Division of Medical Genetics, Department of Pediatrics, University of California San Francisco, San Francisco, CA, USA.
Institute for Human Genetics, University of California San Francisco, San Francisco, CA, USA.
Genet Med. 2020 Mar;22(3):640-645. doi: 10.1038/s41436-019-0695-7. Epub 2019 Nov 26.
Clear and accurate genetic information should be available to health-care consumers at an individualized level of comprehension. The objective of this study is to evaluate the complexity of common online resources and to simplify text content using automated text processing tools.
We extracted all text from Genetics Home Reference and MedlinePlus in bulk and analyzed content using natural language processing. We applied custom tools to improve the readability and compared readability before and after text optimization.
Commonly used educational materials were more complex than the recommended reading level for the general public. Genetic health information entries from Genetics Home Reference (n = 1279) were written at a median 13.0 grade level. MedlinePlus entries, which are not exclusively genetic (n = 1030), had a median grade level of 7.7. When we optimized text for the 59 actionable conditions by prioritizing medical details using a standard structure, the average reading grade level improved.
Factors that increase complexity are long sentences and difficult words. Future strategies to reduce complexity include prioritizing relevant details and using more illustrations. Simplifying and providing standardized online health resources would benefit diverse consumers and promote inclusivity.
医疗保健消费者应能在个人理解层面上获得清晰准确的遗传信息。本研究的目的是评估常见在线资源的复杂性，并使用自动化文本处理工具简化文本内容。
我们从 Genetics Home Reference 和 MedlinePlus 中批量提取所有文本，并使用自然语言处理进行内容分析。我们应用自定义工具来提高可读性，并比较文本优化前后的可读性。
常用的教育材料比普通公众的推荐阅读水平更为复杂。Genetics Home Reference（n=1279）中的遗传健康信息条目平均写作水平为 13.0 级。并非专门针对遗传信息的 MedlinePlus 条目（n=1030）的平均写作水平为 7.7 级。当我们通过使用标准结构优先考虑医疗细节来优化针对 59 种可操作条件的文本时，平均阅读水平得到了提高。
增加复杂性的因素是长句子和难词。未来降低复杂性的策略包括优先考虑相关细节和使用更多插图。简化和提供标准化的在线健康资源将使不同的消费者受益，并促进包容性。
