As the Electronic Health Record (EHR) data keeps growing in volume at an unprecedented rate, there is an increasing need for a more collaborative and scalable approach for designing and engineering clinical data pipelines. To address these two critical needs, we present a scalable analytics pipeline architecture, designed from the bottom-up to harness the power of FHIR (Fast Healthcare Interoperability Resources) for improving collaborative efforts in health data analytics and indicator reporting.

We sought to determine the relationship between continuity of care and adherence to clinic appointments among patients receiving HIV care in high vs. low clinician-to-patient (C:P) ratios facilities in western Kenya. This retrospective analysis included 12,751 patients receiving HIV care from the Academic Model Providing Access to Healthcare (AMPATH) program, between February 2016–2019. We used logistic regression analysis with generalized estimating equations to estimate the relationship between continuity of care (two consecutive visits with the same provider) and adherence to clinic appointments (within 7 days of a scheduled appointment) over time. Adjusting for covariates, patients in low C:P ratio facilities who had continuity of care, were more likely to be adherent to their appointments compared to those without continuity (adjusted odds ratio = 1.50; 95% confidence interval, 1.33–1.69). Continuity in HIV care may be a factor in clinical adherence among patients in low C:P ratio facilities and should therefore be promoted.

BACKGROUND:As the coronavirus pandemic progressed through the United States, Indianapolis Emergency Medical Services (IEMS) identified a gap between the health system capacity and the projected need to support an overwhelmed health care system. In addressing emergencies or special cases, each medical institution in a metropolitan area typically has a siloed process for capturing emergency patient records. These approaches vary in technical capabilities and may include use of an electronic medical record system (EMR) or a hybrid paper/EMR process. Given the projected volume of patients for the COVID-19 pandemic and the proposed multi-institutional team approach needed in case of significant provider illness, IEMS sought a simple, efficient, consolidated EMR solution to support planning for the potential capacity gap. IEMS approached Regenstrief Institute (RI), an established partner with experience in supporting OpenMRS, a global good EMR platform that had been deployed in multiple settings globally.OBJECTIVE:The purpose of this project was to determine if OpenMRS, a global good, could be used to quickly stand up a system that would meet the needs for health emergency data collection and reporting.DESIGN AND IMPLEMENTATION METHODS:The team used an "all hands on deck" approach, bringing together technical and subject matter experts, and a human-centered and iterative process to ensure the system met the key needs of IEMS. The OpenMRS Reference Application was adapted to the specific need and deployed as Docker containers to servers within the Indiana Health Information Exchange.PROJECT OUTCOMES AND LESSONS LEARNED:In less than two weeks, the Regenstrief team was able to install, configure and set up a working version of OpenMRS to support the desired electronic record requirements for the IEMS disaster field clinics. Using a human-centered approach, the RI team developed, tested, and released a user-friendly, installation-ready solution complete with an end user manual and a base support plan. IEMS and RI are sharing this approach to demonstrate how a global good can quickly generate a solution for COVID-19 and other disaster responses.CONCLUSIONS:Open source global goods can rapidly be adapted to meet local needs in an emergency. OpenMRS can be adapted to meet the needs of basic emergency medical services registration, triage, and basic data collection.

Abstract Background Human immunodeficiency virus (HIV) viral failure occurs when antiretroviral therapy fails to suppress and sustain a person’s viral load count below 1,000 copies of viral ribonucleic acid per milliliter. For those newly diagnosed with HIV and living in a setting where healthcare resources are limited, such as a low- and middle-income country, the World Health Organization recommends viral load monitoring six months after initiation of antiretroviral treatment and yearly thereafter. Deviations from this schedule are made in cases where viral failure occurs or at the discretion of the clinician. Failure to detect viral failure in a timely fashion can lead to delayed administration of essential interventions. Clinical prediction models based on information available in the patient medical record are increasingly being developed and deployed for decision support in clinical medicine and public health. This raises the possibility that prediction models can be used to detect potential for viral failure in advance of viral measurements, particularly when those measurements occur infrequently. Objective Our goal is to use electronic health record data from a large HIV care program in Kenya to characterize and compare the predictive accuracy of several statistical machine learning methods for predicting viral failure at the first and second measurements following initiation of antiretroviral therapy. Predictive accuracy is measured in terms of sensitivity, specificity and area under the receiver-operator characteristic curve. Methods We trained and cross-validated 10 statistical machine learning models and algorithms on data from over 10,000 patients in the Academic Model Providing Access to Healthcare care program in western Kenya. These included parametric, non-parametric, ensemble, and Bayesian methods. The input variables included 50 items from the clinical record, hand picked in consultation with clinician experts. Predictive accuracy measures were calculated using 10-fold cross validation. Results Viral load failure rate is about 20% in this patient cohort at both the first and second measurements. Ensemble techniques generally outperformed other methods. For predicting viral failure at the first follow up measure, specificity was over 90% for these methods, but sensitivity was typically in the 50–60% range. Predictive accuracy was greater for the second follow up measure, with sensitivities over 80%. Super Learner, gradient boosting and Bayesian additive regression trees consistently outperformed other methods. For a viral failure rate of 20%, the positive predictive value for the top-performing methods is between 75 and 85%, while the negative predictive value is over 95%. Conclusion Evidence from this study suggests that machine learning techniques have potential to identify patients at risk for viral failure prior to their scheduled measurements. Ultimately, prognostic virologic assessment can help guide the administration of earlier targeted intervention such as enhanced drug resistance monitoring, rigorous adherence counseling, or appropriate next-line therapy switching. External validation studies should be used to confirm the results found here.

