RESUMO
BACKGROUND: An increasing number of users are turning to web-based sources as an important source of health care guidance information. Thus, trustworthy sources of information should be automatically identifiable using objective criteria. OBJECTIVE: The purpose of this study was to automate the assessment of the Health on the Net Foundation Code of Conduct (HONcode) criteria, enhancing our ability to pinpoint trustworthy health information sources. METHODS: A data set of 538 web pages displaying health content was collected from 43 health-related websites. HONcode criteria have been considered as web page and website levels. For the website-level criteria (confidentiality, transparency, financial disclosure, and advertising policy), a bag of keywords has been identified to assess the criteria using a rule-based model. For the web page-level criteria (authority, complementarity, justifiability, and attribution) several machine learning (ML) approaches were used. In total, 200 web pages were manually annotated until achieving a balanced representation in terms of frequency. In total, 3 ML models-random forest, support vector machines (SVM), and Bidirectional Encoder Representations from Transformers (BERT)-were trained on the initial annotated data. A second step of training was implemented for the complementarity criterion using the BERT model for multiclass classification of the complementarity sentences obtained by annotation and data augmentation (positive, negative, and noncommittal sentences). Finally, the remaining web pages were classified using the selected model and 100 sentences were randomly selected for manual review. RESULTS: For web page-level criteria, the random forest model showed a good performance for the attribution criterion while displaying subpar performance in the others. BERT and SVM had a stable performance across all the criteria. BERT had a better area under the curve (AUC) of 0.96, 0.98, and 1.00 for neutral sentences, justifiability, and attribution, respectively. SVM had the overall better performance for the classification of complementarity with the AUC equal to 0.98. Finally, SVM and BERT had an equal AUC of 0.98 for the authority criterion. For the website level criteria, the rule-based model was able to retrieve web pages with an accuracy of 0.97 for confidentiality, 0.82 for transparency, and 0.51 for both financial disclosure and advertising policy. The final evaluation of the sentences determined 0.88 of precision and the agreement level of reviewers was computed at 0.82. CONCLUSIONS: Our results showed the potential power of automating the HONcode criteria assessment using ML approaches. This approach could be used with different types of pretrained models to accelerate the text annotation, and classification and to improve the performance in low-resource cases. Further work needs to be conducted to determine how to assign different weights to the criteria, as well as to identify additional characteristics that should be considered for consolidating these criteria into a comprehensive reliability score.
RESUMO
BACKGROUND: During recent years, Quality of Life (QoL) is a significant assessment factor in clinical trials and epidemiological researches due to the advent of Antiretroviral Therapy (ART), Human Immunodeficiency Virus (HIV) has become a manageable,chronic disease. With regards, more attention must be paid to the QoL of infected patients. Limited evidence exists on the impact of ART on QoL among HIV infected patients. Due to lacking of a systematic approach to summarizing the available evidence on the clinical determinants of People Who Live with HIV/AIDS (PWLHs') QoL, this study aimed to analyze the impact of clinical determinants (ART experience, CD4 count < 200, co-morbidities, time diagnosis and accessibility to cares) on QoL among PWLHs'. METHODS: This study was designed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). PubMed, Science Direct, Web of Science, and Cochrane electronic databases were searched in February 2017 to identify all past studies that discussed social and behavioral characteristics of QoL in PLWHA. To recognize effective factors on social and behavioral QoL, a meta-analysis was conducted. Polled Odds Ratios (ORs) were utilized at a 95% confidence level. Since sampling methods differed between articles in the systematic review, we evaluated pooled estimates using a random effect model. Metan, metareg, metacum, and metabias commands in STATA version 13.0 were applied to analyze the data. RESULTS: Our findings indicated that ART has a positive impact on QoL, with a pooled effect size at approximately 1.04 with a confidence interval between 0.42 to 1.66 which indicates this impact is not very considerable and may be relatively neutral. The pooled effect size for CD4 count on QoL was .29 (95%CI = .22-.35), indicating that there is a negative associate between CD4 count and QoL. The co-morbidity as a negative determinant for QoL among HIV/AIDS infected people. The pooled effect size implies on a relative neutral association, although the confidence interval is wide and ranges between 0.32 to 1.58. The pooled effect size is about 1.82 with confidence interval 1.27 to 2.37 which indicates a considerable positive association with lowest level of heterogeneity. CONCLUSIONS: The results illustrated that time diagnosing and availability to hospital services had significant relationship with a higher QoL and CD4 < 200 was associated with a lower QoL. In conclusion, policy makers should set an agenda setting to provide a suitable diagnostic and therapeutic facilities to early detecting and continues monitoring the health status of People Who Live with HIV/AIDS (PWLHs').