A treatment recommender clinical decision support system for personalized medicine: method development and proof-of-concept for drug resistant tuberculosis.

BACKGROUND: Personalized medicine tailors care based on the patient's or pathogen's genotypic and phenotypic characteristics. An automated Clinical Decision Support System (CDSS) could help translate the genotypic and phenotypic characteristics into optimal treatment and thus facilitate implementation of individualized treatment by less experienced physicians. METHODS: We developed a hybrid knowledge- and data-driven treatment recommender CDSS. Stakeholders and experts first define the knowledge base by identifying and quantifying drug and regimen features for the prototype model input. In an iterative manner, feedback from experts is harvested to generate model training datasets, machine learning methods are applied to identify complex relations and patterns in the data, and model performance is assessed by estimating the precision at one, mean reciprocal rank and mean average precision. Once the model performance no longer iteratively increases, a validation dataset is used to assess model overfitting. RESULTS: We applied the novel methodology to develop a treatment recommender CDSS for individualized treatment of drug resistant tuberculosis as a proof of concept. Using input from stakeholders and three rounds of expert feedback on a dataset of 355 patients with 129 unique drug resistance profiles, the model had a 95% precision at 1 indicating that the highest ranked treatment regimen was considered appropriate by the experts in 95% of cases. Use of a validation data set however suggested substantial model overfitting, with a reduction in precision at 1 to 78%. CONCLUSION: Our novel and flexible hybrid knowledge- and data-driven treatment recommender CDSS is a first step towards the automation of individualized treatment for personalized medicine. Further research should assess its value in fields other than drug resistant tuberculosis, develop solid statistical approaches to assess model performance, and evaluate their accuracy in real-life clinical settings.

A machine-learning based model for automated recommendation of individualized treatment of rifampicin-resistant tuberculosis.

Background: Rifampicin resistant tuberculosis remains a global health problem with almost half a million new cases annually. In high-income countries patients empirically start a standardized treatment regimen, followed by an individualized regimen guided by drug susceptibility test (DST) results. In most settings, DST information is not available or is limited to isoniazid and fluoroquinolones. Whole genome sequencing could more accurately guide individualized treatment as the full drug resistance profile is obtained with a single test. Whole genome sequencing has not reached its full potential for patient care, in part due to the complexity of translating a resistance profile into the most effective individualized regimen. Methods: We developed a treatment recommender clinical decision support system (CDSS) and an accompanying web application for user-friendly recommendation of the optimal individualized treatment regimen to a clinician. Results: Following expert stakeholder meetings and literature review, nine drug features and 14 treatment regimen features were identified and quantified. Using machine learning, a model was developed to predict the optimal treatment regimen based on a training set of 3895 treatment regimen-expert feedback pairs. The acceptability of the treatment recommender CDSS was assessed as part of a clinical trial and in a routine care setting. Within the clinical trial setting, all patients received the CDSS recommended treatment. In 8 of 20 cases, the initial recommendation was recomputed because of stock out, clinical contra-indication or toxicity. In routine care setting, physicians rejected the treatment recommendation in 7 out of 15 cases because it deviated from the national TB treatment guidelines. A survey indicated that the treatment recommender CDSS is easy to use and useful in clinical practice but requires digital infrastructure support and training. Conclusions: Our findings suggest that global implementation of the novel treatment recommender CDSS holds the potential to improve treatment outcomes of rifampicin resistant tuberculosis.

Convalescent plasma in the treatment of moderate to severe COVID-19 pneumonia: a randomized controlled trial (PROTECT-Patient Trial).

There is a need for effective therapy for COVID-19 pneumonia. Convalescent plasma has antiviral activity and early observational studies suggested benefit in reducing COVID-19 severity. We investigated the safety and efficacy of convalescent plasma in hospitalized patients with COVID-19 in a population with a high HIV prevalence and where few therapeutic options were available. We performed a double-blinded, multicenter, randomized controlled trial in one private and three public sector hospitals in South Africa. Adult participants with COVID-19 pneumonia requiring non-invasive oxygen were randomized 1:1 to receive a single transfusion of 200 mL of either convalescent plasma or 0.9% saline solution. The primary outcome measure was hospital discharge and/or improvement of ≥ 2 points on the World Health Organisation Blueprint Ordinal Scale for Clinical Improvement by day 28 of enrolment. The trial was stopped early for futility by the Data and Safety Monitoring Board. 103 participants, including 21 HIV positive individuals, were randomized at the time of premature trial termination: 52 in the convalescent plasma and 51 in the placebo group. The primary outcome occurred in 31 participants in the convalescent plasma group and and 32 participants in the placebo group (relative risk 1.03 (95% CI 0.77 to 1.38). Two grade 1 transfusion-related adverse events occurred. Participants who improved clinically received convalescent plasma with a higher median anti-SARS-CoV-2 neutralizing antibody titre compared with those who did not (298 versus 205 AU/mL). Our study contributes additional evidence for recommendations against the use of convalescent plasma for COVID-19 pneumonia. Safety and feasibility in this population supports future investigation for other indications.