Background Hypertension is a major risk factor for cardiovascular disease (CVD), yet treatment and control rates for hypertension are very low in low- and middle-income countries (LMICs). Lack of effective referral networks between different levels of the health system is one factor that threatens the ability to achieve adequate blood pressure control and prevent CVD-related morbidity. Health information technology and peer support are two strategies that have improved care coordination and clinical outcomes for other disease entities in other settings; however, their effectiveness and cost-effectiveness in strengthening referral networks to improve blood pressure control and reduce CVD risk in low-resource settings are unknown. Methods/design We will use the PRECEDE-PROCEED framework to conduct transdisciplinary implementation research, focused on strengthening referral networks for hypertension in western Kenya. We will conduct a baseline needs and contextual assessment using a mixed-methods approach, in order to inform a participatory, community-based design process to fully develop a contextually and culturally appropriate intervention model that combines health information technology and peer support. Subsequently, we will conduct a two-arm cluster randomized trial comparing 1) usual care for referrals vs 2) referral networks strengthened with our intervention. The primary outcome will be one-year change in systolic blood pressure. The key secondary clinical outcome will be CVD risk reduction, and the key secondary implementation outcomes will include referral process metrics such as referral appropriateness and completion rates. We will conduct a mediation analysis to evaluate the influence of changes in referral network characteristics on intervention outcomes, a moderation analysis to evaluate the influence of baseline referral network characteristics on the effectiveness of the intervention, as well as a process evaluation using the Saunders framework. Finally, we will analyze the incremental cost-effectiveness of the intervention relative to usual care, in terms of costs per unit decrease in systolic blood pressure, per percentage change in CVD risk score, and per disability-adjusted life year saved. Discussion This study will provide evidence for the implementation of innovative strategies for strengthening referral networks to improve hypertension control in LMICs. If effective, it has the potential to be a scalable model for health systems strengthening in other low-resource settings worldwide. Trial registration Clinicaltrials.gov, NCT03543787 . Registered on 29 June 2018.

BACKGROUND:Ineffective referral networks in low- and middle-income countries hinders access to evidence-based therapies by hypertensive patients, leading to high cardiovascular mortality and morbidity. The STRENGTHS (Strengthening Referral Networks for Management of Hypertension Across Health Systems) study evaluates strategies to improve referral processes utilizing the International Association of Public Participation framework to engage stakeholders.OBJECTIVES:This study sought to identify and engage key stakeholders involved in referral of patients in the Ministry of Health, western Kenya.METHODS:Key stakeholders involved in policy formulation, provision, or consumption of public health care service were mapped out and contacted by phone, letters, and emissaries to schedule meetings, explain research objectives, and obtain feedback.RESULTS:Key stakeholders identified were the Ministry of Health, the Academic Model Providing Access to Healthcare, health professionals, communities and their leadership, and patients. Engaging them resulted in permission to contact research in their areas of jurisdiction and enabled collaboration in updating care protocols with emphasis on timely and appropriate referrals.CONCLUSIONS:Early stakeholder identification and engagement using the International Association of Public Participation model eased explanation of research objectives, building consensus, and shaping the interventions to improve the referral process.

