Factors Predicting Health Information Seeking and the Construction of a Cognitive Model: Based on the Meta-Analysis Approach

健康資訊乃涉及大眾日常生活之醫學及健康照護相關資訊，而健康資訊尋求行為相關研究則多以個體處於健康受到威脅、參與醫療決策，或行為改變與預防性行為之情境，探討個體如何、為何及從何種管道取得健康資訊，偏好為何及如何使用等。由於此議題廣受醫護、公共衛生、圖書資訊及傳播學等領域之學者的重視，故累積了豐碩的研究成果。然而，相關研究結論分歧，相同預測因素可能在不同情境中呈現截然不同的預測效果，相關理論模型僅適用於特定情境，應用性十分有限，亟需系統性地回顧及分析以彌平紛爭，或發展出一般性的健康資訊尋求理論模型。 基此，本研究採統合分析法，完整地蒐集健康資訊尋求行為相關實徵研究，歸納出媒體使用、患病經驗、健康情形、知識素養、自我效能、情緒反應、有用性、易用性、可信度、主觀規範、搜尋意圖等11項因素，常用於預測健康資訊尋求行為。計算各因素之效果量後發現，除了健康情形之外，其餘各項因素皆能顯著預測尋求行為，並以主觀規範具有較高的預測效果。各研究文獻或因樣本性質、研究特色、文獻性質及研究方法等面向之變數的不同，造成各因素之預測力呈現顯著差異；惟未發現情緒反應之調節變數，仍待後續研究之探討。此外，基於重要預測變數之分析結果，本研究進一步建構認知取徑之健康資訊尋求統合模型，以個人及環境兩個面向之因素來解釋健康資訊尋求行為。模型檢驗結果顯示，健康資訊尋求統合模型之各項適配度指標皆通過考驗，且多項假設均獲得支持。意即不論是個體特徵為何，透過何種資訊來源，尋求何種主題之健康資訊，健康資訊尋求統合模型可用於解釋上述各類情境之健康資訊尋求行為。 健康資訊需求常引發情緒反應，故部分研究結果指出，個體有時會因為避免更多的負面情緒而迴避資訊，在本研究中雖未獲得證實，但的確發現人們並非總是希望更多的資訊。效果量計算或是模型建構結果均顯示，健康資訊尋求深受情境之相關因素的影響，若欲改變個體之健康資訊尋求行為，經由其親友或意見領袖之影響則較容易達成，故形塑追求健康生活之社群信念便顯得格外重要。透過資訊素養之培養、網路健康資訊之可信度的提升，亦有助於促進健康資訊尋求之意願。Health information is any information related to medicine and health care in everyday life of individuals, health information seeking is studied within the context of “coping with a health-threatening situation,” “participation and involvement in medical decision making,” and “behavior change and preventive behavior”. Researchers are interested in understanding how and why individuals obtain health information, where they retrieve, what particular types of information they prefer, and how the health information is used. Thus, studies relevant to health information seeking (HIS) are abundant and wild spread in various disciplines—such as medicine, nursing, public health, communication, and library and information science. However, the empirical findings of factors’ predicting abailty on HIS are mixed, same factor represented totally different predicting ability within diverse context, theories or models with limited generality only applicable to specific situation. A systematic review and analysis to eliminate these controversies and develop a general theory or model of HIS is crucial. Therefore, meta-analysis was adopted in this study, empirical studies related to HIS was completely retrieved, eleven factors such as media use, ill experience, health status, self-efficacy, health literacy, seeking intention, affective responses, usefulness, easy to use, credibility, and subjective norms commonly used in HIS prediction were inducted, every factor has significant predicting ability on HIS in addition to health status and subjective norms is the highest one according to the result of effect size calculation. Variable attributed to different dimension such as sample property, study characteristic, publication attribute, and research method has moderating effect on HIS prediction, but no moderator of affective responses has been found and further study is needed. Besides, based on the result of effect size calculation, a cognitive model of HIS was proposed, factors that attributed to personal and situational dimension were used to annotate HIS. All goodness of fit were passed and most hypothesis were supported of the meta-analystical model of HIS, indicated this model is applicabile to all kinds of individuals, information channels, or subjects of health information. Since health information need leads to affective responses all the time, some empirical evidences suggested that individuals may avoiding information to prevent more negative effect. Althought this point of view has not been confirmed but people not always need more information was found in this study. HIS is highly associated with contextual factors based on the result of effect size calculation or model construction. It is much easier to modify HIS behavior of individuals by means of their friends, family members, or opinion leaders, so cultivating community belief of health life pursuit is particularly important. We can promote HIS intention through the developing of information literacy and credibility elevating of online health information as well