Analysis of user interactions with a digital health wallet for enabling care continuity in the context of an ongoing pandemic.

BACKGROUND: The onset of COVID-19 and related policy responses made it difficult to study interactive health informatics solutions in clinical study settings. Instrumented log and event data from interactive systems capture temporal details that can be used to generate insights about care continuity during ongoing pandemics. OBJECTIVE: To investigate user interactions with a digital health wallet (DHW) system for addressing care continuity challenges in chronic disease management in the context of an ongoing pandemic. MATERIALS AND METHODS: We analyzed user interaction log data generated by clinicians, nurses, and patients from the deployment of a DHW in a feasibility study conducted during the COVID-19 pandemic in Kenya. We used the Hamming distance from Information Theory to quantify deviations of usage patterns extracted from the events data from predetermined workflow sequences supported by the platform. RESULTS: Nurses interacted with all the user interface elements relevant to triage. Clinicians interacted with only 43% of elements relevant to consultation, while patients interacted with 67% of the relevant user interface elements. Nurses and clinicians deviated from the predetermined workflow sequences by 42% and 36%, respectively. Most deviations pertained to users going back to previous steps in their usage workflow. CONCLUSIONS: User interaction log analysis is a valuable alternative method for generating and quantifying user experiences in the context of ongoing pandemics. However, researchers should mitigate the potential disruptions of the actual use of the studied technologies as well as use multiple approaches to investigate user experiences of health technology during pandemics.

Quantifying the impact of COVID-19 on essential health services: a comparison of interrupted time series analysis using Prophet and Poisson regression models.

BACKGROUND: Coronavirus disease 2019 (COVID-19) altered healthcare utilization patterns. However, there is a dearth of literature comparing methods for quantifying the extent to which the pandemic disrupted healthcare service provision in sub-Saharan African countries. OBJECTIVE: To compare interrupted time series analysis using Prophet and Poisson regression models in evaluating the impact of COVID-19 on essential health services. METHODS: We used reported data from Uganda's Health Management Information System from February 2018 to December 2020. We compared Prophet and Poisson models in evaluating the impact of COVID-19 on new clinic visits, diabetes clinic visits, and in-hospital deliveries between March 2020 to December 2020 and across the Central, Eastern, Northern, and Western regions of Uganda. RESULTS: The models generated similar estimates of the impact of COVID-19 in 10 of the 12 outcome-region pairs evaluated. Both models estimated declines in new clinic visits in the Central, Northern, and Western regions, and an increase in the Eastern Region. Both models estimated declines in diabetes clinic visits in the Central and Western regions, with no significant changes in the Eastern and Northern regions. For in-hospital deliveries, the models estimated a decline in the Western Region, no changes in the Central Region, and had different estimates in the Eastern and Northern regions. CONCLUSIONS: The Prophet and Poisson models are useful in quantifying the impact of interruptions on essential health services during pandemics but may result in different measures of effect. Rigor and multimethod triangulation are necessary to study the true effect of pandemics on essential health services.

Subtyping Gastrointestinal Surgical Outcomes from Real World Data: A Comprehensive Analysis of UK Biobank.

Chronic gastrointestinal (GI) conditions, such as inflammatory bowel diseases (IBD), offer a promising opportunity to create classification systems that can enhance the accuracy of predicting the most effective therapies and prognosis for each patient. Here, we present a novel methodology to explore disease subtypes using our open-sourced BiomedSciAI toolkit. Applying methods available in this toolkit on the UK Biobank, including subpopulation-based feature selection and multi-dimensional subset scanning, we aimed to discover unique subgroups from GI surgery cohorts. Of a 12,073-patient cohort, a subgroup of 440 IBD patients was discovered with an increased risk of a subsequent GI surgery (OR: 2.21, 95% CI [1.81-2.69]). We iteratively demonstrate the discovery process using an additional cohort (with a narrower definition of GI surgery). Our results show that the iterative process can refine the subgroup discovery process and generate novel hypotheses to investigate determinants of treatment response.

Hierarchical Representation of Complex Intervention Sequences for Automated Subgroup Analysis in Critical Care Settings.