The Academic Model Providing Access to Healthcare (AMPATH) has been a model academic partnership in global health for nearly three decades, leveraging the power of a public-sector academic medical center and the tripartite academic mission – service, education, and research – to the challenges of delivering health care in a low-income setting. Drawing our mandate from the health needs of the population, we have scaled up service delivery for HIV care, and over the last decade, expanded our focus on non-communicable chronic diseases, health system strengthening, and population health more broadly. Success of such a transformative endeavor requires new partnerships, as well as a unification of vision and alignment of strategy among all partners involved. Leveraging the Power of Partnerships and Spreading the Vision for Population Health. We describe how AMPATH built on its collective experience as an academic partnership to support the public-sector health care system, with a major focus on scaling up HIV care in western Kenya, to a system poised to take responsibility for the health of an entire population. We highlight global trends and local contextual factors that led to the genesis of this new vision, and then describe the key tenets of AMPATH’s population health care delivery model: comprehensive, integrated, community-centered, and financially sustainable with a path to universal health coverage. Finally, we share how AMPATH partnered with strategic planning and change management experts from the private sector to use a novel approach called a ‘Learning Map®’ to collaboratively develop and share a vision of population health, and achieve strategic alignment with key stakeholders at all levels of the public-sector health system in western Kenya. We describe how AMPATH has leveraged the power of partnerships to move beyond the traditional disease-specific silos in global health to a model focused on health systems strengthening and population health. Furthermore, we highlight a novel, collaborative tool to communicate our vision and achieve strategic alignment among stakeholders at all levels of the health system. We hope this paper can serve as a roadmap for other global health partners to develop and share transformative visions for improving population health globally.

ow-to middle-income countries (LMICs) are disproportionately affected by the rise in prevalence of non-communicable diseases (NCDs).According to the World Health Organization, four groups of diseases-cardiovascular disease, cancer, respiratory disease and diabetes -comprise 82% of NCD deaths worldwide and three-quarters of the deaths related to NCD occur in LMIC [1].In Sub-Saharan Africa, the World Bank estimates that nearly 46% of all deaths will be attributable to NCDs by 2030, and 41% of all deaths for those aged 15-59 will be due to NCDs [2].

Even with the best health care available, patients with chronic illnesses typically spend no more than a few hours a year in a health care setting, while their outcomes are largely determined by their activities during the remaining 5,000 waking hours of the year. As a widely available, low-cost technology, mobile phones are a promising tool to use in engaging patients in behavior change and facilitating self-care between visits. We examined the impact of a six-month mobile health (mHealth) demonstration project among adults with diabetes who belonged to an academic medical center's employee health plan. In addition to pre-post improvements in glycemic control (p=0.01) and patients' satisfaction with overall care (p=0.04), we observed a net cost savings of 8.8 percent. Those early results suggest that mHealth programs can support health care organizations' pursuit of the triple aim of improving patients' experiences with care, improving population health, and reducing the per capita cost of health care

Background: There is increasing interest in finding novel approaches to reduce health disparities in readmissions for acute decompensated heart failure (ADHF). Text messaging is a promising platform for improving chronic disease self-management in low-income populations, yet is largely unexplored in ADHF.Objective: The purpose of this pre-post study was to assess the feasibility and acceptability of a text message-based (SMS: short message service) intervention in a largely African American population with ADHF and explore its effects on self-management.Methods: Hospitalized patients with ADHF were enrolled in an automated text message-based heart failure program for 30 days following discharge. Messages provided self-care reminders and patient education on diet, symptom recognition, and health care navigation. Demographic and cell phone usage data were collected on enrollment, and an exit survey was administered on completion. The Self-Care of Heart Failure Index (SCHFI) was administered preintervention and postintervention and compared using sample t tests (composite) and Wilcoxon rank sum tests (individual). Clinical data were collected through chart abstraction.Results: Of 51 patients approached for recruitment, 27 agreed to participate and 15 were enrolled (14 African-American, 1 White). Barriers to enrollment included not owning a personal cell phone (n=12), failing the Mini-Mental exam (n=3), needing a proxy (n=2), hard of hearing (n=1), and refusal (n=3). Another 3 participants left the study for health reasons and 3 others had technology issues. A total of 6 patients (5 African-American, 1 White) completed the postintervention surveys. The mean age was 50 years (range 23-69) and over half had Medicaid or were uninsured (60%, 9/15). The mean ejection fraction for those with systolic dysfunction was 22%, and at least two-thirds had a prior hospitalization in the past year. Participants strongly agreed that the program was easy to use (83%), reduced pills missed (66%), and decreased salt intake (66%). Maintenance (mean composite score 49 to 78, P=.003) and management (57 to 86, P=.002) improved at 4 weeks, whereas confidence did not change (57 to 75, P=.11). Of the 6 SCHFI items that showed a statistically significant improvement, 5 were specifically targeted by the texting intervention.Conclusions: Over half of ADHF patients in an urban, largely African American community were eligible and interested in participating in a text messaging program following discharge. Access to mobile phones was a significant barrier that should be addressed in future interventions. Among the participants who completed the study, we observed a high rate of satisfaction and preliminary evidence of improvements in heart failure self-management.

