Feature selection and prediction of treatment failure in tuberculosis.

BACKGROUND: Tuberculosis is a major cause of morbidity and mortality in the developing world. Drug resistance, which is predicted to rise in many countries worldwide, threatens tuberculosis treatment and control. OBJECTIVE: To identify features associated with treatment failure and to predict which patients are at highest risk of treatment failure. METHODS: On a multi-country dataset managed by the National Institute of Allergy and Infectious Diseases we applied various machine learning techniques to identify factors statistically associated with treatment failure and to predict treatment failure based on baseline demographic and clinical characteristics alone. RESULTS: The complete-case analysis database consisted of 587 patients (68% males) with a median (p25-p75) age of 40 (30-51) years. Treatment failure occurred in approximately one fourth of the patients. The features most associated with treatment failure were patterns of drug sensitivity, imaging findings, findings in the microscopy Ziehl-Nielsen stain, education status, and employment status. The most predictive model was forward stepwise selection (AUC: 0.74), although most models performed at or above AUC 0.7. A sensitivity analysis using the 643 original patients filling the missing values with multiple imputation showed similar predictive features and generally increased predictive performance. CONCLUSION: Machine learning can help to identify patients at higher risk of treatment failure. Closer monitoring of these patients may decrease treatment failure rates and prevent emergence of antibiotic resistance. The use of inexpensive basic demographic and clinical features makes this approach attractive in low and middle-income countries.

Hacking Hackathons: Preparing the next generation for the multidisciplinary world of healthcare technology.

OBJECTIVE: Machine learning in healthcare, and innovative healthcare technology in general, require complex interactions within multidisciplinary teams. Healthcare hackathons are being increasingly used as a model for cross-disciplinary collaboration and learning. The aim of this study is to explore high school student learning experiences during a healthcare hackathon. By optimizing their learning experiences, we hope to prepare a future workforce that can bridge technical and health fields and work seamlessly across disciplines. METHODS: A qualitative exploratory study utilizing focus group interviews was conducted. Eight high school students from the hackathon were invited to participate in this study through convenience sampling Participating students (n = 8) were allocated into three focus groups. Semi structured interviews were completed, and transcripts evaluated using inductive thematic analysis. FINDINGS: Through the structured analysis of focus group transcripts three major themes emerged from the data: (1) Collaboration, (2) Transferable knowledge and skills, and (3) Expectations about hackathons. These themes highlight strengths and potential barriers when bringing this multidisciplinary approach to high school students and the healthcare community. CONCLUSION: This study found that students were empowered by the interdisciplinary experience during a hackathon and felt that the knowledge and skills gained could be applied in real world settings. However, addressing student expectations of hackathons prior to the event is an area for improvement. These findings have implications for future hackathons and can spur further research into using the hackathon model as an educational experience for learners of all ages.

Guidelines and mHealth to Improve Quality of Hypertension and Type 2 Diabetes Care for Vulnerable Populations in Lebanon: Longitudinal Cohort Study.

BACKGROUND: Given the protracted nature of the crisis in Syria, the large noncommunicable disease (NCD) caseload of Syrian refugees and host Lebanese, and the high costs of providing NCD care, the implications for Lebanon's health system are vast. OBJECTIVE: The aim of this study was to evaluate the effectiveness of treatment guidelines and a mobile health (mHealth) app on quality of care and health outcomes in primary care settings in Lebanon. METHODS: A longitudinal cohort study was implemented from January 2015 to August 2016 to evaluate the effectiveness of treatment guidelines and an mHealth app on quality of care and health outcomes for Syrian and Lebanese patients in Lebanese primary health care (PHC) facilities. RESULTS: Compared with baseline record extraction, recording of blood pressure (BP) readings (-11.4%, P<.001) and blood sugar measurements (-6.9%, P=.03) significantly decreased following the implementation of treatment guidelines. Recording of BP readings also decreased after the mHealth phase as compared with baseline (-8.4%, P=.001); however, recording of body mass index (BMI) reporting increased at the end of the mHealth phase from baseline (8.1%, P<.001) and the guidelines phase (7.7%, P<.001). There were a great proportion of patients for whom blood sugar, BP, weight, height, and BMI were recorded using the tablet compared with in paper records; however, only differences in BMI were statistically significant (31.6% higher in app data as compared with paper records; P<.001). Data extracted from the mHealth app showed that a higher proportion of providers offered lifestyle counseling compared with the counseling reported in patients' paper records (health diet counseling; 77.3% in app data vs 8.8% in paper records, P<.001 and physical activity counseling and 59.7% in app vs 7.1% in paper records, P<.001). There were statistically significant increases in all four measures of patient-provider interaction across study phases. Provider inquiry of medical history increased by 16.6% from baseline following guideline implementation and by 28.2% from baseline to mHealth implementation (P<.001). From baseline, patient report of provider inquiry regarding medication complications increased in the guidelines and mHealth phases by 12.9% and 59.6%, respectively, (P<.001). The proportion of patients reporting that providers asked other questions relevant to their illness increased from baseline through guidelines implementation by 27.8% and to mHealth implementation by 66.3% (P<.001). Follow-up scheduling increased from baseline to the guidelines phase by 20.6% and the mHealth phase by 39.8% (P<.001). CONCLUSIONS: Results from this study of an mHealth app in 10 PHC facilities in Lebanon indicate that the app has potential to improve adherence to guidelines and quality of care. Further studies are necessary to determine the effects of patient-controlled health record apps on provider adherence to treatment guidelines, as well as patients' long-term medication and treatment adherence and disease control.

A "datathon" model to support cross-disciplinary collaboration.

In recent years, there has been a growing focus on the unreliability of published biomedical and clinical research. To introduce effective new scientific contributors to the culture of health care, we propose a "datathon" or "hackathon" model in which participants with disparate, but potentially synergistic and complementary, knowledge and skills effectively combine to address questions faced by clinicians. The continuous peer review intrinsically provided by follow-up datathons, which take up prior uncompleted projects, might produce more reliable research, either by providing a different perspective on the study design and methodology or by replication of prior analyses.