Towards an understanding of the relationship between sleep and cardiovascular disease risk in adults of African descent living in a low socioeconomic status community

Forshaw, Philippa

Towards an understanding of the relationship between sleep and cardiovascular disease risk in adults of African descent living in a low socioeconomic status community

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Human Biology

Faculty

Faculty of Health Sciences

Abstract

Background: Individuals of African descent, specifically African Americans and those from Sub-Saharan Africa, experience a higher burden of cardiovascular disease (CVD) and its associated risk factors (such as obesity, diabetes and hypertension) as well as shorter, poorer sleep quality, compared to Caucasian American individuals. Blood pressure (BP) non-dipping (i.e. the failure of BP to decrease at night during sleep) and nocturnal hypertension are important markers of CVD risk and are substantially more prevalent in African American, compared to Caucasian American, individuals. Two additional layers that need consideration are socioeconomic status (SES) and sex. In contrast to much of the research in the Global North predominantly demonstrating shorter objective and subjective sleep durations (around 6- 7h per night) among African descended individuals living in low SES environments, South Africans of African descent living in low SES communities report much longer sleep durations, around 8-10h per night. Thus, while on one hand, lower SES has been associated with higher risk for CVD possibly through shorter sleep, the nature of the relationship between long sleep duration and CVD risk in the South African context is not well understood. Given the significant sex-specific differences in both CVD risk and sleep, there is a need for research focused on understanding sex-specific associations between CVD and sleep health. Thus, the purpose of this thesis was to investigate sex-specific relationships between CVD risk, nocturnal BP and sleep health in adults of African descent living in a low SES community in South Africa. This purpose was achieved through the following aims: i) to investigate sex-specific relationships between self-reported sleep characteristics and CVD risk among individuals of African descent living in a low SES community, ii) to investigate sex-specific relationships between actigraphy-derived sleep characteristics and CVD risk in these same individuals, iii) to systematically review the literature on sleep and BP dipping in apparently healthy individuals, iv) to explore sex-specific associations between actigraphy-derived sleep characteristics, nocturnal BP and CVD risk among adults of African descent living in a low SES community and v) to conduct qualitative interviews with these same individuals to explore how external (e.g. environmental barriers to and promoters of good sleep) and internal (e.g. individual knowledge, attitudes, beliefs and perceptions around sleep) factors might impact sleep health. This thesis explored the hypothesis that adverse environmental conditions associated with living in low SES communities are not conducive to healthy sleep, driving BP non-dipping, nocturnal hypertension and higher CVD risk. Methods: For Chapters 2 and 3, individuals of African descent (56% women, 29-51y, 40% employed) living in Khayelitsha (an informal settlement in South Africa characterised by high rates of crime, violence and poverty) were recruited and studied. Sleep characteristics were measured subjectively using self-reported questionnaires (Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), Insomnia Severity Index (ISI); n=412) and objectively with seven days of wrist-worn actigraphy (n=194). CVD risk was assessed using the body mass index (BMI)-modified Framingham 10-year CVD risk score and clinical measures (BMI, waist circumference, resting BP, fasting glucose). We then conducted a systematic review (Chapter 4) exploring associations between BP dipping and sleep in healthy individuals. In Chapter 5 we measured twenty-four hour ambulatory BP in a sub-set of individuals from the original cohort (n=59) to explore associations between BP dipping, nocturnal hypertension, sleep characteristics and CVD risk scores. Finally, we conducted one-on-one qualitative interviews (Chapter 6) in a further sub-set of participants (n=15) to explore possible external (e.g. environmental barriers to and promoters of good sleep) and internal (e.g. individual knowledge, attitudes, beliefs and perceptions around sleep) factors related to sleep health in this population.Results: When we examined associations between subjectively measured sleep and CVD risk (Chapter 2), found that men (n=178) reporting poor sleep quality (PSQI>5, OR: 1.95, 95%CI: 1.07, 3.51, p=0.025) and earlier bedtimes (OR: 0.54, 95%CI: 0.39, 0.74, p<0.001)were more likely to have higher CVD risk scores. Women (n=234) reporting earlier bedtimes (OR: 0.72, 95%CI: 0.55, 0.95, p=0.020) and wake-up times (OR: 0.30, 95%CI: 0.13, 0.73, p=0.007), longer sleep onset latencies (OR: 1.47, 95%CI: 1.43, 1.88, p=0.003), shorter total sleep times (OR: 0.84, 95%CI: 0.72, 0.98, p=0.029), higher PSQI global scores (OR: 1.93, 95%CI: 1.29, 2.90, p=0.001) and more moderate to severe insomnia symptoms (ISI≥15, OR: 3.24, 95%CI: 1.04, 10.04, p=0.042) were more likely to have higher CVD risk scores. We confirm actigraphy derived long (men: 9.4 ± 1.4h, women: 8.9 ± 1.2h) but disturbed sleep (low sleep efficiencies [men: 81.8 (76.8, 85.7)%), women: 79.9 (72.5, 84.6)%], high sleep fragmentation indices [men: 58.3 (52.5, 65.2)%, women: 63.4 (56.3, 68.2)%] and high wake