OBJECTIVES:Behavioral models for mobile phone-based diabetes interventions are lacking. This study explores the potential mechanisms by which a text message-based diabetes program affected self-management among African-Americans. METHODS:We conducted in-depth, individual interviews among 18 African-American patients with type 2 diabetes who completed a 4-week text message-based diabetes program. Each interview was audio-taped, transcribed verbatim, and imported into Atlas.ti software. Coding was done iteratively. Emergent themes were mapped onto existing behavioral constructs and then used to develop a novel behavioral model for mobile phone-based diabetes self-management programs. RESULTS:The effects of the text message-based program went beyond automated reminders. The constant, daily communications reduced denial of diabetes and reinforced the importance of self-management (Rosenstock Health Belief Model). Responding positively to questions about self-management increased mastery experience (Bandura Self-Efficacy). Most surprisingly, participants perceived the automated program as a "friend" and "support group" that monitored and supported their self-management behaviors (Barrera Social Support). CONCLUSIONS:A mobile phone-based diabetes program affected self-management through multiple behavioral constructs including health beliefs, self-efficacy, and social support. PRACTICE IMPLICATIONS:Disease management programs that utilize mobile technologies should be designed to leverage existing models of behavior change and can address barriers to self-management associated with health disparities.

Background. Self-management support and team-based care are essential elements of the Chronic Care Model but are often limited by staff availability and reimbursement. Mobile phones are a promising platform for improving chronic care but there are few examples of successful health system implementation. Program Development. An iterative process of program design was built upon a pilot study and engaged multiple institutional stakeholders. Patients identified having a "human face" to the pilot program as essential. Stakeholders recognized the need to integrate the program with primary and specialty care but voiced concerns about competing demands on clinician time. Program Description. Nurse administrators at a university-affiliated health plan use automated text messaging to provide personalized self-management support for member patients with diabetes and facilitate care coordination with the primary care team. For example, when a patient texts a request to meet with a dietitian, a nurse-administrator coordinates with the primary care team to provide a referral. Conclusion. Our innovative program enables the existing health system to support a de novo care management program by leveraging mobile technology. The program supports self-management and team-based care in a way that we believe engages patients yet meets the limited availability of providers and needs of health plan administrators.

PURPOSE:We pilot-tested a text message-based diabetes care program in an urban African-American population in which automated text messages were sent to participants with personalized medication, foot care, and appointment reminders and text messages were received from participants on adherence.METHODS:Eighteen patients participated in a 4-week pilot study. Baseline surveys collected data about demographics, historical cell phone usage, and adherence to core diabetes care measures. Exit interview surveys (using close-coded and open-ended questions) were administered to patients at the end of the program. A 1-month follow-up interview was conducted surveying patients on perceived self-efficacy. Wilcoxon signed-rank tests were used to compare baseline survey responses about self-management activities to those at the pilot's end and at 1-month follow-up.RESULTS:Eighteen urban African-American participants completed the pilot study. The average age was 55 and the average number of years with diabetes was 8. Half the participants were initially uncomfortable with text messaging. Example messages include "Did you take your diabetes medications today" and "How many times did you check your feet for wounds this week?" Participants averaged 220 text messages with the system, responded to messages 80% of the time, and on average responded within 6 minutes. Participants strongly agreed that text messaging was easy to perform and helped with diabetes self-care. Missed medication doses decreased from 1.6 per week to 0.6 (p = .003). Patient confidence in diabetes self-management was significantly increased during and 1 month after the pilot (p = .002, p = .008).CONCLUSIONS:Text messaging may be a feasible and useful approach to improve diabetes self-management in urban African Americans.

Local information access (LIA) programs tap into existing public health data flows, and present data in simple and useful ways to ground staff. LIAs hold great potential for improving rural health systems in developing regions; benefits include more evidence-based decision making and optimizations at a local scale, as well as improved service delivery and data quality. Our fledgling LIA program in rural Uganda currently provides clinicians with a small set of static data visualizations for discussion. To increase the program’s effectiveness, we want to automatically identify relevant data visualizations. We propose an adaptive tool that learns from local clinicians’ decision-making processes to predict and generate visualizations that show actionable anomalies.