Abstract
Early experience of using national health service databases to promote safe herbs use among laypersons in developing country
Krit Pongpirul, MD, MPH, PhD.1, Thitima Paluekpet2, Tingta Massakarn2, Panasun Sunanta3, Verayuth Lertnattee4, Somboon Boonkittichaipan, MD.2 and Apichai Limanond, MD, MPH.5
(1)Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand, (2)Krabi Provincial Health Office, Krabi, Thailand, (3)True Digital Group Co.,Ltd., Bangkok, Thailand, (4)Silpakorn University, Bangkok, Thailand, (5)Kalasin Provincial Health Offfice, Kalasin, Thailand
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Background Herbal medicine has gained popularity among laypersons during the past decade. Despite its clinical benefits, herbs might have an adverse effect on some individuals who are taking conventional drugs and/or have impaired renal or liver functions.
Methods Three major databases were cleaned and merged: 1) Health Data Center (HDC), Ministry of Public Health (diagnosis, procedure, drugs, and laboratory data of 300+ million out-patient visits and 30+ million admissions annually from 2013-2018); 2) National List of Essential Medicine (NLEM) 2018; and 3) Herb-Drug Interactions Database, Mahidol University. A rule-based algorithm and a prototype web-based application were developed in order to estimate a risk score based on potential drug interactions as well as kidney and liver functions. This study was approved by the ethics committee, Krabi provincial health office.
Results With only the citizen identification number entered by an authorized healthcare professional, the application retrieved relevant data and estimated a risk score for the individual. The score is categorized and presented as red, yellow, and green colors to represent high, moderate, and low risks. The application also shows the demographics, clinical characteristics, drug use history, as well as a list of herbs to be avoided. The application was well received by the laypersons for its fast data retrieval and score estimation as well as a simple and clean user interface; however, some raised concern about data privacy and confidentiality. The prototype is being evaluated for its ability to improve safe herbs use.
Conclusion Big data with a simple rule-based algorithm could be useful for promoting safe herbs use among laypersons in a developing country.
Communication and informatics Implementation of health education strategies, interventions and programs Other professions or practice related to public health Planning of health education strategies, interventions, and programs
Abstract
Developing and evaluating public health data literacy competencies: A data analytics fellowship for health librarians
Peace Ossom Williamson, MLS, MS, AHIP1, Tiffany Kindratt, PhD MPH2, Erin Carlson, DrPH, MPH2, Kyrah Brown, PhD2 and Hammad Khan1
(1)The University of Texas at Arlington, Arlington, TX, (2)University of Texas at Arlington, Arlington, TX
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Background:There is an urgent need to train health information professionals (HIPs) who are increasingly moving into data-related positions that support and collaborate in research; however, few informatics and data literacy trainings are available for HIPs. To address this need, we developed the Data Analytics Research Training (DART) Fellowship to allow HIPs working in areas with limited resources to acquire data literacy skills. The DART Fellowship is a free, applied data science program delivered in two in-person sessions and five weeks of online modules. Participants were matched with research mentors and complete a final data project. Methods: We developed curriculum to align with data literacy competencies and principles from Quality Matters, a certification focused on quality online and hybrid education development and experiential learning. We recruited HIPs from diverse ethnic and work backgrounds who were working with underserved communities. Results: Most DART participants were female (88%) with the largest proportions from the National Network of Libraries of Medicine South Central region (37.5%) and Middle Atlantic (25%) region. Of the 24 participants, 18 worked in medical and academic libraries, three worked in health departments, and two worked in public libraries. We will discuss lessons learned from developing the program and evaluation results (using the Kirkpatrick model). Key lessons include the importance of building a culture of support and learning, using project management tools, and applying strategies to effectively streamline information. Conclusions: Intentionally designed, competency-based data science trainings can have a significant impact by building the data literacy capacity of HIPs in underserved communities.
