Abstract
Background:
Sleep-related symptoms are known to influence cardiovascular health (CVH), but the relative importance of sleep duration versus quality remains uncertain. This study evaluates whether a combined measure – “satisfactory sleep health,” incorporating both adequate sleep duration and good sleep quality – better correlates with CVH status in middle-aged and older adults living near the Equator.
Methods:
In this population-based, cross-sectional study, 1781 middle-aged and older adults living in rural villages of coastal Ecuador were assessed for sleep duration and quality using structured interviews and the Pittsburgh Sleep Quality Index (PSQI). Satisfactory sleep health was defined as 7- to 8 h of sleep plus a PSQI score ≤5 points. CVH status was evaluated using the American Heart Association’s Life’s Simple 7 construct. Multivariate logistic regression models were used to determine the investigated associations, after adjustment for relevant confounders.
Results:
While adequate sleep duration alone showed no significant association with CVH, good sleep quality, and satisfactory sleep health were inversely associated with poor CVH in unadjusted models. After adjustment for demographics, level of education, and symptoms of depression, satisfactory sleep health was associated with a 21% reduction in the odds of poor CVH status (OR = 0.79; 95% CI = 0.64-0.99; P < 0.05).
Conclusions:
A combined assessment of sleep duration and quality provides a more robust predictor of CVH than either component alone. These findings support the inclusion of multidimensional sleep metrics in public health frameworks evaluating cardiovascular risk and highlight their potential utility in guiding targeted prevention strategies.
Introduction
Studies associating sleep-related symptoms with adverse cardiovascular outcomes – arterial hypertension, atherosclerosis, myocardial infarction, and stroke – mainly focused on sleep duration.1 -4 Prior research has mostly examined self-reported sleep duration as the primary or sole sleep determinant of these outcomes, often relying on non-validated surveys that address only a limited subset of sleep symptoms. However, asking how many hours a person sleeps per night may be prone to inaccuracies due to seasonal variations, time changes, or recall bias, which contribute to heterogeneity across studies. 5 Moreover, perceived sleep duration can reflect unrelated comorbid conditions. 6 If sleep-related symptoms are going to be used for cardiovascular risk evaluation, assessments should utilize structured and validated instruments that encompass multiple dimensions of sleep, not duration alone.
Comprehensive measurements of sleep disturbances may better predict cardiovascular risk than the simplistic approach of a single question about sleep duration.7 -9 Although a recent meta-analysis reported stronger associations for long sleep duration with cardiovascular events and mortality, 10 it lacked standardized measures for sleep quality, which may have skewed the conclusions.
It remains uncertain whether sleep quality or a better construct is more strongly associated with adverse vascular outcomes than sleep duration. This study evaluates the relative strength of associations between sleep duration, sleep quality, and cardiovascular health (CVH) status in middle-aged and older adults living in rural communities near the Equator, where uniform exposure to environmental conditions supports a more stable evaluation of sleep determinants.
Methods
Study Population
The study involved residents from 3 rural villages in coastal Ecuador. Their inhabitants share a homogeneous profile in terms of ethnicity (Amerindian ancestry), educational attainment, socioeconomic status, and dietary patterns.11,12 Additionally, these communities experience 12 h of daylight year-round with minimal nighttime light pollution and limited shift work, creating a setting that reduces circadian disruptions and facilitates consistent sleep assessments.
Study Design
Community-dwellers aged ≥40 years residing in the targeted communities were identified through door-to-door surveys from 2012 to 2019. Inclusion required completion of sleep questionnaires, CVH metrics assessments, and evaluations of key covariates. Those interviews were carried out through face-to-face interviews by trained rural doctors. Written informed consent was obtained from all participants, and the study was approved by an internationally accredited Ethics Committee.
Assessment of Sleep-Related Symptoms
Nighttime sleep duration and sleep quality were evaluated through face-to-face interviews. Sleep duration was assessed using a single item asking participants how many hours they typically sleep at night, with adequate duration defined as 7 to 8 h. 13 Sleep quality was measured using the Pittsburgh Sleep Quality Index (PSQI), a validated instrument comprising 18 items across 7 components: duration, disturbances, latency, daytime dysfunction, efficiency, perceived quality, and sleep medication use. 14 Using this scale, a score of ≤5 indicated good sleep quality. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI), a widely validated instrument for evaluating subjective sleep disturbances over the preceding month. The PSQI has demonstrated strong psychometric properties across diverse populations, including older adults and Spanish-speaking cohorts, with acceptable internal consistency and construct validity. 15 Although sleep duration is embedded within the PSQI, each construct was analyzed separately. To improve measurement precision, we developed a composite measure of “satisfactory sleep health”, defined as both adequate duration and good sleep quality, following precedent from prior research. 8
Evaluation of Cardiovascular Health Metrics
CVH metrics determinations used the Life’s Simple 7 construct of the American Heart Association (AHA), 16 which considers 7 metrics in the poor range, including: (1) smoking status: current or quit <1 year; (2) body mass index: ≥30 kg/m2; (3) no moderate or intense physical activity; (4) poor diet: regular consumption of none or only 1 component of the suggested AHA healthy diet; (5) blood pressure: ≥140/≥90 mmHg; (6) fasting glucose: ≥126 mg/dL; and (7) total cholesterol blood levels: ≥240 mg/dL. This construct recognized 3 categories of CVH status: (1) ideal, if all the 7 metrics were in the ideal range; (2) intermediate, if metrics were in the ideal or intermediate range but there were no metrics in the poor range; and (3) poor, if at least 1 metric was in the poor range. We did not use the Life’s Essential 8, the most recent construct of the AHA for assessing cardiovascular health status, 17 since the data for study participants were collected at the study entry, well before the introduction of this new framework.
