health-priority. 41. Perry GS, Patil SP, Presley-Cantrell LR. Raising awareness of sleep as a healthy behavior. Prev Chronic Dis. 10(2013), p 130081, 10.5888/pcd10.130081. 42. Benca RM. Diagnosis and treatment of chronic insomnia: a review. Psychiatr Serv 2005;56(3):332–43. 43. Kapur V, Strohl KP, Redline S, Iber C, O’Connor G, Nieto J. Underdiagnosis of sleep apnea syndrome in US communities. Sleep Breath 2002;6(2):49–54.

182

Anujuo_binnenwerk_FINAL.indd 182 30/07/2018 21:19 General discussion

44. Singh M, Drake CL, Roth T. The prevalence of multiple sleep-onset REM periods in a population-based sample. Sleep 2006;29(7):890–5. 45. Hing E, Cherry DK, Woodwell DA. National ambulatory medical care survey: 2004 summary. Adv Data 2006;(374):1–33. 46. Papp KK, Penrod CE, Strohl KP. Knowledge and attitudes of primary care physicians toward sleep and sleep disorders. Sleep Breath. 2002;6(3):103-9. 47. Resnick B. The racial inequality of sleep. The Atlantic Daily. Oct. 27, 2015. Assessed on 10/01/2018 from: https://www.theatlantic.com/health/ archive/2015/10/the-sleep-gap-and-racial-inequality/412405/. 48. KL: Does Inadequate Sleep Play a Role in Vulnerability to Obesity?: American Journal of Human Biology 2012; 24:361-371. 49. Robbins R, Senathirajah Y, Williams NJ, Hutchinson C, Rapoport DM, Allegrante JP, Cohall A, Rogers A, Ogedegbe O, Jean-Louis G. Developing a tailored website for promoting awareness about obstructive sleep apnea (OSA) among blacks in community-based settings. Health Commun. 2018;(0):1-9 doi: 10.1080/10410236.2018.1423865.

8

183

Anujuo_binnenwerk_FINAL.indd 183 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 184 30/07/2018 21:19 SUMMARY

Anujuo_binnenwerk_FINAL.indd 185 30/07/2018 21:19 Summary

SUMMARY

CVD is still the leading cause of morbidity and mortality across the globe. A wide body of evidence has shown that the prevalence of CVD and risk factors is higher among ethnic minority groups compared to high-income countries` host populations. With the emergence of sleep research, there is a growing evidence of a link between sleep duration and CVD and risk factors among various ethnic groups, but European studies on this subject are lacking or inadequate. This thesis provides first-hand information of a possible association/link between sleep duration and CVD and risk factors among ethnic minority groups living in Amsterdam, The Netherlands. The first objective was to assess ethnic inequalities in sleep duration among adult population, and preschool children, and also to explore possible factors that may explain this inequalities. The second objective was to assess the association between short sleep and CVD risk factors in both adults and preschool children, and the third objective was to determine whether short sleep was associated with CVD, and whether short sleep contributed to ethnic differences in CVD independent of other CVD risk factors. The final objective was to provide evidence of a possible link between depressed mood and sleep duration among various ethnic minority groups. Chapter one described the background, objectives, and outline of this thesis.

Ethnic inequalities in sleep duration among adults and preschool children

The purpose was to assess ethnic differences in sleep patterns among adults and preschool children resident in Amsterdam. In chapter 2, we commenced by investigating ethnic differences in sleep duration between ethnic Dutch and various ethnic minority groups, and the contribution of socioeconomic status (SES) to the observed differences. We found that all ethnic minority groups were more likely to report shorter sleep duration than ethnic Dutch in both men and women. Among adults, the prevalence of short sleep was 2-3 times higher in ethnic minority groups than in Dutch people in both men and women. Similarly, the prevalence of short sleep was 3-4 times higher in ethnic minority children than in Dutch children. The prevalence of long sleep was 2 times higher in all ethnic minority women (except in African Surinamese and South-Asian Surinamese) than in Dutch women. The prevalence of long sleep was also higher in ethnic minority men (except in African Surinamese and Ghanaians) than in Dutch. Contribution of

186

Anujuo_binnenwerk_FINAL.indd 186 30/07/2018 21:19 Summary

ethnic differences in educational level to differences in short sleep was small. We concluded that SES explained the observed ethnic differences in long sleep duration, but contribution of SES to ethnic differences in short sleep duration was small.

