Background: Patient-Generated Health Data (PGHD) in remote monitoring programs is a promising source of precise, personalized data, encouraged by expanding growth in the health technologies market. However, PGHD utilization in clinical settings is low. One of the critical challenges that impedes confident clinical use of PGHD is that these data are not managed according to any recognized approach for data quality assurance. Objective: This article aims to identify the PGHD management and quality challenges that such an approach must address, as these are expressed by key PGHD stakeholder groups. Materials and Methods: In-depth interviews were conducted with 20 experts who have experience in the use of PGHD in remote patient monitoring, including: healthcare providers, health information professionals within clinical settings, and commercial providers of remote monitoring solutions. Participants were asked to describe PGHD management processes in the remote monitoring programs in which they are involved, and to express their perspectives on PGHD quality challenges during the data management stages. Results: The remote monitoring programs in the study did not follow clear PGHD management or quality assurance approach. Participants were not fully aware of all the considerations of PGHD quality. Digital health literacy, wearable accuracy, difficulty in data interpretation, and lack of PGHD integration with electronic medical record systems were among the key challenges identified that impact PGHD quality. Conclusion: Co-development of PGHD quality guidelines with relevant stakeholders, including patients, is needed to ensure that quality remote monitoring data from wearables is available for use in more precise and personalized patient care.

https://minerva-access.unimelb.edu.au/bitstream/handle/11343/240481/Patient-generated health data management and quality challenges in remote patient monitoring.pdf

Patient-generated health data management and quality challenges in remote patient monitoring.

Smoking is the leading cause of preventable death in the United States Obtaining patients' smoking status is the first step in delivering smoking cessation counseling In this study, we assessed the quality of smoking status captured in an electronic health record from a large academic medical center We analyzed data from structured notes, finding that smoking status was documented in 98% of 64,451 hospital encounters in 2016 32% hospital encounters had discrepant documentation, and 545% of patients had implausible changes (eg, changes from "current smoker" to "never smoker") Overall, only 29% of patients were documented as active smokers, but 364% were documented as "unknown" or had discrepancies in their smoking status These results suggest that patients that smoke are not appropriately identified Centralized documentation with clinically actionable smoking status categories and implementation of patient-facing tools that allow patients to directly record their information could improve data quality of smoking status

Challenges with Collecting Smoking Status in Electronic Health Records.

Objective The growing prevalence of chronic conditions requiring changes in lifestyle and at-home self-management has increased interest in and need for supplementing clinic visits with data generated by patients outside the clinic. Patient-generated health data (PGHD) support the ability to diagnose and manage chronic conditions, to improve health outcomes, and have the potential to facilitate more "connected health" between patients and their care teams; however, health systems have been slow to adopt PGHD use in clinical care. Materials and methods We surveyed current and potential users of PGHD to catalog how PGHD is integrated into clinical care at an academic health center. The survey included questions about data type, method of collection, and clinical uses of PGHD. Current users were asked to provide detailed case studies of PGHD use in research and care delivery. Results Thirty-one respondents completed the survey. Seventeen individuals contributed detailed case studies of PGHD use across diverse areas of care, including behavioral health, metabolic and gastrointestinal conditions, musculoskeletal/progressive functional conditions, cognitive symptoms, and pain management. Sensor devices and mobile technologies were the most commonly reported platforms for collection. Clinicians and researchers involved in PGHD use cited the potential for PGHD to enhance care delivery and outcomes, but also indicated substantial barriers to more widespread PGHD adoption across healthcare systems. Conclusion The results of our survey illustrate how PGHD is used in targeted areas of one healthcare system and provide meaningful insights that can guide health systems in supporting the widespread use of PGHD in care delivery.

/pdf/use-of-patient-generated-health-data-across-healthcare-5fx4vr1fjr.pdf

Use of patient-generated health data across healthcare settings: implications for health systems.

Background Patient-generated health data (PGHD) collected digitally with mobile health (mHealth) technology has garnered recent excitement for its potential to improve precision management of chronic conditions such as atrial fibrillation (AF), a common cardiac arrhythmia. However, sustained engagement is a major barrier to collection of PGHD. Little is known about barriers to sustained engagement or strategies to intervene upon engagement through application design. Objective This article investigates individual patient differences in sustained engagement among individuals with a history of AF who are self-monitoring using mHealth technology. Methods This qualitative study involved patients, health care providers, and research coordinators previously involved in a randomized, controlled trial involving electrocardiogram (ECG) self-monitoring of AF. Patients were adults with a history of AF randomized to the intervention arm of this trial who self-monitored using ECG mHealth technology for 6 months. Semistructured interviews and focus groups were conducted separately with health care providers and research coordinators, engaged patients, and unengaged patients. A validated model of sustained engagement, an adapted unified theory of acceptance and use of technology (UTAUT), guided data collection, and analysis through directed content analysis. Results We interviewed 13 patients (7 engaged, 6 unengaged), 6 providers, and 2 research coordinators. In addition to finding differences between engaged and unengaged patients within each predictor in the adapted UTAUT model (perceived ease of use, perceived usefulness, facilitating conditions), four additional factors were identified as being related to sustained engagement in this population. These are: (1) internal motivation to manage health, (2) relationship with health care provider, (3) supportive environments, and (4) feedback and guidance. Conclusion Although it required some modification, the adapted UTAUT model was useful in understanding of the parameters of sustained engagement. The findings of this study provide initial requirement specifications for the design of applications that engage patients in this unique population of adults with AF.