The knowledge and perceptions regarding antibiotic stewardship of the interns rotating at the Bloemfontein Academic Complex.

BACKGROUND: Antibiotic resistance (ABR) is a global problem with the overuse of antibiotics accelerating this process. Antibiotic stewardship aims to optimise antibiotic treatment to enable cost-effective therapy and improve patients' outcome whilst limiting ABR. The study aimed to evaluate intern medical doctors' knowledge and perceptions about antibiotic stewardship and their perceptions regarding education on relevant topics. METHODS: This was a cross-sectional study on interns rotating at Bloemfontein Academic Complex. An anonymous, self-administered questionnaire was completed. The questionnaire recorded demographic information, perception and knowledge of antibiotic stewardship, and the quality of education as perceived by the interns. RESULTS: Of the 120 possible participants, 92 (76.7%) responded to all or part of the questionnaire. The median age of the respondents was 25 years, and 56.7% of the respondents were female. The mean score for the knowledge-based case scenarios was 5.4 out of 10. Only 4.4% participants could manage a drip site infection correctly, whilst 18.5% could treat Escherichia coli (E. coli) bacteraemia. The interns perceived that they have a lack of training and preparedness in certain areas of prescribing antibiotics. Though 77.2% of the interns had received education on starting antibiotic treatment, 29.3% claimed to be unsure when to start antibiotic therapy. Interns indicated that formal lectures (81.3%) and bedside tutorials (86.7%) have a high educational value. CONCLUSION: Intern medical doctors do not have sufficient knowledge to establish antibiotic stewardship but have a desire for improvement. The results identified specific areas where better antibiotic training is required.

Adalimumab treatment of life-threatening tuberculosis.

Most guidelines call for the discontinuation of treatment with tumor necrosis factor blockers in patients who develop tuberculosis. We report a case of life-threatening tuberculosis paradoxical reaction involving the lungs caused by withdrawal of the anti-tumor necrosis factor antibody adalimumab. Clinical improvement occurred only after resumption of adalimumab treatment.

South African medical students' perceptions and knowledge about antibiotic resistance and appropriate prescribing: Are we providing adequate training to future prescribers?

BACKGROUND: Education of medical students has been identified by the World Health Organization as an important aspect of antibiotic resistance (ABR) containment. Surveys from high-income countries consistently reveal that medical students recognise the importance of antibiotic prescribing knowledge, but feel inadequately prepared and require more education on how to make antibiotic choices. The attitudes and knowledge of South African (SA) medical students regarding ABR and antibiotic prescribing have never been evaluated. OBJECTIVE: To evaluate SA medical students' perceptions, attitudes and knowledge about antibiotic use and resistance, and the perceived quality of education relating to antibiotics and infection. METHODS: This was a cross-sectional survey of final-year students at three medical schools, using a 26-item self-administered questionnaire. The questionnaires recorded basic demographic information, perceptions about antibiotic use and ABR, sources, quality, and usefulness of current education about antibiotic use, and questions to evaluate knowledge. Hard-copy surveys were administered during whole-class lectures. RESULTS: A total of 289 of 567 (51%) students completed the survey. Ninety-two percent agreed that antibiotics are overused and 87% agreed that resistance is a significant problem in SA - higher proportions than those who thought that antibiotic overuse (63%) and resistance (61%) are problems in the hospitals where they had worked (p<0.001). Most reported that they would appreciate more education on appropriate use of antibiotics (95%). Only 33% felt confident to prescribe antibiotics, with similar proportions across institutions. Overall, prescribing confidence was associated with the use of antibiotic prescribing guidelines (p=0.003), familiarity with antibiotic stewardship (p=0.012), and more frequent contact with infectious diseases specialists (p<0.001). There was an overall mean correct score of 50% on the knowledge questionnaire, with significant differences between institutions. Students who used antibiotic prescribing guidelines and found their education more useful scored higher on knowledge questionnaires. CONCLUSION: There are low levels of confidence with regard to antibiotic prescribing among final-year medical students in SA, and most students would like more education in this area. Perceptions that ABR is less of a problem in their local setting may contribute to inappropriate prescribing behaviours. Differences exist between medical schools in knowledge about antibiotic use, with suboptimal scores across institutions. The introduction and use of antibiotic prescribing guidelines and greater contact with specialists in antibiotic prescribing may improve prescribing behaviours.