Presentation
HE9 - Developing a Dashboard for Monitoring System Equity in Maternal Health
SessionPoster Session 2
DescriptionIn 2022, approximately 817 maternal deaths were recorded in the United States (US) [1]. Racial and ethnic minority groups such as American Indian or Indigenous American, Black, some Asian populations, and Hispanic subgroups experience disproportionately higher rates of maternal deaths, comorbidities, and adverse perinatal outcomes compared to non-Hispanic White women [3]. For instance, the maternal mortality rate among Black women in the United States was 2.6 times higher than for non-Hispanic White women in 2021 and 2022 [1,2]. These disparities also extend to care interventions, for example, Hispanic women are less likely to receive epidural analgesia during childbirth [3].
Current efforts to mitigate these disparities in maternal care include educating healthcare professionals about racial and ethnic inequities and implementing quality improvement initiatives like Perinatal Quality Collaboratives [3,5]. While these measures have facilitated collaborative efforts and increased awareness of inequities, they often lack the tools for real-time monitoring of disparities at the hospital level [3]. One recommended actionable step to improve equity in maternal care is to integrate equity dashboards within hospitals to identify and monitor variations in maternal care processes and outcomes [3].
Previous implementations of dashboards within hospitals have demonstrated effectiveness in identifying risk areas which led to the development of interventions, such as training programs for healthcare professionals [7,8]. However, these dashboards often lack focus on equity as data are usually not disaggregated by race and may not incorporate real-time data updates, limiting their utility in identifying variations in care across racial groups [6]. Additionally, existing health equity dashboards have predominantly utilized static, retrospective data and are often limited to state or county levels [5,6]. This limitation hampers the ability to monitor ongoing disparities in day-to-day healthcare activities within individual hospitals. Moreover, the reliance on publicly accessible data restricts the granularity needed to implement targeted interventions [5]. There is a clear need for hospital-specific dashboards that can provide actionable insights to healthcare providers and administrators [7]. Hence, the aim of our research is to develop a hospital-level maternal equity dashboard that supports the monitoring and identification of variations in maternal care processes and outcomes across different racial/ethnic groups. By integrating institutional data with disparity analysis, we seek to create a tool that not only visualizes key maternal care metrics but also promptly supports decision-making to improve patient safety and equity.
The maternal care equity dashboard we developed utilized data collected from the labor and delivery (L&D) unit and the mother-baby unit (MBU) of a large academic health system in the Southeastern United States. The dataset includes incident reports data, and delivery data of patients who delivered in 2019 and 2020. Patients' demographic data that included medical record numbers, delivery age, race, ethnicity, spoken language and length of stay (LOS) collected from the clinical data warehouse of the hospital was also included in this study. The data were cleaned, and patient delivery data were automatically linked to their demographic data using Microsoft Excel 2020 and Python scripts Jupyter Notebook 7.0.8 (Project Jupyter, Version: 7.0.8, Berkeley, California, USA). The data were then disaggregated by race/ethnicity: non-Hispanic White (NHW), non-Hispanic Black (NHB), Hispanic and Others. Measures related to process and outcomes included on the dashboard were selected based on previously identified disparities in maternal health outcomes, areas of concern identified through hospital-level data analysis, other measures modifiable through the delivery of care and insights gathered from clinicians. The dashboard was iteratively developed adhering to Human-Centered Design (HCD) principles, such as accessibility, ease of use, and user engagement, to ensure it effectively meets the needs of its users [10,11]. Tableau Desktop 2023.2.0 (Tableau Software, Version: 2023, MountainView, California, USA) was used to develop visualizations for the dashboard. Statistical analysis including chi-square test and Kruskal-Wallis rank sum test, was performed to assess racial differences in dashboard measures using R version 4.3.1 (R Core Team, 2023).