/pdf/factors-influencing-sustained-engagement-with-ecg-self-ejji9kk603.pdf

Factors Influencing Sustained Engagement with ECG Self-Monitoring: Perspectives from Patients and Health Care Providers.

Objectives: In this study, we aimed: 1) to conceptualize the theoretical challenges facing health information systems (HIS) to represent patients’ decisions about health and medical treatments in everyday life; 2) to suggest approaches for modeling these processes. Methods: The conceptualization of the theoretical and methodological challenges was discussed in 2015 during a series of interdisciplinary meetings attended by health informatics staff, epidemiologists and health professionals working in quality management and primary and secondary prevention of chronic diseases of the Hospital Italiano de Buenos Aires, together with sociologists, anthropologists and e-health stakeholders. Results: HIS are facing the need and challenge to represent social human processes based on constructive and complexity theories, which are the current frameworks of human sciences for understanding human learning and socio-cultural changes. Computer systems based on these theories can model processes of social construction of concrete and subjective entities and the interrelationships between them. These theories could be implemented, among other ways, through the mapping of health assets, analysis of social impact through community trials and modeling of complexity with system simulation tools. Conclusions: This analysis suggested the need to complement the traditional linear causal explanations of disease onset (and treatments) that are the bases for models of analysis of HIS with constructive and complexity frameworks. Both may enlighten the complex interrelationships among patients, health services and the health system. The aim of this strategy is to clarify people’s decision making processes to improve the efficiency, quality and equity of the health services and the health system.

/pdf/representation-of-people-s-decisions-in-health-information-5dy377fpbo.pdf

Representation of People's Decisions in Health Information Systems.* A Complementary Approach for Understanding Health Care Systems and Population Health.

Patient-facing technologies are increasingly utilized for direct patient data entry for potential incorporation into the electronic health record. We analyzed patient-entered data during implementation of a patient-facing data entry technology using an online patient portal and clinic-based tablet computers at a University-based tertiary medical center clinic, including entries for past medical history, past surgical history, and social history. Entries were assessed for granularity, clinical accuracy, and the addition of novel information into the record. We found that over half of patient-generated diagnoses were duplicates of lesser or equal granularity compared to previous provider-entered diagnoses. Approximately one fifth of patient-generated diagnoses were found to meet the criteria for new, meaningful additions to the medical record. Our findings demonstrate that while patient-generated data provides important additional information, it may also present challenges including generating inaccurate or less granular information.

Characterizing Patient-Generated Clinical Data and Associated Implications for Electronic Health Records.

Objective To prospectively evaluate a logistic regression-based machine learning (ML) prognostic algorithm implemented in real-time as a clinical decision support (CDS) system for symptomatic persons under investigation (PUI) for Coronavirus disease 2019 (COVID-19) in the emergency department (ED). Methods We developed in a 12-hospital system a model using training and validation followed by a real-time assessment. The LASSO guided feature selection included demographics, comorbidities, home medications, vital signs. We constructed a logistic regression-based ML algorithm to predict “severe” COVID-19, defined as patients requiring intensive care unit (ICU) admission, invasive mechanical ventilation, or died in or out-of-hospital. Training data included 1,469 adult patients who tested positive for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) within 14 days of acute care. We performed: 1) temporal validation in 414 SARS-CoV-2 positive patients, 2) validation in a PUI set of 13,271 patients with symptomatic SARS-CoV-2 test during an acute care visit, and 3) real-time validation in 2,174 ED patients with PUI test or positive SARS-CoV-2 result. Subgroup analysis was conducted across race and gender to ensure equity in performance. Results The algorithm performed well on pre-implementation validations for predicting COVID-19 severity: 1) the temporal validation had an area under the receiver operating characteristic (AUROC) of 0.87 (95%-CI: 0.83, 0.91); 2) validation in the PUI population had an AUROC of 0.82 (95%-CI: 0.81, 0.83). The ED CDS system performed well in real-time with an AUROC of 0.85 (95%-CI, 0.83, 0.87). Zero patients in the lowest quintile developed “severe” COVID-19. Patients in the highest quintile developed “severe” COVID-19 in 33.2% of cases. The models performed without significant differences between genders and among race/ethnicities (all p-values > 0.05). Conclusion A logistic regression model-based ML-enabled CDS can be developed, validated, and implemented with high performance across multiple hospitals while being equitable and maintaining performance in real-time validation.

/pdf/a-12-hospital-prospective-evaluation-of-a-clinical-decision-1foy0vlo.pdf

A 12-hospital prospective evaluation of a clinical decision support prognostic algorithm based on logistic regression as a form of machine learning to facilitate decision making for patients with suspected COVID-19

Daren Scheppmann

Papers

Characterizing Patient-Generated Clinical Data and Associated Implications for Electronic Health Records.

A 12-hospital prospective evaluation of a clinical decision support prognostic algorithm based on logistic regression as a form of machine learning to facilitate decision making for patients with suspected COVID-19