
A Health Information Quality Assessment Tool for Korean Online Newspaper Articles: Development Study
Author(s) -
Naae Lee,
Seung-Won Oh,
Belong Cho,
SeungKwon Myung,
Seung-Sik Hwang,
Goo Hyeon Yoon
Publication year - 2021
Publication title -
jmir. journal of medical internet research/journal of medical internet research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.446
H-Index - 142
eISSN - 1439-4456
pISSN - 1438-8871
DOI - 10.2196/24436
Subject(s) - newspaper , inter rater reliability , reliability (semiconductor) , quality (philosophy) , medicine , psychology , medical physics , statistics , mathematics , rating scale , advertising , power (physics) , physics , philosophy , epistemology , quantum mechanics , business
Background Concern regarding the reliability and accuracy of the health-related information provided by online newspaper articles has increased. Numerous criteria and items have been proposed and published regarding the quality assessment of online information, but there is no standard quality assessment tool available for online newspapers. Objective This study aimed to develop the Health Information Quality Assessment Tool (HIQUAL) for online newspaper articles. Methods We reviewed previous health information quality assessment tools and related studies and accordingly developed and customized new criteria. The interrater agreement for the new assessment tool was assessed for 3 newspaper articles on different subjects (colorectal cancer, obesity genetic testing, and hypertension diagnostic criteria) using the Fleiss κ and Gwet agreement coefficient. To compare the quality scores generated by each pair of tools, convergent validity was measured using the Kendall τ ranked correlation. Results Overall, the HIQUAL for newspaper articles comprised 10 items across 5 domains: reliability, usefulness, understandability, sufficiency, and transparency. The interrater agreement for the article on colorectal cancer was in the moderate to substantial range (Fleiss κ=0.48, SE 0.11; Gwet agreement coefficient=0.74, SE 0.13), while for the article introducing obesity genetic testing it was in the substantial range, with values of 0.63 (SE 0.28) and 0.86 (SE 0.10) for the two measures, respectively. There was relatively low agreement for the article on hypertension diagnostic criteria at 0.20 (SE 0.10) and 0.75 (SE 0.13), respectively. Validity of the correlation assessed with the Kendall τ showed good correlation between tools (HIQUAL vs DISCERN=0.72, HIQUAL vs QUEST [Quality Evaluation Scoring Tool]=0.69). Conclusions We developed a new assessment tool to evaluate the quality of health information in online newspaper articles, to help consumers discern accurate sources of health information. The HIQUAL can help increase the accuracy and quality of online health information in Korea.