After the cleaning and linking processes, delivery data for 5,786 patients and 528 incident reports was incorporated into the equity dashboard. The dashboard included 18 visualizations of process and outcome measures, including anesthesia type, delivery method, surgical blood loss, and maternal morbidity indicators. These visualizations offer user-friendly elements like drop-down menus and search filters, allowing for interactive exploration of the dashboard. Significant differences were identified in several measures, such as anesthesia type, delivery method, labor and delivery (L&D) complications, and preeclampsia/eclampsia rates across racial/ethnic groups. NHB patients had the highest usage of general anesthesia (5.5%) and the highest rate of L&D complications (25.2%). We are currently conducting usability testing of the dashboard with 10-12 maternal care healthcare professionals, using an evaluation framework that assesses task performance, interaction workflow, perceived engagement, and potential utility [12]. The findings from this usability testing will inform improvements in subsequent iterations of the dashboard.
Addressing racial and ethnic disparities in maternal care requires innovative solutions that are both data-driven and user-centric. Our proposed maternal equity dashboard bridges the gap between awareness and action by providing hospitals with the tools necessary to monitor and identify variations in maternal care processes and outcomes. This approach not only enhances patient safety but also contributes to the broader goal of achieving equity in maternal care. The development of a hospital-level maternal equity dashboard represents a significant advancement in addressing systemic biases in maternal healthcare. By leveraging on accurate data, maternal care equity dashboards can empower healthcare providers to make informed decisions that can reduce disparities and improve outcomes for all patients.
Acknowledgement：
This project was supported by Agency for Healthcare Research and Quality [Grant no. 1R03HS027680]; Natural Sciences and Engineering Research Council of Canada [Grant no. RGPIN-2022-04878]; NSERC Undergraduate Student Research Awards; Summer Undergraduate Data Science Opportunities Program (The Data Sciences Institute, University of Toronto); and MIE Summer Research Awards (Department of Mechanical and Industrial Engineering, University of Toronto).
Reference:
[1] D. L. Hoyert, “Maternal Mortality Rates in the United States, 2021,” 2023.
[2] D. L. Hoyert, “Maternal Mortality Rates in the United States, 2022,” 2024.
[3] E. A. Howell et al., “Reduction of Peripartum Racial and Ethnic Disparities: A Conceptual Framework and Maternal Safety Consensus Bundle,” Journal of Obstetric, Gynecologic & Neonatal Nursing, vol. 47, no. 3, pp. 275–289, May 2018.
[4] A. Thomas, S. Krevat, and R. M. Ratwani, “Policy Changes To Address Racial/Ethnic Inequities In Patient Safety,” Health Affairs Forefront.
[5] J. Gallifant et al., “Disparity dashboards: an evaluation of the literature and framework for health equity improvement,” The Lancet Digital Health, vol. 5, no. 11, pp. e831–e839, 2023.
[6] O. Olakotan, J. N. Lim, and T. Pillay, “Challenges and Opportunities in perinatal public health: The utility of perinatal health inequality dashboards in addressing disparities in maternal and neonatal outcomes,” Oct. 2023.
[7] S. Yi et al., “Designing and developing a digital equity dashboard for the emergency department,” J Am Coll Emerg Physicians Open, vol. 4, no. 4, p. e12997, Jun. 2023.
[8] “New York State Maternal and Child Health (MCH) Dashboard.” Available: https://apps.health.ny.gov/public/tabvis/PHIG_Public/mch/reports/#state
[9] R. Harte et al., “A Human-Centered Design Methodology to Enhance the Usability, Human Factors, and User Experience of Connected Health Systems: A Three-Phase Methodology,” JMIR Human Factors, vol. 4, no. 1, p. e5443, Mar. 2017.
[10] I. Göttgens and S. Oertelt-Prigione, “The Application of Human-Centered Design Approaches in Health Research and Innovation: A Narrative Review of Current Practices,” JMIR mHealth and uHealth, vol. 9, no. 12, p. e28102, Dec. 2021.
