Background Studies in Northern America report a lower prevalence of coronary artery disease (CAD) in Black than White people despite higher prevalence of vascular risk factors. However, it remains unclear to what extent these differences are driven by biology or reflect ethnic disparities in smoking, socioeconomic deprivation, healthcare access, and health-seeking behaviour. We aimed to determine if ethnic differences in coronary and peripheral artery disease persist in people with stroke living in the United Kingdom (UK), considering that most people with stroke seek medical attention and the National Health Service provides equitable access to care. Methods In a prospective population-based cohort of all patients with suspected TIA or stroke in Oxfordshire, UK (Oxford Vascular Study, OxVasc, April 2002 to March 2023), we fitted multivariable logistic regression models to quantify the association of ethnicity and extracranial atherosclerotic disease (coronary or peripheral) while adjusting for age, sex, vascular risk factors, and socioeconomic deprivation. We verified our findings in a systematic review of UK-based studies reporting prevalence of extracranial atherosclerotic disease and associated vascular risk factors across ethnic groups. Prevalence estimates were pooled by random-effects meta-analysis. Results In 7297 OxVasc participants (52.6% women, 71.0 ± 15.5 years), non-White people (n = 335) had a higher prevalence of diabetes mellitus (OR = 3.26, 2.55–4.16). However, Black people had a lower prevalence of CAD (OR = 0.31, 0.10–0.98, p = 0.046) or any extracranial atherosclerotic disease (OR = 0.42, 0.18–0.96, p = 0.04) than White people. After adjustment for age, sex, vascular risk factors, and deprivation, the difference was attenuated both for CAD (aOR = 0.50, 0.18–1.41, p = 0.19) and any extracranial atherosclerotic disease (aOR = 0.43, 0.18–1.04, p = 0.06). Asian and White people had similar prevalence of CAD and extracranial atherosclerotic disease. Among UK-based studies of patients with stroke, Black and Asian people were more likely than White people to have diabetes (pooled OR: 2.78, 2.40–3.22, n = 3 studies; 4.15, 3.11–5.53 n = 6), hypertension (1.95, 1.47–2.60, n = 3; 1.47, 1.02–2.12, n = 6), and obesity (1.52, 1.27–1.83, n = 1; 1.2, 0.66–2.29, n = 3) but less likely to be smoking (0.70, 0.50–0.97, n = 2; 0.57, 0.44–0.74, n = 5). Yet, Black people had less CAD than White (0.48, 0.31–0.76, n = 3) and Asian (0.30, 0.10–0.80, n = 2) people. Interpretation We confirm the existence of a coronary artery disease paradox in Black but not Asian people with TIA or stroke, which is likely independent of biases in healthcare access or health-seeking behaviour, and may suggest different biological susceptibilities.
Background In an era in which time is increasingly read from digital devices, the Clock Drawing Test (CDT) may capture not only cognitive impairment but also construct-irrelevant variance associated with reduced analog clock usage. Objective We aimed to derive updated normative data for CDT while modeling the impact of digital-like (DL) contamination on score interpretation. Methods This cross-sectional study enrolled cognitively healthy adults aged 18–94 years. Clock drawings were scored using six established methods. DL patterns were identified, including five-minute sequencing (e.g., 5–55) and post-meridiem mapping (e.g., 13–23). We then examined whether the frequency of digital timekeeping was associated with CDT performance and the probability of relying on DL strategies. Psychometric rules for score adjustment and interpretation were computed. Results The sample included 560 participants (357 women). The scoring methods showed moderate-to-strong intercorrelations (ρ = 0.45–0.79), with good-to-excellent inter-rater reliability (ICC = 0.78–0.95). DL patterns occurred in 45 individuals (8%) and were not associated with cognitive functioning. As participants' age increased, the more frequent the use of digital devices for time-checking, the poorer the CDT performance (b = −0.051, 95% CI [−0.086, −0.017]) and the greater the likelihood of producing DL patterns (b = 0.052, 95% CI [0.017, 0.087], OR = 1.053, 95% CI [1.017, 1.091]). Both age and DL contamination emerged as consistent predictors of CDT performance. Conclusions The clinical validity of CDT may be deteriorating. This study provides updated norms with optional DL adjustments. Their effectiveness in decreasing misclassifications requires diagnostic studies involving clinical samples.