In chapter 3, we investigated whether the sleep patterns as observed in the adult population is similar to those of preschool children with similar ethnic background. Hence, in the first part of this chapter, we investigated ethnic differences in sleep duration at 5 years among various ethnic groups in Amsterdam using Amsterdam Born Children and their Development (ABCD) Study data between 2003 and 2008. We found that ethnic minority children were more likely than Dutch children to report short sleep duration. We concluded that this finding is consistent with the results in the adult population. The findings of the second part of chapter 3 was reported in the following chapter 4.

Association between sleep duration and CVD risk factors among adults and preschool children of various ethnic groups.

The purpose was to investigate whether there is a relationship between sleep duration and CVD risk factors in both adults and preschool children. Chapter 4 examined the association between short sleep duration and CVD risk factors such as obesity, type 2 diabetes, hypertension and dyslipidaemia (HDL cholesterol, LDL cholesterol, total cholesterol, and increased triglycerides) among ethnic minority groups and ethnic Dutch; as well as the contribution of socioeconomic and lifestyle factors to the observed association. We found that short sleep was associated with obesity in 4 out of 6 ethnic groups; and with diabetes in 3 out of 6 ethnic groups. Whereas, the association of short sleep with hypertension and dyslipidaemia remained ethnic-specific. We concluded that the association of short sleep with obesity and diabetes was partly consistent across ethnic groups, but the association of short sleep with hypertension and dyslipidaemia was inconsistent across ethnic groups. SES and lifestyle factors contribute little to the association between short sleep duration and CVD risk factors between the ethnic groups.

In the last part of chapter 3, we also examined the relationship between short sleep duration and overweight and blood pressure among preschool children S

187

Anujuo_binnenwerk_FINAL.indd 187 30/07/2018 21:19 Summary

of various ethnic groups. We found increased prevalence of overweight among preschool children short sleepers in Dutch and Moroccans only, but the association was no longer significant after adjusting for SES. Short sleep was not associated with systolic blood pressure, or diastolic blood pressure in any of the groups. We concluded that the observed association of short sleep with overweight/obesity and blood pressure/hypertension as reported in adults cannot be duplicated in children.

In chapter 5, we assessed the relationship between short or long sleep duration and arterial stiffness (measured by pulse wave velocity – PWV), and whether the association differed among the groups in both men and women. We found that neither short, nor long sleep was associated with PWV in all the ethnic groups, except for long sleep in Dutch men which was associated with higher PWV (indication of stiffer arteries). We concluded that our study provided no convincing evidence that sleep duration was associated with arterial stiffness among various ethnic groups.

Association between short sleep duration and CVD/ contribution of short sleep to ethnic differences in CVD.

The purpose was to assess the association between short sleep duration and CVD among ethnic groups, and to determine how much short sleep contributes to ethnic differences in CVD. Chapter 6 examined the association between short sleep duration and CVD and the contribution of short sleep to ethnic inequalities in CVD, independent of known traditional risk factors of CVD. We found that short sleep was associated with CVD among all ethnic groups, and contributed to ethnic inequalities in CVD. We conclude that short sleep contributed substantially to ethnic inequalities in CVD, independent of well-known CVD risk factors, especially in Surinamese and Ghanaian groups.

Association between depressed mood and sleep duration in a multi-ethnic cohort/contribution of depression to the prevalence of short or long sleep.