Covariables of Interest
Based on previous studies, demographics, educational attainment, and symptoms of depression were used as clinical covariables.11,12 Demographics and level of education were assessed by self-report, and symptoms of depression by the depression axis of the Depression-Anxiety-Stress Scale-21 (DASS-21). 18
Statistical Analysis
Data analysis was carried out using STATA version 19 (College Station, TX, USA). In unadjusted analyses, continuous variables were compared by linear models and categorical variables by the chi-square or Fisher exact test, as appropriate. Several logistic regression models were fitted to estimate Odds Ratios (OR) with their 95% confidence intervals for the association between each of the 3 sleep determinations of interest (sleep duration, sleep quality, and sleep health) as separate independent variables and the total CVH status, which was treated as the dependent variable. Multivariable models were initially adjusted for age and sex, and then for all the above-mentioned clinical covariables. To get more insights on the association between sleep parameters and the CVH status, different models were fitted using each of the CVH metrics as separate dependent variables. The study was designed to have enough statistical power to detect a protective effect (Odds Ratio) of 0.82. Missing values were not imputed, and complete data was utilized for the logistic regression models.
Results
Of 1838 individuals aged ≥40 years enrolled in a parent population-based cohort study, 1781 (97%) completed all the required face-to-face interviews and procedures for assessing sleep-related symptoms, CVH status, and covariables of interest. The remaining 57 individuals, 8 died or emigrated between enrollment and the invitation to participate in this study, 12 declined consents, and 37 were unable to complete the tests due to aphasia or severe cognitive impairment. There were no differences in the mean age (P = .138), the percentage of women (P = .425), or the level of education (P = .264) across participants who completed the tests compared to those who were excluded.
The mean (±SD) age of 1781 study participants was 55.8 ± 12.5 years (median age: 54 years), 1001(56%) were women, 1025 (58%) had primary school education only, and 204 (11%) participants had symptoms of depression. The average nighttime sleep duration for the entire population was 6.8 ± 1.4 h, with 843 (47%) participants classified under the category of adequate sleep duration, while 812 (46%) were categorized as short sleepers and 126 (7%) as long sleepers. A total of 1060 (60%) participants had good sleep quality. A satisfactory sleep health (combination of adequate sleep duration and good sleep quality) was noted in 617 (35%) participants.
Poor CVH metrics included: smoking status: 91 (5%); body mass index: 543 (30%); physical activity: 140 (8%); diet: 309 (17%); blood pressure: 518 (29%); fasting glucose: 426 (24%); and total cholesterol blood levels: 200 (11%). A total of 1281 (72%) had a poor CVH status (at least 1 metric in the poor range).
Table 1 depicts characteristics of study participants across the different sleep determinations. As noted, there were differences across sleep-related symptoms and the investigated covariables. More importantly, good sleep quality and satisfactory sleep health were associated with poor CVH status, whereas adequate nighttime sleep duration was not. Unadjusted logistic regression models confirmed the significant inverse associations between good sleep quality (OR = 0.80; 95% CI = 0.64-0.99), and satisfactory sleep health (OR = 0.78; 95% CI = 0.63-0.96), with poor CVH status. The association was non-significant for adequate sleep duration (OR = 0.95; 95% CI = 0.77-1.17).
Characteristics of 1781 Study Participants According to Sleep Determinations of Interest (Unadjusted Analyses).
Statistically significant results.
When these models were adjusted for all the aforementioned covariables, only satisfactory sleep health was associated with a 21% reduction in the odds of poor CVH status (OR = 0.79; 95% CI = 0.64-0.99; P = .038; P < .05). However, the association between good sleep quality and adequate sleep duration were not significant; increasing age and symptoms of depression remained significant (Figure 1). Interaction models did not indicate any effect modification by independent variables or between these variables and the confounders.
Forest plot showing the association for each of the sleep variables included in this study and a poor cardiovascular health status (dependent variable), after adjustment for confounders.
Discussion
This study showed, in unadjusted analyses, that good sleep quality but not adequate sleep duration, is inversely associated with poor CVH status. When both variables were combined into a single metric (satisfactory sleep health), the association became stronger, suggesting that sleep duration had no independent impact on the cardiovascular risk. After adjustment, satisfactory sleep health was the sole sleep variable to demonstrate an inverse significant association with poor CVH status.