Biostatistics, economics Diversity and culture Epidemiology Other professions or practice related to public health Public health or related education Public health or related research
Abstract
Using administrative datasets to generate local area population analyses of adverse childhood events
Leslie Ray, MPH, MPPA, MA1, Christopher O'Malley, MPH2, Sanaa Abedin, MPH1, Isabel Corcos, PhD, MPH1, Maria Peńa, MPH1 and Ryan Smith, MPH1
(1)County of San Diego Health and Human Services Agency, San Diego, CA, (2)County of San Diego, San Diego, CA
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Background: Negative health and well being consequences among adults have been associated with adverse events that occurred during childhood. These consequences have included higher rates of chronic disease, substance use disorder, mental health disease as well as poverty, low education and incarceration.
Objective/Purpose: The purpose of this project was to generate population based ACES data at the small area level to provide the local public health agency with baseline and ongoing measurements that can be linked at aggregate level to corresponding health, demographic and economic data.
Methods: This project used data from the ACES component of the Behavioral Risk Factor Survey. Pooling data for multiple years produced statistically significant results for regions within the county. These data were then matched geographically with demographic, economic, behavioral and disease rates.
Results: At the same geographic level the percent of adults who experienced ACES was integrated with the percentage who also experienced demographic, economic and health consequences. For example:
Percent of adults who as children:
20.3% resided with someone who was a problem drinker or alcoholic
12.9% resided with someone who was depressed, mentally ill or suicidal
9.8% resided with someone who used street drugs or abused prescription drugs
Percent of adults who:
13.3% lived below poverty level
388.2/100,000 were discharged with any mention of alcohol use disorder
540.8/100,000 were discharged with any mention of ADRD
The result of this project was the development of a geographically based database system that incorporated ACES, demographic, economic, behavioral and health data.
Discussion/Conclusions: Establishing small area estimates of ACES and pairing those estimates with other health and well being measures in a single data system allows epidemiologists to generate correlations between disease and social determinants. This provides a robust system for considering factors that may contribute to negative long-term health and well being consequences.
Administer health education strategies, interventions and programs Biostatistics, economics Implementation of health education strategies, interventions and programs Planning of health education strategies, interventions, and programs Social and behavioral sciences
Abstract
Supporting communities by enhancing data analytics and integration capacity of health-related organizations
Joshua Colasurdo, BS, Anne King, MBA, Christie Pizzimenti, PhD and David Dorr, MD, MS
Oregon Health & Science University, Portland, OR
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Suboptimal health outcomes are common in vulnerable populations, including those defined by race, socioeconomic status, geography, age, ethnicity, and gender identity. Addressing these disparities solely through medical care – especially in underserved rural settings – often misses the underlying factors that drive such disparities. Addressing such disparities requires data that comes from multiple sources, including social determinants of health, chronic illness diagnoses, and community services. We describe a framework for advancing the standardization and processing of data for advanced analytics within community and health-related organizations by integrating individual and area biopsychosocial data from multiple sources at the county and community level. Our consensus-based approach was built with 12 different organizations, including payers, public health departments, health systems, and individual clinics. It focuses on increasing data integration capabilities through standardization and harmonization of the types of data, integrating and augmenting data where possible, managing data granularity, and sharing data quality assessments and data structures to organizations using a federated data process. For instance, we assess missingness and inaccuracy and suggest areal sources for supplements. To enhance analytics, we leverage common data visualization tools implemented at each organization with compiled analytic techniques through containerization. This results in secure, robust visualizations that can combine diverse data such as social determinants with claims and clinical data to identify populations with multiple biopsychosocial risks; then, plan tailored interventions based on needs. For instance, public health and health systems teams can address patients with both diabetes and food insecurity by creating cross-organizations programs to access healthy foods.