In this final part of the study, we assessed the association between depressed mood and the prevalence of short or long sleep duration among ethnic groups, and to determine how much depression accounted for the prevalence

188

Anujuo_binnenwerk_FINAL.indd 188 30/07/2018 21:19 Summary

of short or long sleep. We found that depressed mood was associated with prevalence of short sleep duration in all ethnic groups, but not with long sleep except in Dutch and Moroccans. Depressed mood accounted substantially for the prevalence of short sleep. We concluded by stressing the importance of management of depression (as inclusive) in strategies aimed to reduce the increased prevalence of short sleep especially in those ethnic minority groups with a high prevalence of depression, such as the Turkish and Moroccan group in the Netherlands.

S

189

Anujuo_binnenwerk_FINAL.indd 189 30/07/2018 21:19 Anujuo_binnenwerk_FINAL.indd 190 30/07/2018 21:19 SAMENVATTING

Anujuo_binnenwerk_FINAL.indd 191 30/07/2018 21:19 Samenvatting

SAMENVATTING

Hart- en vaatziekten (HVZ) zijn nog steeds de belangrijkste veroorzakers van morbiditeit en sterfte wereldwijd. Steeds meer onderzoek toont aan dat etnische minderheidsgroepen belast zijn met een hogere prevalentie van HVZ en hun risicofactoren dan de autochtone populatie. Met de opkomst van slaaponderzoek komt er meer bewijs voor verschillen in het verband tussen slaapduur en HVZ, en de risicofactoren tussen etnische groepen, maar Europese studies over dit onderwerp blijven beperkt. Dit proefschrift onderzocht de associatie tussen slaapduur en HVZ en risico factoren onder etnische minderheidsgroepen in Amsterdam. Het eerste doel van dit proefschrift was het meten van etnische verschillen in slaapduur in volwassenen en kleuters, en welke mogelijke factoren deze verschillen verklaren. Het tweede doel was het meten van de associatie tussen korte slaap en risicofactoren van HVZ bij zowel volwassenen als kleuters. Het derde doel was om vast te stellen of korte slaap geassocieerd was met HVZ en of korte slaap bijdroeg aan etnische verschillen in HVZ, onafhankelijk van andere HVZ-risicofactoren. De laatste doelstelling was het aantonen van een mogelijk verband tussen depressieve stemming en slaapduur bij verschillende etnische minderheidsgroepen. Hoofdstuk 1 beschreef de achtergrond, doelstellingen en hoofdlijnen van dit proefschrift.

Etnische ongelijkheden in slaapduur bij volwassenen en kleuters

In Hoofdstuk 2 zijn etnische verschillen in slaapduur onder volwassenen en kleuters, en de bijdrage van sociaaleconomische status (SES) aan de waargenomen verschillen bestudeerd. De resultaten laten zien dat alle etnische minderheidsgroepen gemiddeld een kortere slaapduur rapporteerden dan autochtonen, in zowel mannen als vrouwen. Bij volwassenen was de prevalentie van korte slaap 2-3 keer hoger in etnische minderheidsgroepen dan in autochtone (Nederlanders), zowel bij mannen als vrouwen. Evenzo was de prevalentie van korte slaap 3-4 keer hoger onder kleuters met migratie achtergrond dan onder Nederlandse kinderen. De prevalentie van lange slaap was 2 keer hoger bij alle vrouwen van etnische minderheden (behalve onder Surinaamse groepen) dan bij Nederlandse vrouwen. De prevalentie van lange slaapduur was ook hoger bij allochtone mannen (behalve bij Afrikaanse Surinamers en Ghanezen) dan in Nederlanders. SES kan de waargenomen etnische verschillen in lange slaapduur verklaren, maar de

192

Anujuo_binnenwerk_FINAL.indd 192 30/07/2018 21:19 Samenvatting

bijdrage van SES aan etnische verschillen in korte slaapduur was klein.

In Hoofdstuk 3 is onderzocht of slaappatronen in de volwassen populatie vergelijkbaar zijn met die van kleuters met een vergelijkbare etnische achtergrond. In het eerste deel van dit Hoofdstuk, worden etnische verschillen in slaapduur van 5-jarigen tussen 2003 en 2008 bestudeerd in de ‘Amsterdam Born Children and their Development’ (ABCD) studie. Het hoofdstuk laat zien dat kinderen van etnische minderheden vaker dan Nederlandse kinderen een korte slaapduur rapporteerden. Dit resultaat is consistent met patronen in de volwassen populatie. De resultaten van het tweede deel van hoofdstuk 3 zijn gerapporteerd in het volgende hoofdstuk 4.