[11] D. Norman, “The Design of Everyday Things,” New York: Basic Books, 2013.
[12] M. Karami, M. Langarizadeh, and M. Fatehi, “Evaluation of Effective Dashboards: Key Concepts and Criteria,” Open Med Inform J, vol. 11, pp. 52–57, Oct. 2017.
Current efforts to mitigate these disparities in maternal care include educating healthcare professionals about racial and ethnic inequities and implementing quality improvement initiatives like Perinatal Quality Collaboratives [3,5]. While these measures have facilitated collaborative efforts and increased awareness of inequities, they often lack the tools for real-time monitoring of disparities at the hospital level [3]. One recommended actionable step to improve equity in maternal care is to integrate equity dashboards within hospitals to identify and monitor variations in maternal care processes and outcomes [3].
Previous implementations of dashboards within hospitals have demonstrated effectiveness in identifying risk areas which led to the development of interventions, such as training programs for healthcare professionals [7,8]. However, these dashboards often lack focus on equity as data are usually not disaggregated by race and may not incorporate real-time data updates, limiting their utility in identifying variations in care across racial groups [6]. Additionally, existing health equity dashboards have predominantly utilized static, retrospective data and are often limited to state or county levels [5,6]. This limitation hampers the ability to monitor ongoing disparities in day-to-day healthcare activities within individual hospitals. Moreover, the reliance on publicly accessible data restricts the granularity needed to implement targeted interventions [5]. There is a clear need for hospital-specific dashboards that can provide actionable insights to healthcare providers and administrators [7]. Hence, the aim of our research is to develop a hospital-level maternal equity dashboard that supports the monitoring and identification of variations in maternal care processes and outcomes across different racial/ethnic groups. By integrating institutional data with disparity analysis, we seek to create a tool that not only visualizes key maternal care metrics but also promptly supports decision-making to improve patient safety and equity.
The maternal care equity dashboard we developed utilized data collected from the labor and delivery (L&D) unit and the mother-baby unit (MBU) of a large academic health system in the Southeastern United States. The dataset includes incident reports data, and delivery data of patients who delivered in 2019 and 2020. Patients' demographic data that included medical record numbers, delivery age, race, ethnicity, spoken language and length of stay (LOS) collected from the clinical data warehouse of the hospital was also included in this study. The data were cleaned, and patient delivery data were automatically linked to their demographic data using Microsoft Excel 2020 and Python scripts Jupyter Notebook 7.0.8 (Project Jupyter, Version: 7.0.8, Berkeley, California, USA). The data were then disaggregated by race/ethnicity: non-Hispanic White (NHW), non-Hispanic Black (NHB), Hispanic and Others. Measures related to process and outcomes included on the dashboard were selected based on previously identified disparities in maternal health outcomes, areas of concern identified through hospital-level data analysis, other measures modifiable through the delivery of care and insights gathered from clinicians. The dashboard was iteratively developed adhering to Human-Centered Design (HCD) principles, such as accessibility, ease of use, and user engagement, to ensure it effectively meets the needs of its users [10,11]. Tableau Desktop 2023.2.0 (Tableau Software, Version: 2023, MountainView, California, USA) was used to develop visualizations for the dashboard. Statistical analysis including chi-square test and Kruskal-Wallis rank sum test, was performed to assess racial differences in dashboard measures using R version 4.3.1 (R Core Team, 2023).
After the cleaning and linking processes, delivery data for 5,786 patients and 528 incident reports was incorporated into the equity dashboard. The dashboard included 18 visualizations of process and outcome measures, including anesthesia type, delivery method, surgical blood loss, and maternal morbidity indicators. These visualizations offer user-friendly elements like drop-down menus and search filters, allowing for interactive exploration of the dashboard. Significant differences were identified in several measures, such as anesthesia type, delivery method, labor and delivery (L&D) complications, and preeclampsia/eclampsia rates across racial/ethnic groups. NHB patients had the highest usage of general anesthesia (5.5%) and the highest rate of L&D complications (25.2%). We are currently conducting usability testing of the dashboard with 10-12 maternal care healthcare professionals, using an evaluation framework that assesses task performance, interaction workflow, perceived engagement, and potential utility [12]. The findings from this usability testing will inform improvements in subsequent iterations of the dashboard.
