Scholarlink

Background <p>Stroke is a major public health concern in sub-Saharan Africa, including Ghana. Although its clinical and demographic risk factors are well established globally, nationally representative evidence on stroke prevalence and associated factors in Ghana remains limited. Also, previous research using WHO SAGE Wave 1 data has examined stroke prevalence and associated factors among older adults in Ghana; however, important methodological gaps remain, particularly regarding the explicit consideration of the complex survey design and unobserved cluster-level heterogeneity.</p> Aim <p>This study examined the prevalence and risk factors of stroke among adults in Ghana.</p> Methods <p>Data were obtained from the World Health Organization’s Study on Global AGEing and Adult Health (WHO SAGE, Wave 1), comprising 5,066 Ghanaian participants, including individuals aged 50 years and older and a comparative group aged 18–49 years. Descriptive statistics and chi-square tests were used to assess variable distributions and bivariate associations with stroke. Given the hierarchical and complex survey structure of the data, with individuals nested within primary sampling units (PSUs), survey multivariable logistic regression was applied to estimate adjusted associations, while Bayesian multilevel logistic regression models were fitted to account for cluster-level random variation and improve estimate stability. Analyses were conducted in R (v4.5.1) using the survey and R-INLA packages, with statistical significance set at <i>p</i> &lt; 0.05.</p> Results <p>The two models produced consistent findings, with the Bayesian multilevel model offering the most stable and precise estimates after accounting for cluster-level heterogeneity at the PSU level and the within-cluster dependence among sampled individuals. Cluster-level heterogeneity was negligible (ICC = 0.0001; 95% CrI: 0.0000–0.0005), indicating minimal residual variation in stroke across PSUs. The Bayesian multilevel logistic regression analysis identified several significant predictors of stroke. Individuals who had ever attended school had higher odds of stroke compared to those who had not (aOR = 1.48, 95% CrI: 1.00-2.21). Obesity (aOR = 1.68, 95% CrI: 1.05–2.70), arthritis (aOR = 1.74, 95% CrI: 1.08–2.79), diabetes (aOR = 2.67, 95% CrI: 1.53–4.66), and hypertension (aOR = 3.46, 95% CrI: 2.28–5.23) were all associated with increased odds of stroke. Participants aged 50 years and above also had higher odds compared to younger adults (aOR = 2.76, 95% CrI: 1.18–6.44), whereas engagement in vigorous work-related physical activity was associated with reduced odds of stroke (aOR = 0.60, 95% CrI: 0.38–0.95).</p> Conclusion <p>Public health strategies should emphasize physical activity, healthy diets, weight control, and improved management of chronic diseases. These efforts align with the sustainable development goal 3 (SDG 3); Good Health and Well-being, particularly Target 3.4, which aims to reduce premature mortality from non-communicable diseases through prevention and treatment by 2030.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Stroke burden and associated predictors in Ghana: a Bayesian analysis of the WHO study on global ageing and adult health

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月20日

查看全文
  • Abdul-Karim Iddrisu,
  • Adam Mohammed

摘要

Background

Stroke is a major public health concern in sub-Saharan Africa, including Ghana. Although its clinical and demographic risk factors are well established globally, nationally representative evidence on stroke prevalence and associated factors in Ghana remains limited. Also, previous research using WHO SAGE Wave 1 data has examined stroke prevalence and associated factors among older adults in Ghana; however, important methodological gaps remain, particularly regarding the explicit consideration of the complex survey design and unobserved cluster-level heterogeneity.

Aim

This study examined the prevalence and risk factors of stroke among adults in Ghana.

Methods

Data were obtained from the World Health Organization’s Study on Global AGEing and Adult Health (WHO SAGE, Wave 1), comprising 5,066 Ghanaian participants, including individuals aged 50 years and older and a comparative group aged 18–49 years. Descriptive statistics and chi-square tests were used to assess variable distributions and bivariate associations with stroke. Given the hierarchical and complex survey structure of the data, with individuals nested within primary sampling units (PSUs), survey multivariable logistic regression was applied to estimate adjusted associations, while Bayesian multilevel logistic regression models were fitted to account for cluster-level random variation and improve estimate stability. Analyses were conducted in R (v4.5.1) using the survey and R-INLA packages, with statistical significance set at p < 0.05.

Results

The two models produced consistent findings, with the Bayesian multilevel model offering the most stable and precise estimates after accounting for cluster-level heterogeneity at the PSU level and the within-cluster dependence among sampled individuals. Cluster-level heterogeneity was negligible (ICC = 0.0001; 95% CrI: 0.0000–0.0005), indicating minimal residual variation in stroke across PSUs. The Bayesian multilevel logistic regression analysis identified several significant predictors of stroke. Individuals who had ever attended school had higher odds of stroke compared to those who had not (aOR = 1.48, 95% CrI: 1.00-2.21). Obesity (aOR = 1.68, 95% CrI: 1.05–2.70), arthritis (aOR = 1.74, 95% CrI: 1.08–2.79), diabetes (aOR = 2.67, 95% CrI: 1.53–4.66), and hypertension (aOR = 3.46, 95% CrI: 2.28–5.23) were all associated with increased odds of stroke. Participants aged 50 years and above also had higher odds compared to younger adults (aOR = 2.76, 95% CrI: 1.18–6.44), whereas engagement in vigorous work-related physical activity was associated with reduced odds of stroke (aOR = 0.60, 95% CrI: 0.38–0.95).

Conclusion

Public health strategies should emphasize physical activity, healthy diets, weight control, and improved management of chronic diseases. These efforts align with the sustainable development goal 3 (SDG 3); Good Health and Well-being, particularly Target 3.4, which aims to reduce premature mortality from non-communicable diseases through prevention and treatment by 2030.