The University of Sydney School of Pharmacy, Faculty of Medicine and Health, The University of Sydney, Australia.
The University of Sydney Library, The University of Sydney, Australia.
J Med Internet Res. 2021 Jul 2;23(7):e21680. doi: 10.2196/21680.
People engage in health information-seeking behavior to support health outcomes, and being able to predict such behavior can inform the development of interventions to guide effective health information seeking. Obtaining a comprehensive list of the predictors of health information-seeking behavior through a systematic search of the literature and exploring the interrelationship of these predictors are critical first steps in this process.
This study aims to identify significant predictors of health information-seeking behavior in the primary literature, develop a common taxonomy for these predictors, and identify the evolution of the concerned research field.
A systematic search of PsycINFO, Scopus, and PubMed was conducted for all years up to and including December 10, 2019. Quantitative studies identifying significant predictors of health information-seeking behavior were included. Information seeking was broadly defined and not restricted to any source of health information. Data extraction of significant predictors was performed by 2 authors, and network analysis was conducted to observe the relationships between predictors with time.
A total of 9549 articles were retrieved, and after the screening, 344 studies were retained for analysis. A total of 1595 significant predictors were identified. These predictors were categorized into 67 predictor categories, with the most central predictors being age, education, gender, health condition, and financial income. With time, the interrelationship of predictors in the network became denser, with the growth of new predictor grouping reaching saturation (1 new predictor identified) in the past 7 years, despite increasing publication rates.
A common taxonomy was developed to classify 67 significant predictors of health information-seeking behavior. A time-aggregated network method was developed to track the evolution of the research field, showing the maturation of new predictor terms and an increase in primary studies reporting multiple significant predictors of health information-seeking behavior. The literature has evolved with a decreased characterization of novel predictors of health information-seeking behavior. In contrast, we identified a parallel increase in the complexity of predicting health information-seeking behavior, with an increase in the literature describing multiple significant predictors.
人们进行健康信息搜索行为是为了支持健康结果，能够预测这种行为可以为干预措施的开发提供信息，以指导有效的健康信息搜索。通过系统地搜索文献来获取健康信息搜索行为的预测因素的综合清单，并探索这些预测因素之间的相互关系，是这一过程的关键第一步。
本研究旨在确定主要文献中健康信息搜索行为的显著预测因素，为这些预测因素制定通用分类法，并确定相关研究领域的发展演变。
对 PsycINFO、Scopus 和 PubMed 进行了系统搜索，检索范围为截至 2019 年 12 月 10 日的所有年份。纳入的研究包括确定健康信息搜索行为显著预测因素的定量研究。信息搜索被广泛定义，不限于任何健康信息来源。由 2 位作者对显著预测因素进行数据提取，采用网络分析观察预测因素随时间的关系。
共检索到 9549 篇文章，经过筛选，保留 344 篇进行分析。共确定 1595 个显著预测因素。这些预测因素被分为 67 个预测因素类别，其中最核心的预测因素是年龄、教育、性别、健康状况和经济收入。随着时间的推移，网络中预测因素的相互关系变得更加紧密，尽管发表率不断提高，但在过去 7 年中，新预测因素分组的增长达到了饱和（仅发现 1 个新预测因素）。
开发了一个通用分类法来对健康信息搜索行为的 67 个显著预测因素进行分类。开发了一种时间聚合网络方法来跟踪研究领域的发展演变，显示出预测健康信息搜索行为的新预测因素的成熟以及报告多个健康信息搜索行为的主要研究的增加。文献的发展伴随着对健康信息搜索行为的新预测因素的特征描述减少。相比之下，我们发现预测健康信息搜索行为的复杂性增加，描述多个显著预测因素的文献也在增加。