Our understanding of the impact of interventions in critical care is limited by the lack of techniques that represent and analyze complex intervention spaces applied across heterogeneous patient populations. Existing work has mainly focused on selecting a few interventions and representing them as binary variables, resulting in oversimplification of intervention representation. The goal of this study is to find effective representations of sequential interventions to support intervention effect analysis. To this end, we have developed Hi-RISE (Hierarchical Representation of Intervention Sequences), an approach that transforms and clusters sequential interventions into a latent space, with the resulting clusters used for heterogeneous treatment effect analysis. We apply this approach to the MIMIC III dataset and identified intervention clusters and corresponding subpopulations with peculiar odds of 28-day mortality. Our approach may lead to a better understanding of the subgroup-level effects of sequential interventions and improve targeted intervention planning in critical care settings.

Principled Subpopulation Analysis of the BetterBirth Study and the Impact of WHO's Safe Childbirth Checklist Intervention.

The World Health Organization (WHO) developed the Safe Childbirth Checklist as an intervention to improve care and outcomes in maternal and newborn health. The original study reported that the intervention did not significantly improve the outcomes. In this work, we employ a principled data-driven analysis to identify subpopulations with divergent characteristics: 1) vulnerable subgroups with the highest risk of neonatal deaths and 2) subgroups in the intervention arm that benefited from the Checklist intervention with significantly reduced risks of deaths and complications. Results demonstrate that low birth weight represented the most vulnerable group, whereas mother-baby dyads described by normal gestational age at birth, known parity, and unknown number of abortions was found to benefit from the Checklist intervention (OR : 0.70, 95%CI : 0.62-0.79, p < 0.001). Generally, the flexibility of our approach helps to answer subgroup-based queries in the broader global health domain, which also provides further insights to domain experts.

Data-level Linkage of Multiple Surveys for Improved Understanding of Global Health Challenges.

Data-driven approaches can provide more enhanced insights for domain experts in addressing critical global health challenges, such as newborn and child health, using surveys (e.g., Demographic Health Survey). Though there are multiple surveys on the topic, data-driven insight extraction and analysis are often applied on these surveys separately, with limited efforts to exploit them jointly, and hence results in poor prediction performance of critical events, such as neonatal death. Existing machine learning approaches to utilise multiple data sources are not directly applicable to surveys that are disjoint on collection time and locations. In this paper, we propose, to the best of our knowledge, the first detailed work that automatically links multiple surveys for the improved predictive performance of newborn and child mortality and achieves cross-study impact analysis of covariates.

Successes and Misses of Global Health Development: Detecting Temporal Concept Drift of Under-5 Mortality Prediction Models with Bias Scan.

Under-5 Mortality rates have been decreasing across Africa for the past two decades. Contributing factors include policy changes, technology, and health investments. This study identifies sub-populations that have experienced more-than-expected change in mortality rates (either increasing or decreasing) during this time period. We train under-5 mortality predictive models on Demographic and Health Survey (DHS) datasets from the early 2000s and apply those models to data collected in more recent versions of the survey. This provides an estimate of the risk current families would have faced in the past. We then apply techniques from anomalous pattern detection to identify sub-populations that have the most divergence between their predicted and observed mortality rates; higher and lower. These detected groups are examples of successes and possible misses of the health progress observed in Africa over the course of decades. Identifying these groups through data-driven discovery may lead to a better understanding of health policies in developing countries.

Detection of Anomalous Patterns Associated with the Impact of Medications on 30-Day Hospital Readmission Rates in Diabetes Care.

Improving quality of care in diabetes requires a good understanding of variations in diabetes outcomes and related interventions. However, little is known about the impact of diabetes interventions on outcome measures at the subpopulation-level. In this study, we developed methods that combine causal inference techniques with subset scanning techniques to study the heterogeneous effects of treatments on binary health outcomes. We analyzed a diabetes dataset consisting of 70,000 initial inpatient encounters to investigate the anomalous patterns associated with the impact of 4 anti-diabetic medication classes on 30-day readmission in diabetes. We discovered anomalous subpopulations where the likelihood of readmission was up to 1.8 times higher than that of the overall population suggesting subpopulation-level heterogeneity. Identifying such subpopulations may lead to a better understanding of the heterogeneous effects of treatments and improve targeted intervention planning.