Associatie tussen slaapduur en risicofactoren voor hart- en vaatziekten bij volwassenen en kleuters van verschillende etnische groepen

Hoofdstuk 4 onderzocht de associatie tussen korte slaapduur en risicofactoren voor HVZ zoals obesitas, suikerziekte, hypertensie en dyslipidemie (hoge dichtheid lipoproteïne (HDL) cholesterol, lage dichtheid lipoproteïne (LDL) cholesterol, totaal cholesterol en verhoogde triglyceriden) in etnische minderheidsgroepen en autochtone Nederlanders. Bovendien werd de bijdrage van socio-economische en leefstijlfactoren aan de waargenomen associatie onderzocht. De resultaten laten zien dat korte slaap was geassocieerd met obesitas in 4 van de 6 etnische groepen en met suikerziekte in 3 van de 6 etnische groepen. De associatie van korte slaap met hypertensie en dyslipidemie was daarentegen etnisch specifiek. Concluderend was de associatie van korte slaap met obesitas en diabetes deels consistent in etnische groepen, terwijl de associatie van korte slaap met hypertensie en dyslipidemie inconsistent was tussen etnische groepen heen. Sociaal-economische en leefstijlfactoren dragen weinig bij aan het verband tussen korte slaapduur en HVZ-risicofactoren in etnische minderheidsgroepen.

In het laatste deel van Hoofdstuk 3 hebben we de relatie tussen korte slaapduur en overgewicht en bloeddruk onderzocht in kleuters van verschillende etnische groepen. We vonden enkel een verhoogde prevalentie van overgewicht bij Nederlandse en Marokkaanse kleuters die kort slapen, maar de associatie was niet langer significant na correctie voor SES. In S

193

Anujuo_binnenwerk_FINAL.indd 193 30/07/2018 21:20 Samenvatting

geen van de etnische groepen was korte slaap geassocieerd met systolische bloeddruk of diastolische bloeddruk. We concludeerden dat de associatie van korte slaap met overgewicht/obesitas en bloeddruk/hypertensie zoals waargenomen in volwassenen niet gedupliceerd kan worden bij kinderen.

In hoofdstuk 5 hebben we de relatie tussen slaapduur en arteriële stijfheid (gemeten met pulsgolfsnelheid - PWV) bestudeerd en onderzocht of de associatie verschilde tussen etnische groepen bij zowel mannen als vrouwen. We ontdekten dat in alle etnische groepen zowel korte als lange slaap niet geassocieerd waren met PWV. Nederlandse mannen vormden een uitzondering, omdat langdurige slaap bij hen geassocieerd was met een hogere PWV (indicatie van stijvere slagaders). We concludeerden dat ons onderzoek geen overtuigend bewijs leverde dat slaapduur geassocieerd is met arteriële stijfheid bij verschillende etnische groepen.

Associatie tussen korte slaapduur en HVZ en de bijdrage van korte slaap aan etnische verschillen in HVZ

Hoofdstuk 6 onderzocht de associatie tussen korte slaapduur en HVZ en de bijdrage van korte slaap tot etnische ongelijkheid in HVZ, onafhankelijk van bekende traditionele risicofactoren van HVZ. We ontdekten dat korte slaap geassocieerd was met HVZ bij alle etnische groepen. Bovendien vonden we dat korte slaap bijdraagt aan etnische ongelijkheid in HVZ, onafhankelijk van bekende cardiovasculaire risicofactoren, vooral in Surinaamse en Ghanese groepen.

Verband tussen depressieve stemming en slaapduur in een multi-etnisch cohort en de bijdrage van depressie aan de prevalentie van korte of lange slaap.

