Search

Background: The COVID-19 pandemic has highlighted the inability of health systems to leverage existing system infrastructure in order to rapidly develop and apply broad analytical tools that could inform state- and national-level policymaking, as well as patient care delivery in hospital settings. The COVID-19 pandemic has also led to highlighted systemic disparities in health outcomes and access to care based on race or ethnicity, gender, income-level, and urban-rural divide. Although the United States seems to be recovering from the COVID-19 pandemic owing to widespread vaccination efforts and increased public awareness, there is an urgent need to address the aforementioned challenges., Objective: This study aims to inform the feasibility of leveraging broad, statewide datasets for population health-driven decision-making by developing robust analytical models that predict COVID-19-related health care resource utilization across patients served by Indiana's statewide Health Information Exchange., Methods: We leveraged comprehensive datasets obtained from the Indiana Network for Patient Care to train decision forest-based models that can predict patient-level need of health care resource utilization. To assess these models for potential biases, we tested model performance against subpopulations stratified by age, race or ethnicity, gender, and residence (urban vs rural)., Results: For model development, we identified a cohort of 96,026 patients from across 957 zip codes in Indiana, United States. We trained the decision models that predicted health care resource utilization by using approximately 100 of the most impactful features from a total of 1172 features created. Each model and stratified subpopulation under test reported precision scores >70%, accuracy and area under the receiver operating curve scores >80%, and sensitivity scores approximately >90%. We noted statistically significant variations in model performance across stratified subpopulations identified by age, race or ethnicity, gender, and residence (urban vs rural)., Conclusions: This study presents the possibility of developing decision models capable of predicting patient-level health care resource utilization across a broad, statewide region with considerable predictive performance. However, our models present statistically significant variations in performance across stratified subpopulations of interest. Further efforts are necessary to identify root causes of these biases and to rectify them., (©Suranga N Kasturi, Jeremy Park, David Wild, Babar Khan, David A Haggstrom, Shaun Grannis. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 15.11.2021.)

Background: In the United States, more than 1.6 million new cases of cancer are estimated to be diagnosed each year. However, the burden of cancer among the US population is not shared equally, with racial and ethnic minorities and lower-income populations having a higher cancer burden compared with their counterparts. For example, African Americans have the highest mortality rates and shortest survival rates for most cancers compared with other racial or ethnic groups in the United States. A wide range of technologies (eg, internet-based [electronic health, eHealth] technologies, mobile [mobile health, mHealth] apps, and telemedicine) available to patients are designed to improve their access to care and empower them to participate actively in their care, providing a means to reduce health care disparities; however, little is known of their use among underserved populations., Objective: The aim of this study was to systematically review the current evidence on the use of cancer-specific patient-centered technologies among various underserved populations., Methods: Computer-based search was conducted in the following academic databases: (1) PubMed (cancer subset), (2) MEDLINE, (3) PsycINFO, and (4) CINAHL. We included studies that were peer-reviewed, published in the English language, and conducted in the United States. Each study was individually assessed for relevance, with any disagreements being reconciled by consensus. We used a 3-step inclusion process in which we examined study titles, abstracts, and full-text papers for assessment of inclusion criteria. We systematically extracted information from each paper meeting our inclusion criteria., Results: This review includes 71 papers that use patient-centered technologies that primarily targeted African Americans (n=31), rural populations (n=14), and Hispanics (n=12). A majority of studies used eHealth technologies (n=41) finding them to be leading sources of cancer-related health information and significantly improving outcomes such as screening among nonadherent individuals and increasing knowledge about cancer and cancer screening. Studies on mHealth found that participants reported overall favorable responses to receiving health information via short message service (SMS) text message; however, challenges were experienced with respect to lack of knowledge of how to text among some participants. More complex mobile technologies (eg, a tablet-based risk assessment tool) were also found favorable to use and acceptable among underserved populations; however, they also resulted in more significant barriers, for example, participants expressed concerns regarding security and unfamiliarity with the technology and preferred further instruction and assistance in its use., Conclusions: There is a growing body of literature exploring patient-centered technology and its influence on care of underserved populations. In this review, we find that these technologies seem to be effective, especially when tailored, in improving patient and care-related outcomes. Despite the potential of patient-centered technologies and the receptivity of underserved populations, challenges still exist with respect to their effective use and usability., (©Will L Tarver, David A Haggstrom. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 23.04.2019.)

Background: My HealtheVet (MHV) is the personal health record and patient portal developed by the United States Veterans Health Administration (VA). While millions of American veterans have registered for MHV, little is known about how a patient's health status may affect adoption and use of the personal health record., Objective: Our aim was to characterize the reach of the VA personal health record by clinical condition., Methods: This was a cross-sectional analysis of all veterans nationwide with at least one inpatient admission or two outpatient visits between April 2010 and March 2012. We compared adoption (registration, authentication, opt-in to use secure messaging) and use (prescription refill and secure messaging) of MHV in April 2012 across 18 specific clinical conditions prevalent in and of high priority to the VA. We calculated predicted probabilities of adoption by condition using multivariable logistic regression models adjusting for sociodemographics, comorbidities, and clustering of patients within facilities., Results: Among 6,012,875 veterans, 6.20% were women, 61.45% were Caucasian, and 26.31% resided in rural areas. The mean age was 63.3 years. Nationwide, 18.64% had registered for MHV, 11.06% refilled prescriptions via MHV, and 1.91% used secure messaging with their clinical providers. Results from the multivariable regression suggest that patients with HIV, hyperlipidemia, and spinal cord injury had the highest predicted probabilities of adoption, whereas those with schizophrenia/schizoaffective disorder, alcohol or drug abuse, and stroke had the lowest. Variation was observed across diagnoses in actual (unadjusted) adoption and use, with registration rates ranging from 29.19% of patients with traumatic brain injury to 14.18% of those with schizophrenia/schizoaffective disorder. Some of the variation in actual reach can be explained by facility-level differences in MHV adoption and by differences in patients' sociodemographic characteristics (eg, age, race, income) by diagnosis., Conclusions: In this phase of early adoption, opportunities are being missed for those with specific medical conditions that require intensive treatment and self-management, which could be greatly supported by functions of a tethered personal health record.