Public health or related research Systems thinking models (conceptual and theoretical models), applications related to public health
Abstract
National initiative for cross-model data analytics in early home visiting (HV)
Morgan Taylor, MSPH and Anne Duggan, ScD
Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
background: In 2010, the US began substantial investment in the national scale up of evidence-based early home visiting (HV) for expectant families and families with young children via DHHS’s Maternal, Infant, and Early Childhood Home Visiting (MIECHV) Program. The number of HV models designated as evidence-based has grown from 7 to 20 since then and federal investment is expected to grow. The MIECHV Program supports not only increased HV availability, but also research infrastructure to broaden and strengthen HV impacts. The Home Visiting Applied Research Collaborative (HARC) is a unique and critical part of this infrastructure. HARC is a national research and development platform that brings HV stakeholders together in innovative research to address a consensus-driven research agenda. This presentation focuses on HARC’s partnership with national models and other stakeholders in a Community of Practice (CoP) to advance cross-model data analytics.
purpose: The CoP aims to broaden and strengthen HV impact by increasing HV precision through cross-model research using data analytics to answer high priority questions using existing data.
methods: The CoP began in September 2019. Eleven evidence-based models are participating in Year 1. In Year 2, the CoP will expand to include selected states and other HV stakeholders. CoP tasks across the two years are to: 1) build a shared understanding of existing model data; 2) articulate and prioritize research questions that can be answered using these data; 3) develop research designs and analytic plans to answer a subset of high priority questions; 4) provide technical and capacity support to models and other stakeholders to implement these analytic plans; 5) develop the specifications for data sharing and for data warehousing; and 6) disseminate findings and materials to relevant stakeholders. This presentation will share the CoP’s Year 1 experiences.
Conduct evaluation related to programs, research, and other areas of practice Public health or related research Social and behavioral sciences
Abstract
Data integration to identify infants exposed to opioids in utero: Implications for equitable newborn care
Angela Campbell, PhD, MPH1, Marc Rosenman, MD2, Sami Gharbi, MS1 and Sarah E Wiehe, MD, MPH1
(1)Indiana University School of Medicine, Indianapolis, IN, (2)Chicago, IL
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Data integration between the medical record of a mother and her infant is important for optimal newborn care, but is often logistically problematic. This study utilized electronic health data from a major metropolitan hospital system in Indiana to assess how often a woman’s documented opioid use during pregnancy is recorded in the infant’s medical records and how this relates to the infant’s care. We found that only 55% of mothers who had an ICD code for opioid use during pregnancy were matched to an infant with documented opioid exposure. Black mothers with documented opioid use had 54% decreased odds [95%CI (0.32, 0.71)] of being matched with an infant with documented opioid exposure or neonatal abstinence syndrome (NAS) relative to White mothers. NAS and opioid exposed infants who did not match with an opioid exposed mother either did not receive a Finnegan score to assess withdrawal symptoms, or waited 7.9 hours longer (p<.05), on average, for their first score relative to matched infants. These results suggest that hospitals may be able to identify, formally assess, and adequately manage more infants experiencing opioid withdrawal if there is increased data integration between mother/infant medical charts. Furthermore, this data integration may reduce disparities in care among black and white infants experiencing opioid withdrawal. Given that infants who match with an opioid exposed mother are receiving increased in-hospital monitoring of symptoms, and may require long-term monitoring for other issues, it is important that attention is given to data integration and monitoring within this vulnerable population.
Communication and informatics Public health or related research
Abstract
Development of a data visualization dashboard for public health using participatory design
Elizabeth Heitkemper, PhD, RN1, Betty Bekemeier, PhD, MPH, FAAN2, Melinda Schultz, MA3, Uba Backonja, PhD, MS, RN4, Greg Whitman3 and Colin Barnett5
(1)The University of Texas at Austin, Austin, TX, (2)Northwest Center for Public Health Practice, Seattle, WA, (3)University of Washington, Seattle, WA, (4)University of Washington, Tacoma, Tacoma, WA, (5)University of Texas at Austin, Austin, TX
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
To ensure system accountability, communicate needs to stakeholders, and establish the link between public health activities and population health outcomes, use of data are crucial. However, financial data are notoriously dense and difficult for leaders from local health departments (LHDs) to utilize and communicate. Evidence shows that complex data are better understood by policymakers and the public through straightforward concepts like rankings, maps, and other visualizations. To assist public health professionals with using financial data, the Public Health Activities and Services Tracking (PHAST) team conducted 10 user-centered participatory design sessions and focus groups (DSFG) with partner LHDs to determine the data visualization needs and preferences of public health leaders. Recruitment consisted of emailing partner LHDs to solicit participation. DSFG with between 2 and 5 participants each were scheduled for two hours and conducted using Zoom video conferencing. All DSFG were video and audio recorded and transcribed by Zoom. Data were analyzed using iterative content analysis by four researchers. Our sample represented LHDs that served a broad range of jurisdictions: 4 large central and/or fringe metropolitan, 2 medium and/or small metropolitan, and 4 rural jurisdictions. The following themes were identified: 1) Visualizations can support public health leaders’ work; 2) Divergent desires exist for dashboard functionality based on data skill; 3) Basic is perceived as better for data visualizations. In the first theme, all participants discussed how desired and useful data visualizations are in their work. The second theme highlighted the varying levels of data expertise within LHDs. Most of our participants wanted a powerful dashboard that could run large data analyses and create complex visualizations, but stated that frontline staff would need something more basic. The final theme illustrates user desire for simple visualizations that are straightforward and clearly labelled. Together these findings will inform the creation of a user-informed data visualization dashboard to allow data-driven decision-making by LHDs and professionals.