In dit laatste deel hoofdstuk 7 van de studie hebben we de associatie tussen depressieve stemming en de prevalentie van korte of lange slaapduur onder etnische groepen onderzocht en om te bepalen in hoeverre depressie verantwoordelijk was voor de prevalentie van korte of lange slaap. We ontdekten dat depressieve stemming geassocieerd was met de prevalentie van korte slaapduur in alle etnische groepen. Depressieve stemming verklaarde een groot deel van de prevalentie van korte slaap. Depressieve stemming was enkel met lange slaap geassocieerd in Nederlanders en Marokkanen.

194

Anujuo_binnenwerk_FINAL.indd 194 30/07/2018 21:20 Samenvatting

We concludeerden door het belang van het verminderen van depressie te benadrukken in strategieën gericht op het verminderen van de verhoogde prevalentie van korte slaap, vooral in die etnische minderheidsgroepen met een hoge prevalentie van depressie, zoals de Turkse en Marokkaanse groep in Nederland.

S

195

Anujuo_binnenwerk_FINAL.indd 195 30/07/2018 21:20 Anujuo_binnenwerk_FINAL.indd 196 30/07/2018 21:20 ACKNOWLEDGEMENTS

Anujuo_binnenwerk_FINAL.indd 197 30/07/2018 21:20 Acknowledgements

I have always aspired to be an academic researcher while growing up. I nursed this ambition throughout the years of my primary, secondary and tertiary education. I embrace the fulfilment of my aspiration now that I have completed my doctorate. Special thanks to all members of my family in Nigeria for their unrelenting prayers and support.

The journey towards completing this PhD would not have been possible if not for the great assistance, support and advice from the following grand academic elites:

First and foremost, I express my heart-felt gratitude to you, Dr Charles Agyemang for your unflinching support and supervision. You discovered my interest in research and gave me the necessary motivation that culminated to my present status. Apart from been my co-promotor, you were also my teacher. I was just an ordinary student with little research experience when I started. You taught me from the scratch and led me to the path of my success story today. What amazes me about you Charles, is how you always provide clear explanations when I am confronted with difficult questions during our regular meetings right from the inception of my research. Your calm and calculative approach towards addressing issues that were not only academic-related, but also of personal concerns unravels excellent innate quality you possess. Thank you for giving me the opportunity to conduct this research, and guiding me to the end of my PhD.

I am very grateful to you, Prof. Karien Stronks for your in-depth supervision throughout my PhD trajectory. You were very visionary to discover my weakness, and with your special skill of imparting knowledge, using demonstrations and illustrations, you practically taught me what I may call “special tricks” to overcome the challenges pertaining to analysis that I may have faced. Sometimes, I wondered how you do think beyond my reasoning whenever you provide feedbacks. How you combined the leadership of the department in the past years with other academic engagements portrays an exemplary qualities you possess. I know for sure, that such qualities shall propel you to greater heights. Thank you for being my promotor.

To Prof. Girardin Jean-Louis, I appreciate your being part of my PhD success story. As an important sleep research expert, you provided useful critical comments and suggestions, alongside other co-authors, that improved the quality and content of manuscripts submitted to journals. Thank you for the collaboration.

198

Anujuo_binnenwerk_FINAL.indd 198 30/07/2018 21:20 Acknowledgements

I wish to thank the doctoral committee for making out time to review my thesis and their role as opponents during my PhD defence. I feel elated to have cherished this moment which has afforded me the opportunity to cross-fertilize ideas with renowned experts in the field.

I also thank Dr Marieke B. Snijder, Dr Bert-Jan van den Born, Prof. Ron Peters, Prof. Gbenga Ogedegbe, Dr Femke Rutters, and Dr Tanja Vrijkotte for all your contributions that led to successful publication of my papers.

To our head of department, Prof. Anton Kunst, I admire the drive you have brought to the department. I have no doubt that your new vision shall widen the horizon of achievements in this department, and I am glad to have been associated with your administration.