after sleep onset (WASO) times [men: 103.1 (76.1, 127.0)min, women: 84.9 (68.8, 110.4)min]) in this population, for whom obesity (specifically among women: 60.3%) and hypertension (men: 48%, women: 44%) are prevalent. Associations between actigraphy-derived sleep measures and CVD risk (Chapter 3) found that among men (n=94), earlier midsleep time was associated with higher CVD risk scores (b =–0.17, 95%CI:–0.33, –0.02, p=0.030) while shorter sleep duration was associated with obesity (OR: 0.48, 95%CI: 0.25, 0.90, p=0.023). Among women (n=100), earlier wake-up times (b: –0.24, 95%CI: –0.41, –0.07, p=0.007) and midsleep times (b: – 0.18, 95%CI: –0.39, 0.00, p=0.046) were associated with higher CVD risk scores. Women with earlier bedtimes (OR: 0.53, 95%CI: 0.33, 0.85, p=0.009) and midsleep times (OR: 0.47, 95%CI: 0.26, 0.83, p=0.010) were more likely to have elevated BP, and those with earlier wake-up times (OR: 0.54, 95%CI: 0.35, 0.81, p=0.003) and midsleep times (OR: 0.46, 95%CI: 0.27, 0.77, p=0.003) were also more likely to be obese. Interaction effects revealed that among women, CVD risk scores were higher in those who had shorter sleep combined with later bedtimes or in those who had longer sleep combined with earlier bedtimes (β:–2.38, 95%CI: –0.35, –0.12, p<0.001). A weaker interaction effect was found for WASO such that CVD risk score was higher in women with longer sleep and more WASO or shorter sleep with less WASO (β: 0.004, 95%CI: 0.00, 0.00, p=0.014). The systematic review (Chapter 4) showed that BP non dipping in apparently healthy individuals was associated with short sleep duration, more sleep fragmentation, less sleep depth and increased variability in sleep timing. Measuring 24h ambulatory BP (Chapter 5) found a high proportion of SBP non-dipping (men: 50%, women: 61%), with 48% of men and 72% of women also presenting with nocturnal hypertension. Among the women (n=36), shorter total sleep times (rho: 0.42, p=0.020) and worse sleep efficiencies (rho: 0.51, p=0.003) were correlated with smaller SBP dipping percentages. Similarly, shorter sleep durations (rho: 0.39, p=0.029), shorter total sleep times (rho: 0.44, p=0.014) and worse sleep efficiencies (rho: 0.37, p=0.037) were correlated with smaller DBP dipping percentages. Women with worse sleep efficiencies (rho: –0.39, p=0.016) has higher nocturnal SBP. Among the men (n=23), worse sleep efficiencies (SBP rho: –0.47, p=0.024; DBP rho: – 0.50, p=0.015), greater WASO (SBP rho: 0.59, p=0.003; DBP rho: 0.50, p=0.014) and greater sleep fragmentation indices (SBP rho: 0.59, p=0.003; DBP rho: 0.59, p=0.003) were correlated with higher nocturnal SBP and DBP. Worse sleep duration regularity scores were correlated with lower SBP (rho: – 0.48, p=0.025) and DBP (rho: –0.52, p=0.013) dipping percentages. Men with nocturnal hypertension had higher WASO (116.8 (88.8, 163.3)min vs. 88.1 (65.1, 98.3)min, p=0.031) and sleep fragmentation indices (36.4(33.4, 40.8)% vs. 29.6(25.8, 34.7)%, p=0.019) compared to those without nocturnal hypertension. Insights from the qualitative interviews (Chapter 6) revealed that external factors such as high-density living, noise, crime, violence and excessive alcohol use within the community primarily contributed to disturbing the sleep of participants. Conclusions: This thesis provides new insights, from a Global South lens, to relationships between sleep and cardiovascular health as they relate to adults of African descent living in a low SES environment. 4 Two main features of sleep emerge as important risk factors for CVD in these study participants: mistimed sleep and disturbed sleep, despite adequate sleep opportunities. By considering the lived experiences of individuals in this low SES community, we gained an understanding of the major role that the adverse conditions of the neighbourhood has on impairing sleep health in this population. We speculate that this earlier timed sleep observed predominantly in women, but to some extent in men, might be a direct consequence of environment-related fear, prompting residents to seek refuge in bed at a time which may be too early, potentially contributing to circadian misalignment, which in turn may increase CVD risk. We further hypothesise that when faced with these adverse neighbourhood conditions, some residents may be in a state of hypervigilance at night, resulting in insufficient sympathetic nervous system (SNS) withdrawal, which in turn leads to disturbed sleep. Disturbed sleep may contribute to BP non-dipping or nocturnal hypertension, which may subsequently increase CVD risk, potentially through insufficient cardiovascular system recovery at night. Interestingly, we note that some participants appear to demonstrate resilience through attaining healthy sleep despite living in a challenging neighbourhood environment. Perhaps these are the individuals in whom appropriate SNS withdrawal takes place at night, improving their sleep health and reducing their CVD risk. Considering all the evidence generated though these studies, this thesis proposes the term Sleep Health Insecurity - a lack of regular access to healthy sleep (that which is of sufficient duration, regular, appropriately timed, consolidated, satisfying and refreshing), which is essential for optimal mental and physical health, emotional well-being and cognition. Although we propose that residents of Khayelitsha are experiencing Sleep Health Insecurity, which may increase their CVD risk, these residents likely represent not only a large sector of the South African population but also other similar low SES populations around the world, making this concept a fundamental global health issue.