AI-assisted tracking of worldwide non-pharmaceutical interventions for COVID-19.

The Coronavirus disease 2019 (COVID-19) global pandemic has transformed almost every facet of human society throughout the world. Against an emerging, highly transmissible disease, governments worldwide have implemented non-pharmaceutical interventions (NPIs) to slow the spread of the virus. Examples of such interventions include community actions, such as school closures or restrictions on mass gatherings, individual actions including mask wearing and self-quarantine, and environmental actions such as cleaning public facilities. We present the Worldwide Non-pharmaceutical Interventions Tracker for COVID-19 (WNTRAC), a comprehensive dataset consisting of over 6,000 NPIs implemented worldwide since the start of the pandemic. WNTRAC covers NPIs implemented across 261 countries and territories, and classifies NPIs into a taxonomy of 16 NPI types. NPIs are automatically extracted daily from Wikipedia articles using natural language processing techniques and then manually validated to ensure accuracy and veracity. We hope that the dataset will prove valuable for policymakers, public health leaders, and researchers in modeling and analysis efforts to control the spread of COVID-19.

Addressing Care Continuity and Quality Challenges in the Management of Hypertension: Case Study of the Private Health Care Sector in Kenya.

BACKGROUND: Hypertension is a major risk factor of cardiovascular disease and a leading cause of morbidity and mortality globally. In Kenya, the rise of hypertension strains an already stretched health care system that has traditionally focused on the management of infectious diseases. Health care provision in this country remains fragmented, and little is known about the role of health information technology in care coordination. Furthermore, there is a dearth of literature on the experiences, challenges, and solutions for improving the management of hypertension and other noncommunicable diseases in the Kenyan private health care sector. OBJECTIVE: The aim of this study is to assess stakeholders' perspectives on the challenges associated with the management of hypertension in the Kenyan private health care sector and to derive recommendations for the design and functionality of a digital health solution for addressing the care continuity and quality challenges in the management of hypertension. METHODS: We conducted a qualitative case study. We collected data using in-depth interviews with 18 care providers and 8 business leads, and direct observations at 18 private health care institutions in Nairobi, Kenya. We analyzed the data thematically to identify the key challenges and recommendations for technology-enabled solutions to support the management of hypertension in the Kenyan private health sector. We subsequently used the generated insights to derive and describe the design and range of functions of a digital health wallet platform for enabling care quality and continuity. RESULTS: The management of hypertension in the Kenyan private health care sector is characterized by challenges such as high cost of care, limited health care literacy, lack of self-management support, ineffective referral systems, inadequate care provider training, and inadequate regulation. Care providers lack the tools needed to understand their patients' care histories and effectively coordinate efforts to deliver high-quality hypertension care. The proposed digital health platform was designed to support hypertension care coordination and continuity through clinical workflow orchestration, decision support, and patient-mediated data sharing with privacy preservation, auditability, and trust enabled by blockchain technology. CONCLUSIONS: The Kenyan private health care sector faces key challenges that require significant policy, organizational, and infrastructural changes to ensure care quality and continuity in the management of hypertension. Digital health data interoperability solutions are needed to improve hypertension care coordination in the sector. Additional studies should investigate how patients can control the sharing of their data while ensuring that care providers have a holistic view of the patient during any encounter.

A Framework for Inferring Epidemiological Model Parameters using Bayesian Nonparametrics.

The use of epidemiological models for decision-making has been prominent during the COVID-19 pandemic. Our work presents the application of nonparametric Bayesian techniques for inferring epidemiological model parameters based on available data sets published during the pandemic, towards enabling predictions under uncertainty during emerging pandemics. We present a methodology and framework that allows epidemiological model drivers to be integrated as input into the model calibration process. We demonstrate our methodology using the stringency index and mobility data for COVID-19 on an SEIRD compartmental model for selected US states. Our results directly compare the use of Bayesian nonparametrics for model predictions based on best parameter estimates with results of inference of parameter values across the US states. The proposed methodology provides a framework for What-If analysis and sequential decision-making methods for disease intervention planning and is demonstrated for COVID-19, while also applicable to other infectious disease models.