To our secretariat colleagues, Henriëtte, Nita and Noor, you have assisted me in different ways. I thank you all. To Dr Marieke Hartman and Dr Mary Nicolaou, those little chats we had provided some relieve! Also greetings and success to all PhD students in our Department. I cannot forget you Lize and Dr Mirte Kuipers, for your help in making the lay-out and design of my thesis.

Special thanks to colleagues whom I shared office with, including Dimitri, Amy, Alieke, Danae, Julia, Lauren, Pichaya, and Wenwei. Also thanks to colleagues next to my office, including Gita, Maria, Nicholas, Wouter, Erik, Hugo and Uchaindani, for common interests we may have shared. To you, Dr Ikenna Ebuenyi and Ben Nieuwenhuizen – my “paranymphs” Thanks for your encouragements! To Bettina, Maryanne and Kirsten of AMC graduate school, thanks for your invaluable advice.

I cannot forget the following friends, Demian Aku, Chinonyerem Uzoma, John Onye, Valentino Meerzorg, Terry Boadum, Timothy Pawironadi, Dr Shahab Bozorgmehri, Dr Marc Hann, mr S. S. Jangali, Abdul-Malik Baldewe, Drs F. Ogbuli, Dr Hugo Eenhoorn, Dr Hadi Zafarmand, Dr Joachim Doua, and Dr Umar Ikram. whose advice and cooperation were priceless throughout the course of my research. To Dr Hellen vd Maazen, I did not forget your advice at the inception of my research. Thank you!

Finally, special thanks to you, Marjella, for your prayers and emotional support. A

199

Anujuo_binnenwerk_FINAL.indd 199 30/07/2018 21:20 Anujuo_binnenwerk_FINAL.indd 200 30/07/2018 21:20 ABOUT THE AUTHOR

Anujuo_binnenwerk_FINAL.indd 201 30/07/2018 21:20 About the author

Kenneth Okwudili Anujuo was born in 1972 in Okohia, Imo State, Nigeria. He completed his secondary education at Secondary Technical School Dikenafai in 1987, and afterwards proceeded to study Science Technology (medical option) at the College of Technology (currently The Polytechnic) Nekede, Owerri in Imo State, Nigeria. In 1988, Kenneth was admitted to Veterinary Medicine and Surgery at the University of Nigeria Nsukka (UNN). He graduated a Veterinary Surgeon in 1995, and completed his National Youth Service Corps in 1996.

Kenneth has vested interest in public health research and education, and to this end, he worked with various organizations with similar interests including Kalvet Agro-allied Services Nig. Ltd, M&G NatureWorks Research Institute, and Corporate Project Education Foundation where he taught Biology and Chemistry to university preparatory-students, all in Nigeria.

In 2007, Kenneth was admitted to master Biomedical Sciences at Vrije University Amsterdam. He opted for International Public Health specialization (IPH) of the Department of Health Sciences, and completed his master in 2009. To further fulfill his research ambition, his proposal to investigate sleep and cardiovascular outcomes among ethnic groups in the Netherlands was accepted in the Department of Public Health, Academic Medical Center (AMC), leading to the commencement of his PhD in 2013, subsequently ending with this book as the end product.

He intends to continue in this area of research so as to make more positive contributions to improving health among various ethnic groups across the globe.

202

Anujuo_binnenwerk_FINAL.indd 202 30/07/2018 21:20 About the author

A

203

Anujuo_binnenwerk_FINAL.indd 203 30/07/2018 21:20 Anujuo_binnenwerk_FINAL.indd 204 30/07/2018 21:20 LIST OF PUBLICATIONS

Anujuo_binnenwerk_FINAL.indd 205 30/07/2018 21:20 List of publications

LIST OF PUBLICATIONS

Articles included in this thesis

Anujuo K, Agyemang C, Snijder MB, Jean-Louis G, van den Born BJ, Peters RJ, Stronks K. Contribution of short sleep duration to ethnic differences in cardiovascular disease: results from a cohort study in the Netherlands. BMJ Open. 2017;7: :e017645. doi: 10.1136/bmjopen-2017-017645.