Several studies have addressed the strength of the association between both sleep duration and quality, and cardiovascular risk factors or cardiac events. For a valid comparison, both sleep variables must have been investigated in the same cohort. However, results have been heterogeneous because of diverse study designs and the use of non-validated instruments to define good sleep quality.
A recent population-based study relied on the PSQI as a comprehensive instrument to assess sleep-related symptoms. This study included more than 6000 adults living in Singapore and aimed to determine whether sleep duration, sleep quality, or a combination of both has a stronger association with various physical and mental health disorders. 19 The study results indicated that the investigated disorders were more closely associated with sleep quality than with sleep duration. Drawbacks of this study, however, included the use of sleep-related symptoms as dependent – instead of independent – variables and the so-called “multiple comparisons problem” which occurs when multiple statistical tests are fitted simultaneously, increasing the possibility of false-positive results. Similarly, a meta-analysis of longitudinal studies revealed that poor sleep quality instead of inadequate sleep duration is associated with increased risk of coronary heart disease. 10 However, most of the 11 studies included in this meta-analysis did not use a validated instrument for assessing sleep-related symptoms and the information might be misleading.
While Mendelian randomization studies have suggested a causal link between short sleep duration and increased cardiovascular risk,20,21 our findings diverge by showing that sleep duration alone was not a significant predictor of poor CVH status in this rural equatorial population. This discrepancy may reflect contextual differences, including environmental stability, cultural sleep practices, and the aforementioned limitations of self-reported sleep duration. Moreover, our results support the notion that integrating sleep quality with duration into a composite metric – as we did with the framework “satisfactory sleep health” – may offer a more nuanced and clinically relevant assessment of cardiovascular risk. Similarly, while it has been proposed that catch-up sleep may be associated with reduced cardiovascular risk in sleep-deprived individuals, 22 such compensatory patterns may be less prevalent or differently expressed in our cohort, warranting further investigation.
Pathogenic mechanisms linking non-breathing sleep disorders to cardiovascular health have been investigated but remain partially understood. One proposed pathway involves the increased production of stress hormones, such as cortisol, often triggered by sleep disruptions. 23 Elevated cortisol levels can contribute to an increased risk of cardiovascular risk factors by exacerbating arterial hypertension, promoting insulin resistance, and facilitating the development of metabolic syndrome. Moreover, sleep disorders are thought to impair endothelial function, a major determinant of vascular health. Endothelial dysfunction, characterized by reduced nitric oxide availability and vascular inflammation, may increase susceptibility to arterial hypertension, atherosclerosis, and other vascular conditions.24,25 In addition, disturbed sleep has been associated with elevated expression of inflammatory cytokines, including tumor necrosis factor-alpha and interleukin-6, which play central roles in chronic inflammation and the progression of cardiovascular diseases.23,24
While the present study emphasizes the link between non-breathing sleep disorders and cardiovascular risk factors, several other conditions – including mental health disorders, cognitive decline, and immune system abnormalities – have also been associated with disturbed sleep.6,8 These findings underscore the need for a timely identification and management of sleep-related symptoms to mitigate risk factors that threaten heart and brain health, even in asymptomatic or pauci-symptomatic individuals. Intervention strategies should adopt a comprehensive approach, addressing various aspects of sleep beyond just sleep duration.
This study has limitations. The cross-sectional design precludes the assessment of causality. We did not measure cortisol levels or biomarkers of inflammation, which may be relevant for the studied associations. Our results may have limited applicability to other settings, potentially impacting their generalizability. As previously mentioned, studied communities experience 12 h of daylight year-round with minimal nighttime light pollution and limited shift work, which may not apply to other populations. In addition, we did not rely on objective measurement of sleep-related symptoms, such as those provided by actigraphy or polysomnography. These limitations are balanced by several strengths of our study, including the population-based design with an unbiased participant selection process, the homogeneity of the characteristics of the study population, and the systematic evaluation of sleep-related symptoms using uniform and standardized protocols.
In conclusion, our findings support the utility of multidimensional sleep assessments in cardiovascular risk stratification. These results argue against relying solely on sleep duration in public health screening or research frameworks. Incorporating validated instruments such as the PSQI may enhance the accuracy of risk prediction models and inform intervention strategies, particularly in low-resource and environmentally stable populations. Further longitudinal research should explore causal mechanisms and elucidate underlying pathophysiologic mechanisms. These will inform personalized preventive strategies in diverse populations.
Footnotes
Author Contributions
OHD: study design and manuscript drafting; RMM: statistical analysis and significant intellectual contribution to manuscript content; EEA: data collection and analysis; DAR: supervision and data collection and analysis; PCR: study design and significant intellectual contribution to manuscript content.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Universidad Espíritu Santo – Ecuador. The sponsor had no role in the design of the study, in the collection, analysis and interpretation of data, or in the decision to submit the manuscript for publication.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.