Identifying Factors Associated with Neonatal Mortality in Sub-Saharan Africa using Machine Learning.

This study aimed at identifying the factors associated with neonatal mortality. We analyzed the Demographic and Health Survey (DHS) datasets from 10 Sub-Saharan countries. For each survey, we trained machine learning models to identify women who had experienced a neonatal death within the 5 years prior to the survey being administered. We then inspected the models by visualizing the features that were important for each model, and how, on average, changing the values of the features affected the risk of neonatal mortality. We confirmed the known positive correlation between birth frequency and neonatal mortality and identified an unexpected negative correlation between household size and neonatal mortality. We further established that mothers living in smaller households have a higher risk of neonatal mortality compared to mothers living in larger households; and that factors such as the age and gender of the head of the household may influence the association between household size and neonatal mortality.

Enhancing Guideline-Based Prescribing and Personalized Medication Scheduling.

Poor communication of medication administration instructions is a preventable cause of medication nonadherence. The Universal Medication Schedule (UMS) framework improves adherence by providing a simplified set of dose timing rules. However, this framework does not readily generalize to individuals with varying daily routines. We propose a point-of-care solution for enhancing guideline-based electronic prescribing and personalizing dose schedules. We describe a JSON-based approach to encode and execute standard treatment guidelines to support electronic prescribing as well as an algorithm for optimizing medication administration schedules based on a patient's daily routine. We evaluated the structure and accuracy of our JavaScript Object Notation (JSON) formalism focusing on Kenya's hypertension treatment guidelines. Our experiments compare the medication schedules generated by our algorithm with those generated by pharmacists. Our findings show that treatment guidelines can be efficiently represented and executed using the JSON formalism, and that different medication administration schedules can be generated automatically and optimized for patients' daily routines.

Medication-related problems among adult chronic kidney disease patients in a sub-Saharan tertiary hospital.

BACKGROUND: Chronic kidney disease (CKD) patients are prone to medication-related problems (MRPs). Few studies address the clinical relevance of MRPs among CKD patients in sub-Saharan Africa. OBJECTIVE: To investigate the frequency and predictors of MRPs among adult CKD patients treated at a tertiary care facility in an urban sub-Saharan setting. SETTING: Kenyatta National Hospital in Nairobi, Kenya. METHOD: A cross-sectional study involving 60 adult patients with CKD was carried out. Data were collected through structured interviews and patient chart reviews between April 2016 and June 2016. MRPs identified from the collected data were classified according to Hepler and Strand classification. The frequencies of the identified MRPs were computed and logistic regression used to investigate the associations between the MRPs and covariates in the data. MAIN OUTCOME MEASURES: frequencies and predictors of MRPs. RESULTS: 271 MRPs were identified. The commonest MRPs were drug interactions (21.8%), indication without drug (18.1%) and medication non-adherence (15.5%). Compared to patients with CKD stage 3, patients with CKD stage 4 were 5.9 times more likely to have an improper drug selection and 4.7 times more likely to experience overdosage. Other significant predictors of MRPs were the number of medications per prescription and the number of comorbidities per patient. CONCLUSION: This study found a high frequency of MRPs among patients with chronic kidney disease receiving care in urban sub-Saharan tertiary hospital settings. The predictors of MRPs among CKD patients in this setting are likely to be multifactorial and include the CKD stage, polypharmacy, and comorbidities.

Validation of the Behavior of a Knowledge Base Implementing Clinical Guidelines for Point-of-Care Antiretroviral Toxicity Monitoring.

This study investigated the automated detection of antiretroviral toxicities in structured electronic health records data. The evaluation compared responses generated by 5 clinical pharmacists and 1 prototype knowledge-based application for 15 randomly selected test cases. The main outcomes were inter-subject dissimilarity of responses quantified by the Jaccard distance, and the mean proportion of correct responses by each subject. The statistical differences in inter-subject Jaccard distances suggested that the prototype was inferior to clinical pharmacists in the detection of possible antiretroviral toxicity associations from structured data. The reason for dissimilarities was attributable to inadequate domain coverage by the prototype. The differences in the mean proportion of correct responses between the clinical pharmacists and the prototype were statistically indistinguishable. Overall, this study suggests that knowledge-based applications have the potential to support automated detection of antiretroviral toxicities from structured patient records. Furthermore, the study demonstrates a systematic approach for validating such applications quantitatively.