Addressing racial and ethnic disparities in maternal care requires innovative solutions that are both data-driven and user-centric. Our proposed maternal equity dashboard bridges the gap between awareness and action by providing hospitals with the tools necessary to monitor and identify variations in maternal care processes and outcomes. This approach not only enhances patient safety but also contributes to the broader goal of achieving equity in maternal care. The development of a hospital-level maternal equity dashboard represents a significant advancement in addressing systemic biases in maternal healthcare. By leveraging on accurate data, maternal care equity dashboards can empower healthcare providers to make informed decisions that can reduce disparities and improve outcomes for all patients.
Acknowledgement：
This project was supported by Agency for Healthcare Research and Quality [Grant no. 1R03HS027680]; Natural Sciences and Engineering Research Council of Canada [Grant no. RGPIN-2022-04878]; NSERC Undergraduate Student Research Awards; Summer Undergraduate Data Science Opportunities Program (The Data Sciences Institute, University of Toronto); and MIE Summer Research Awards (Department of Mechanical and Industrial Engineering, University of Toronto).
Reference:
[1] D. L. Hoyert, “Maternal Mortality Rates in the United States, 2021,” 2023.
[2] D. L. Hoyert, “Maternal Mortality Rates in the United States, 2022,” 2024.
[3] E. A. Howell et al., “Reduction of Peripartum Racial and Ethnic Disparities: A Conceptual Framework and Maternal Safety Consensus Bundle,” Journal of Obstetric, Gynecologic & Neonatal Nursing, vol. 47, no. 3, pp. 275–289, May 2018.
[4] A. Thomas, S. Krevat, and R. M. Ratwani, “Policy Changes To Address Racial/Ethnic Inequities In Patient Safety,” Health Affairs Forefront.
[5] J. Gallifant et al., “Disparity dashboards: an evaluation of the literature and framework for health equity improvement,” The Lancet Digital Health, vol. 5, no. 11, pp. e831–e839, 2023.
[6] O. Olakotan, J. N. Lim, and T. Pillay, “Challenges and Opportunities in perinatal public health: The utility of perinatal health inequality dashboards in addressing disparities in maternal and neonatal outcomes,” Oct. 2023.
[7] S. Yi et al., “Designing and developing a digital equity dashboard for the emergency department,” J Am Coll Emerg Physicians Open, vol. 4, no. 4, p. e12997, Jun. 2023.
[8] “New York State Maternal and Child Health (MCH) Dashboard.” Available: https://apps.health.ny.gov/public/tabvis/PHIG_Public/mch/reports/#state
[9] R. Harte et al., “A Human-Centered Design Methodology to Enhance the Usability, Human Factors, and User Experience of Connected Health Systems: A Three-Phase Methodology,” JMIR Human Factors, vol. 4, no. 1, p. e5443, Mar. 2017.
[10] I. Göttgens and S. Oertelt-Prigione, “The Application of Human-Centered Design Approaches in Health Research and Innovation: A Narrative Review of Current Practices,” JMIR mHealth and uHealth, vol. 9, no. 12, p. e28102, Dec. 2021.
[11] D. Norman, “The Design of Everyday Things,” New York: Basic Books, 2013.
[12] M. Karami, M. Langarizadeh, and M. Fatehi, “Evaluation of Effective Dashboards: Key Concepts and Criteria,” Open Med Inform J, vol. 11, pp. 52–57, Oct. 2017.
Event Type
Poster Presentation
TimeTuesday, April 14:45pm - 6:15pm EDT
LocationFrontenac Foyer