Administration, management, leadership Communication and informatics Program planning Public health administration or related administration Public health or related research
Abstract
Analytics solution for improving veterans transportation service
Burcu Adivar, PhD, Jennifer Bushelle-Edghill, PhD, Bola Ekezue, PhD, Lee Brown, PhD, Ruchi Rani and Brenda Pettaway
Fayetteville State University, Fayetteville, NC
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
This study introduces a mathematical formulation and an open source solution to optimize the transportation services provided by the Veterans Transportation Service (VTS) program in an effort to minimize missed appointments and transit times as well as to maximize number of veterans serviced. Combinatorial optimization, specifically, capacitated vehicle routing problem (CVRP) is utilized to determine a set of routes designated to transport veterans from their residential address to their medical appointment at the VA hospital. Within two hours after the appointment completion time, veterans are transported from VA hospital to the drop off locations. The two-hour constraint ensures that no patient arrives after the appointment time and is in transit at most two hours to and from their location. The homogenous fleet with maximum capacity of eight passengers, fuel cost, speeds, and real driving times were considered. ODL Studio 1.4.1 was used to process and overlay data points onto a local map of real-world roadways. Solving the CVRP reduced the fleet size, decreased the overall route time, reduced daily travel distance and gas consumption. The results showed that optimized routes generated by open source software were able to pick up all required patients under the two-hour constraint with the use of equal or a smaller number of vehicles from the historical data. This study also showed significant idle time for the drivers where they are waiting for additional appointments or they are able to finish their routes much before their day shift end time. The idle time is proof that additional veterans could use the VTS service. The reduction in usage of vehicles economically is a reduction in operation and maintenance costs involved with the additional vehicles and drivers. Finally, optimizing appointments using ODL Studio or another open source software offers advantages which can be used in conjunction with other initiatives to improve resource allocation for the VTS program
Administration, management, leadership Communication and informatics Other professions or practice related to public health Systems thinking models (conceptual and theoretical models), applications related to public health
Abstract
Telehealth, technology and health disparities
Kathy Wibberly, PhD
University of Virginia, Charlottesville, VA
APHA's 2020 VIRTUAL Annual Meeting and Expo (Oct. 24 - 28)
Telehealth technologies are being used as tools to address health disparities in both the United States and globally. These efforts are often overlooked, yet have significant implications on how we address public health and health disparity challenges related to workforce shortages, workforce training, emergency management (including the management of epidemics), access to health services and chronic disease prevention and management. This session is designed to help public health professionals, planners and administrators gain insight into how telehealth technologies are currently being used, as well as provide a glimpse at emerging telehealth technology innovations and practices. Participants will also engage in a facilitated discussion on how telehealth technologies can be leveraged in current work settings, as well as a brief discussion on policy barriers and opportunities that exist around the use of telehealth technologies. This session will conclude with a presentation of free and low cost resources available to assist public health agencies and organizations with the development of actionable telehealth-enabled solutions for enhancing value and benefit, particularly for minority and underserved populations.
Other professions or practice related to public health Program planning Provision of health care to the public Public health or related laws, regulations, standards, or guidelines Public health or related public policy