BackgroundResearch integrity and research fairness have gained considerable momentum in the past decade and have direct implications for global health epidemiology. Research integrity and research fairness principles should be equally nurtured to produce high-quality impactful research—but bridging the two can lead to practical and ethical dilemmas. In order to provide practical guidance to researchers and epidemiologist, we set out to develop good epidemiological practice guidelines specifically for global health epidemiology, targeted at stakeholders involved in the commissioning, conduct, appraisal and publication of global health research.MethodsWe developed preliminary guidelines based on targeted online searches on existing best practices for epidemiological studies and sought to align these with key elements of global health research and research fairness. We validated these guidelines through a Delphi consultation study, to reach a consensus among a wide representation of stakeholders.ResultsA total of 45 experts provided input on the first round of e-Delphi consultation and 40 in the second. Respondents covered a range of organisations (including for example academia, ministries, NGOs, research funders, technical agencies) involved in epidemiological studies from countries around the world (Europe: 19; Africa: 10; North America: 7; Asia: 5; South-America: 3 Australia: 1). A selection of eight experts were invited for a face-to-face meeting. The final guidelines consist of a set of 6 standards and 42 accompanying criteria including study preparation, protocol development, data collection, data management, data analysis, dissemination and communication.ConclusionWhile guidelines will not by themselves guard global health from questionable and unfair research practices, they are certainly part of a concerted effort to ensure not only mutual accountability between individual researchers, their institutions and their funders but most importantly their joint accountability towards the communities they study and society at large.

Objective This systematic review aims to assess how information from unstructured text is used to develop and validate clinical prognostic prediction models. We summarize the prediction problems and methodological landscape and determine whether using text data in addition to more commonly used structured data improves the prediction performance. Materials and Methods We searched Embase, MEDLINE, Web of Science, and Google Scholar to identify studies that developed prognostic prediction models using information extracted from unstructured text in a data-driven manner, published in the period from January 2005 to March 2021. Data items were extracted, analyzed, and a meta-analysis of the model performance was carried out to assess the added value of text to structured-data models. Results We identified 126 studies that described 145 clinical prediction problems. Combining text and structured data improved model performance, compared with using only text or only structured data. In these studies, a wide variety of dense and sparse numeric text representations were combined with both deep learning and more traditional machine learning methods. External validation, public availability, and attention for the explainability of the developed models were limited. Conclusion The use of unstructured text in the development of prognostic prediction models has been found beneficial in addition to structured data in most studies. The text data are source of valuable information for prediction model development and should not be neglected. We suggest a future focus on explainability and external validation of the developed models, promoting robust and trustworthy prediction models in clinical practice.

Introduction: A 6-month isoniazid as tuberculosis preventive therapy (TPT) for people living with HIV (PLHIV) was nationally introduced in Eritrea in 2014. However, its effectiveness in

Objective: This systematic review aims to assess how information from unstructured clinical text is used to develop and validate prognostic risk prediction models. We summarize the prediction problems and methodological landscape and assess whether using unstructured clinical text data in addition to more commonly used structured data improves the prediction performance. Materials and Methods: We searched Embase, MEDLINE, Web of Science, and Google Scholar to identify studies that developed prognostic risk prediction models using unstructured clinical text data published in the period from January 2005 to March 2021. Data items were extracted, analyzed, and a meta-analysis of the model performance was carried out to assess the added value of text to structured-data models. Results: We identified 126 studies that described 145 clinical prediction problems. Combining text and structured data improved model performance, compared to using only text or only structured data. In these studies, a wide variety of dense and sparse numeric text representations were combined with both deep learning and more traditional machine learning methods. External validation, public availability, and explainability of the developed models was limited. Conclusion: Overall, the use of unstructured clinical text data in the development of prognostic prediction models has been found beneficial in addition to structured data in most studies. The EHR text data is a source of valuable information for prediction model development and should not be neglected. We suggest a future focus on explainability and external validation of the developed models, promoting robust and trustworthy prediction models in clinical practice.

Purpose In Eritrea, a 6-month isoniazid preventive therapy (IPT) was introduced in Eritrea in 2014 to prevent/reduce risk of incident tuberculosis in people living with HIV (PLHIV). The global and local uptake of IPT in newly enrolled PLHIV was reported to be low. Anecdotal reports showed that there was resistance from clinicians against its implementation. This study was therefore conducted to explore the factors that affect implementation of IPT in Eritrea from the perspectives of healthcare professionals. Materials and Methods An exploratory qualitative study that used a framework content analysis using inductive approach was employed. Data were collected from a sample of HIV care clinic prescribers from regional and national referral hospitals through in-depth interviews. Senior program officers were also interviewed as key informants. A conceptual framework model was developed using a root cause analysis. Results Overall, five themes and 13 sub-themes emerged from the in-depth interviews with healthcare professionals and key informants. Several multi-level causes/factors related to the healthcare system, HIV control program, healthcare professionals, patients and the product were identified as barriers to the implementation of IPT. Information gap on IPT and fear of isoniazid-induced liver injury were identified as the main reasons for the reluctance in administering IPT. It was observed that healthcare professionals had significant information gap that resulted in rumors and doubts on the benefits and risks of IPT, which ultimately caused reluctance on its implementation. Inadequate planning and operationalization during the introduction of IPT and inadequate laboratory setups were found to be the possible root causes for the aforementioned central problems. Conclusion The root causes/factors for the limited implementation of IPT in Eritrea were mainly related to the HIV control program and the healthcare system. Adequate planning, operationalization and capacitation of the existing laboratory setups are recommended for a successful implementation of IPT.