The Revolving Fund Pharmacy Model: backing up the Ministry of Health supply chain in western Kenya.

OBJECTIVES: A pressing challenge in low and middle-income countries (LMIC) is inadequate access to essential medicines, especially for chronic diseases. The Revolving Fund Pharmacy (RFP) model is an initiative to provide high-quality medications consistently to patients, using revenues generated from the sale of medications to sustainably resupply medications. This article describes the utilization of RFPs developed by the Academic Model Providing Access to Healthcare (AMPATH) with the aim of stimulating the implementation of similar models elsewhere to ensure sustainable access to quality and affordable medications in similar LMIC settings. METHODS: The service evaluation of three pilot RFPs started between April 2011 and January 2012 in select government facilities is described. The evaluation assessed cross-sectional availability of essential medicines before and after implementation of the RFPs, number of patient encounters and the impact of community awareness activities. FINDINGS: Availability of essential medicines in the three pilot RFPs increased from 40%, 36% and <10% to 90%, 94% and 91% respectively. After the first year of operation, the pilot RFPs had a total of 33 714 patient encounters. As of February 2014, almost 3 years after starting up the first RFP, the RFPs had a total of 115 991 patient encounters. In the Eldoret RFP, community awareness activities led to a 51% increase in sales. CONCLUSIONS: With proper oversight and stakeholder involvement, this model is a potential solution to improve availability of essential medicines in LMICs. These pilots exemplify the feasibility of implementing and scaling up this model in other locations.

Using Natural Language Processing and Network Analysis to Develop a Conceptual Framework for Medication Therapy Management Research.

This paper describes a theory derivation process used to develop a conceptual framework for medication therapy management (MTM) research. The MTM service model and chronic care model were selected as parent theories. Review article abstracts targeting medication therapy management in chronic disease care were retrieved from Ovid Medline (2000-2016). Unique concepts in each abstract were extracted using MetaMap and their pairwise cooccurrence determined. The information was used to construct a network graph of concept co-occurrence that was analyzed to identify content for the new conceptual model. 142 abstracts were analyzed. Medication adherence is the most studied drug therapy problem and co-occurred with concepts related to patient-centered interventions targeting self-management. The enhanced model consists of 65 concepts clustered into 14 constructs. The framework requires additional refinement and evaluation to determine its relevance and applicability across a broad audience including underserved settings.

Modelling the Medication Management System for Resource Limited Settings: A Formal Representation of the Prescribing and Dispensing Phases.

We propose a conceptual data model for relational databases targeting the prescribing and dispensing phases of the medication management system. The model was developed using recommendations from existing standards and guidelines, with necessary modifications made to suit adoption in resource-limited settings. We present the model as an entity-relationship diagram with 10 entities, 12 relationships and 48 attributes. It is our hope that this work will help mitigate barriers in the implementation of electronic prescribing and dispensing standards in the developing world.

Towards a Clinical Decision Support System for Drug Allergy Management: Are Existing Drug Reference Terminologies Sufficient for Identifying Substitutes and Cross-Reactants?

Drug allergy cross-reactivity checking is an important component of electronic health record systems. Currently, a single, open-source medication dictionary that can provide this function does not exist. In this study, we assessed the feasibility of using RxNorm and NDF-RT (National Drug File--Reference Terminology) for allergy management decision support. We evaluated the performance of using the Pharmacological Class, Mechanism of Action and Chemical Structure NDF-RT classifications in discriminating between safe and cross-reactive alternatives to a sample of common drug allergens. The positive predictive values for the three approaches were 96.3%, 99.3% and 96.2% respectively. The negative predictive values were 94.7%, 56.8% and 92.6%. Our findings suggest that in the absence of an established medication allergy classification system, using the Pharmacologic Class and Chemical Structure classifications in NDF-RT may still be effective for discriminating between safe and cross-reactive alternatives to potential allergens.