Utilization of Routine Health Information System and Associated Factors Among Health Professionals in Public Health Facilities in Dire Dawa, Eastern Ethiopia: a Cross-sectional Study

Background: Using reliable evidence from routine health information over time is an important aid to improve the health outcome, tackling disparities, enhancing efficiency, and encouraging innovation. In Ethiopia, the utilization of routine health data for improving the performance and quality of care was not well-studied in primary and secondary health facilities. This study assessed the level of routine health information utilization and associated factors among health professionals in public health facilities of Dire Dawa, eastern Ethiopia.Method: An institution-based cross-sectional study was conducted among 378 randomly selected health professionals from June 10 to July 20, 2020. A self-administered pretested structured questionnaire was used to collect data from participants. Data were entered using EpiData version 3.1 and analyzed using Stata version 16.0. Descriptive statistics were used to characterize the participants and binary logistic regression analysis was conducted to identify factors associated with the utilization of routine health information. Adjusted Odds Ratio (AOR) with 95% confidence interval was used to report association and significance was declared at P-value<0.05.Results: Good utilization of routine health information among health professionals was 57.7% (95% CI: 52.6%, 62.6%). Good organizational support (AOR=3.91, 95% CI: 2.01, 7.61), the low perceived complexity of the reporting formats (AOR=2.20, 95% CI: 1.23, 3.97), good self-efficacy (AOR=2.52, 95% CI: 1.25, 5.10), and good decision making autonomy (AOR=3.97, 95% CI: 2.12, 7.43) were important factors associated with good utilization of routine health information.Conclusion: Good utilization of routine health information among health professionals was low. Lack of self-confidence and empowerment of health professionals, the complexity of routine health information system format, and poor organizational support were significantly reducing the level of routine health information utilization. Therefore, improving the self-efficacy and decision-making capacity of health professionals through comprehensive training, empowerment and organizational support would be essential to increase the level of routine health information utilization.

Routine health information utilization and associated factors among health care workers in Ethiopia: A systematic review and meta-analysis

Background Utilization of routine health information plays a vital role for the effectiveness of routine and programed decisions. A proper utilization of routine health information helps to make decisions based on evidence. Considerable studies have been done on the utilization of routine health information among health workers in Ethiopia, but inconsistent findings were reported. Thus, this study was conducted to determine the pooled utilization of routine health information and to identify associated factors among health workers in Ethiopia. Methods Search of PubMed, HINARI, Global Health, Scopus, EMBASE, web of science, and Google Scholar was conducted to identify relevant studies from October 24, 2020 to November 18, 2020. The Newcastle-Ottawa scale tool was used to assess the quality of included studies. Two reviewers extracted the data independently using a standardized data extraction format and exported to STATA software version 11 for meta-analysis. Heterogeneity among studies was checked using Cochrane Q and I2 test statistics. The pooled estimate of utilization of routine health information was executed using a random effect model. Results After reviewing 22924 studies, 10 studies involving 4054 health workers were included for this review and meta-analysis. The pooled estimate of routine health information utilization among health workers in Ethiopia was 57.42% (95% CI: 41.48, 73.36). Supportive supervision (AOR = 2.25; 95% CI: 1.80, 2.82), regular feedback (AOR = 2.86; 95% CI: 1.60, 5.12), availability of standard guideline (AOR = 2.53; 95% CI: 1.80, 3.58), data management knowledge (AOR = 3.04; 95% CI: 1.75, 5.29) and training on health information (AOR = 3.45; 95% CI: 1.96, 6.07) were identified factors associated with utilization of routine health information. Conclusion This systematic review and meta-analysis found that more than two-fifth of health workers did not use their routine health information. This study suggests the need to conduct regular supportive supervision, provision of training and capacity building, mentoring on competence of routine health information tasks, and strengthening regular feedback at all health facilities. In addition, improving the accessibility and availability of standard set of indicators is important to scale-up information use.

APRILE: Exploring the Molecular Mechanisms of Drug Side Effects with Explainable Graph Neural Networks

With the majority of people 65 and over taking two or more medicines (polypharmacy), managing the side effects associated with polypharmacy is a global challenge. Explainable Artificial Intelligence (XAI) is necessary to reliably design safe polypharmacy. Here, we develop APRILE: a predictor-explainer framework based on graph neural networks to explore the molecular mechanisms underlying polypharmacy side effects by explaining predictions made by the predictors. For a side effect and its associated drug pair, or a set of side effects and their drug pairs, APRILE gives a set of proteins (drug targets or non-targets) and Gene Ontology (GO) items as the explanation. Using APRILE, we generate such explanations for 843,318 (learned) + 93,966 (novel) side effect--drug pair events, spanning 861 side effects (472 diseases, 485 symptoms and 9 mental disorders) and 20 disease categories. We show that our two new metrics, pharmacogenomic information utilization and protein-protein interaction information utilization, provide quantitative estimates of mechanism complexity. Explanations were significantly consistent with state of the art disease-gene associations for 232/239 (97%) side effects. Further, APRILE generated new insights into molecular mechanisms of four diverse categories of ADRs: infection, metabolic diseases, gastrointestinal diseases, and mental disorders, including paradoxical side effects. We demonstrate the viability of discovering polypharmacy side effect mechanisms by learning from an AI model trained on massive biomedical data. Consequently, it facilitates wider and more reliable use of AI in healthcare.

Improving data quality and information utilization at Metema Primary Hospital, Amhara national regional state, Northwest Ethiopia 2018: Capstone project

Quality data for evidence-based decision making become a growing concern globally. Available information needs to be disseminated on time and used for decision making. Therefore, an effective Health Management Information System is essential to make evidence-based decision. This study aimed to measure the change in data quality and information utilization before and after intervention. Facility-based pre-post interventional study design was conducted at Metema hospital from September/2016 to December30/2018. A total of 384 individual medical-records, HMIS registration-books and reports were reviewed. Training, supportive supervision and feedback were intervention packages. About 309 (80.5%) of charts were from outpatient department. Data recording completeness increased from 69.0% to 96.0%, data consistency increased from 84.0% to 99.5% and report timeliness enhanced from 66.0% to 100%. There was a statistically significant difference for data recording completeness between pre and post-intervention results with mean difference of −0.246 (−0.412, −0.081). Also, after the intervention, gap-filling feedback and supportive supervision were given to all departments. In addition, four quality improvement projects were developed at post-intervention phase. The level of data quality and use was improved after the intervention. So, designing and implementing intervention strategies based on the root causes will help to improve data quality and use.