Anujuo KO, Vrijkotte TGM, Stronks K, Girardin JL, Agyemang CO, Ethnic differences in sleep duration at 5 years, and its relationship with overweight and blood pressure. EUR J Public Health 2016;6:1001-1006. doi: 10.1093/eurpub/ckw084. Epub 2016 Jul 1.

Anujuo K, Stronks K, Snijder MB, Jean-Louis G, van den Born BJ, Peters RJ, Agyemang C, Relationship between sleep duration and arterial stiffness in a multi-ethnic population: The HELIUS study. Chronobiol Int. 2016;33(5):543-52. doi: 10.3109/07420528.2016.1158721. Epub 2016 Apr 8.

Anujuo K, Stronks K, Snijder MB, Jean-Louis G, Rutters F, van den Born BJ, Peters RJ, Agyemang C. Relationship between short sleep duration and cardiovascular risk factors in a multi-ethnic cohort - the HELIUS study. Sleep Med. 2015;16 (12):1482-1488.

Anujuo K, Stronks K, Snijder MB, Jean-Louis G, Ogedegbe G, Agyemang C. Ethnic differences in self-reported sleep duration in the Netherlands - the HELIUS study. Sleep Med. 2014;15 (9):1115-1121.

Anujuo K, Stronks K, Snijder MB, Lok A, Jean-Louis G, Agyemang C. Association between depressed mood and sleep duration among various ethnic groups – the HELIUS study (Submitted).

206

Anujuo_binnenwerk_FINAL.indd 206 30/07/2018 21:20 List of publications

Articles not included in this thesis

Agyemang C, Goosen S, Anujuo K, Ogedegbe G, Relationship between post-traumatic stress disorder and diabetes among 105 180 asylum seekers in the Netherlands. EUR J Public Health. 2012;22 (5):658-662.

Agyemang C, Kunst AE, Bhopal R, Anujuo K, Zaninotto P, Nazroo J, Nicolaou M, Unwin N, van Valkengoed I, Redekop WK, Stronks K, Diabetes prevalence in populations of South Asian Indian and African origins: a comparison of England and the Netherlands. Epidemiology. 2011;22 (4):563-567.

K Anujuo, MB Snijder, K Stronks, C Agyemang, Sleep patterns and sleep problems in a multi-ethnic population in Amsterdam, the Netherlands. European J Public Health. 2013; DOI: 10.1093/eurpub/ckt126.277

L

207

Anujuo_binnenwerk_FINAL.indd 207 30/07/2018 21:20 Anujuo_binnenwerk_FINAL.indd 208 30/07/2018 21:20 PORTFOLIO

Anujuo_binnenwerk_FINAL.indd 209 30/07/2018 21:20 Portfolio

Name PhD student: Kenneth Okwudili Anujuo PhD period: April 2013 – April 2018 Names PhD supervisors: Prof. Dr. K. Stronks Dr C.O. Agyemang

Main Courses Organisation Year ECTS* General courses 12.6 Computing in R AMC 2013 0.4 Systematic Reviews AMC 2013 0.7 AMC World of Science AMC 2013 0.7 Qualitative Health Research AMC 2013 1.9 Clinical Data Management AMC 2014 0.2 Clinical Epidemiology AMC 2014 0.6 Scientific Writing in English for Publication AMC 2014 1.5 Infectious Diseases AMC 2014 1.3 Practical Biostatistics AMC 2016 1.1 Searching for a Systematic Review AMC 2016 0.1 Entrepreneurship in Health and Life Sciences AMC 2016 1.5 Oral Presentation in English AMC 2016 0.8 Citation and Impact factor analysis AMC 2017 0.1 Endnote AMC 2017 0.1 Project Management AMC 2017 0.6 Didactic Skills UVA 2018 0.4 Clinical Epidemiology (Randomized Clinical Trials) UVA 2018 0.6 Seminars, workshops and master classes 4.2 Masterclass Prof. Stevenson N. Goodman (USA) AMC 2016 0.3 Cancer Center Amsterdam Seminar Series AMC 2016 0.3 Amsterdam Public Health annual meeting UVA/VU 2016 0.4 Amsterdam Public Health annual meeting UVA/VU 2017 0.4 2nd Health Behaviour and Chronic Disease UVA/VU 2017 0.4 3rd Health Behaviour and Chronic Disease UVA/VU 2017 0.4 HBCD GECCO workshop UVA/VU 2017 0.4 Global Health/HBCD workshop UVA/VU 2017 0.4 HBCD Health Behaviour and Chronic Disease UVA/VU 2018 0.4 Amsterdam Public Health annual meeting UVA/VU 2018 0.4 Amsterdam Public Health spring meeting UVA/VU 2018 0.4

