Clinical and Experimental Optometry Research Lab (CEORLab), Center of Physics - School of Sciences, University of Minho, Minho, Portugal.
J Optom. 2023 Jul-Sep;16(3):182-188. doi: 10.1016/j.optom.2022.07.003. Epub 2022 Aug 29.
The aim of this study was to systematically review and meta-analyse epidemiological data of refractive error prevalence in Portugal.
A structured search strategy and systematic literature review was applied to multiple databases, such as MEDLINE/PubMed, Web of Science, Scopus, Google Scholar, official organizations and academic repositorium's, to identify all relevant epidemiological studies in Portugal until February 2021. The outcome measure was the prevalence of refractive error among the Portuguese population. The events and sample size were entered as raw data and the effect size parameters were computed by Comprehensive Meta-Analysis Software.
A total of 9 studies were pooled for the meta-analysis. The fixed effects model points to an estimated effect size of 43% (95% CI: 41.9-44.1%). However, the statistics of heterogeneity (Q-value p < 0.001; I-squared =99.344) showed very high heterogeneity among studies and recommends using a random-effects model. The random effects model points to an estimated effect size of 31.9% (95% CI: 19.8-47.0%) prevalence of refractive error in the Portuguese population.
A prevalence of refractive error in Portugal of 31.9% (95% CI: 20.0-47.0%) can be considered as a conservative approach to the real burden of this condition. However, it translates into at least 2 to 4.5 million Portuguese individuals with a refractive error. The high heterogeneity between studies, the wide estimate and the random effects involved demonstrate the need for more studies and consistent sources to obtain narrower estimates.
本研究旨在系统回顾和荟萃分析葡萄牙屈光不正患病率的流行病学数据。
采用结构化搜索策略和系统文献回顾，对多个数据库（如 MEDLINE/PubMed、Web of Science、Scopus、Google Scholar、官方组织和学术资源库）进行检索，以确定截至 2021 年 2 月葡萄牙所有相关的流行病学研究。主要结局指标为葡萄牙人群中屈光不正的患病率。将事件和样本量作为原始数据输入，使用 Comprehensive Meta-Analysis Software 计算效应大小参数。
共有 9 项研究纳入荟萃分析。固定效应模型估计的效应大小为 43%（95%CI：41.9-44.1%）。然而，异质性统计（Q 值 p<0.001；I-squared=99.344）表明研究之间存在高度异质性，建议使用随机效应模型。随机效应模型估计的屈光不正患病率为 31.9%（95%CI：19.8-47.0%）。
葡萄牙屈光不正的患病率为 31.9%（95%CI：20.0-47.0%），这可以被认为是对该疾病真实负担的保守估计。然而，这意味着至少有 200 万到 450 万葡萄牙人患有屈光不正。研究之间存在高度异质性、广泛的估计值和随机效应表明，需要进行更多的研究和采用一致的来源，以获得更窄的估计值。