210

Anujuo_binnenwerk_FINAL.indd 210 30/07/2018 21:20 Portfolio

Main Courses Organisation Year ECTS* Presentations 13 Public Health Department seminar (3x) AMC 2013-2017 4.5 (Inter)national conferences 8.5 European Society of Cardiology Lisbon (oral) ESC 2015 1.5 Oslo Migrant Health (oral) EUPHA 2016 1.5 European Sleep Research Society Bologna ESRS 2016 2.0 (2 posters) Canadian Sleep Society Calgary CSS 2017 2.0 (oral and poster) MERH world congress Edinburgh (oral) EUPHA 2018 1.5 Other 3 2-day training in practical polysomnography Sleep 2017 2.0 Edinburgh Consult Ltd Peer reviewing 1.0 International Journal of Hypertension 2014 0.1 Sleep Medicine 2014 0.1 Sleep Medicine (2 x) 2015 0.2 Canadian Medical Association 2016 0.1 Int. Jour Human Nutrition and Dietetics 2016 0.1 BMJ Open 2016 0.1 Sleep Medicine 2016 0.1 Sleep Medicine 2018 0.1 Biomed and Environmental Sciences 2018 0.1

* European Credit Transfer and Accumulation System: 1 ECTS = 28 hours workload.

P

211

Anujuo_binnenwerk_FINAL.indd 211 30/07/2018 21:20 Anujuo_binnenwerk_FINAL.indd 212 30/07/2018 21:20

THETHE ROLEROLE OFOF SLEEPSLEEP ININ ETHNICETHNIC INVITATIONINVITATION The role of sleep in ethnic inequalities health The role of sleep in ethnic inequalities health INEQUALITIESINEQUALITIES ININ HEALTH:HEALTH: YouYou are are cordially cordially invited invited to to cardiovascularcardiovascular diseasedisease andand riskrisk factorsfactors attendattend the the public public defense defense ofof the the doctoral doctoral thesis thesis of of

KENNETHKENNETH OKWUDILI OKWUDILI ANUJUO ANUJUO

titledtitled

THETHE ROLE ROLE OF OF SLEEP SLEEP IN IN ETHNIC ETHNIC INEQULITIESINEQULITIES IN IN HEALTH: HEALTH: CARDIOVASCULARCARDIOVASCULAR DISEASE DISEASE ANDAND RISK RISK FACTORS FACTORS

OnOn Friday, Friday, September September 21, 21, 2018 2018 AtAt 12:00 12:00 InIn the the Agnietenkapel Agnietenkapel OudezijdsOudezijds Voorburgwal Voorburgwal 229-231 229-231 AmsterdamAmsterdam ReceptionReception follows follows immediately immediately afterafter the the defense defense in in the the reception reception areaarea of of the the Agnietenkapel Agnietenkapel

PARANYMPHS:PARANYMPHS: K K IkennaIkenna Ebuenyi Ebuenyi ENNETH ENNETH [email protected]@vu.nl BenBen Nieuwenhuizen Nieuwenhuizen O O b.p.vannieuwenhuizenb.p.vannieuwenhuizen@@amc.uva.nlamc.uva.nl KWUDILI KWUDILI A A NUJUO NUJUO

KennethKenneth OkwudiliOkwudili AnujuoAnujuo

Anujuo_omslag_FINAL.inddAnujuo_omslag_FINAL.indd All All Pages Pages 30/07/201830/07/2018 21:13 21:13