This article brought to you by Melissa. Explore how Health Data Compass is tapping into Census, elevation, altitude datasets to deepen healthcare research insights.
Today’s medical researchers are armed with data, with better-than-ever resources helping clinicians steadily improve patient diagnosis, treatment, and outcomes. Going beyond data related to the human body makes a difference too, accessing insights around external factors such as living conditions, Census coding, and geographic and demographic characteristics. Does elevation or altitude have impact on microtia in newborns? Physicians can now theorize these factors and readily access the data to determine the validity of their hypotheses.
- Health Data Compass supports a number of research projects at any given time
- A principal investigator’s research variables of interest can be diverse and disconnected
- Sourcing clean, correct, and standardized data is difficult; enhancing that data to serve research demands requires sophisticated tools and expertise
- Do elevation and altitude impact the occurrence of microtia in newborns? This physician’s theory required additional data that would meaningfully answer the research hypothesis
- HIPAA compliance must be applied at all points in the data chain, ensuring the privacy and security of protected health information
- Melissa’s fine-grained tools enabled elevation and altitude data to seamlessly integrate with patient records
- A ‘finder file’ including key identifiers helped ensure data cleanliness and proper alignment with formatted records
- Data is multi-sourced and consistently available, synthesized into the Health Data Compass enterprise data warehouse through ETL processes
- Including elevation and altitude data enhancements allows Compass partners to easily query data and analyze results with deep detail
- With enhanced data, researchers can address questions not previously given consideration
- Data enhancements are delivered in compliance with Compass’ data use agreements, featuring end-to-end compliance with HIPAA and HITECH regulations for privacy and security
- Overall quality of data improves, as standardization solves missing data records and inconsistencies in how data is entered by clinicians in the field
- New opportunities to apply data are on the horizon, such as improved specificity of patient geographic information which lends to identification and analysis of rurality and the subsequent impacts on healthcare
The Customer: Health Data Compass
Health Data Compass is an enterprise health data warehouse, integrating patient data from electronic medical records, provider billing data, and omics data to empower its partners with advanced health data analytics. Maintained by the Colorado Center for Personalized Medicine (CCPM) at the University of Colorado Anschutz Medical Campus, Compass is funded by and collects patient data from UCHealth hospitals, Children’s Hospital Colorado, and the University of Colorado Medicine physicians’ group. In its role as data steward and honest broker of the research done on campus, Compass creates enriched single-source data records. Data delivery is accessible to its partners through a variety of options, ranging from self-service de-identified cohort analyses to fully-identified, line-level datasets suitable for advanced analytics. Compass’ cloud analytics infrastructure enables partners with the computational environments necessary to service advanced analytics requirements including statistics and visualizations for bioinformatics, natural language processing, and more.
The Challenge: Improve patient data for wider research use
Health Data Compass traditionally takes on projects instigated by a principal investigator – meeting with campus faculty to identify project needs as well as the variables required to answer specific research questions. Once variables are defined, the team extracts and reports the data, creating a final data product for analysis for grant consideration.
Data gathered by Compass initially exists in Caboodle, the relational database of EPIC, after structured and unstructured data pass through a variety of database types to become consolidated more and more succinctly into tables. This data is then copied into the Google Cloud Platform and made available for partner use via Compass’ analytics infrastructure services. The subsequent data warehouse is the foundation of Compass’ ability to connect patient data across all its different domains.
Supplemental data enhancements from Melissa are commonly applied to Compass data, helping it become more meaningful to the research process. “Our mandate is to ensure data is the most valuable asset it can be – clean, up to date, correct, and enhanced before it is shared with any of our partner stakeholders,” said Tacker Patton, Data Delivery Projects Manager, Health Data Compass. “Melissa broadly supports our mission to bolster patient information and, in turn, strengthens our ability to serve broader research demands across campus.”
In this case, a group of ICU physicians was interested in the potential impact of elevation and altitude on the development of microtia in newborns. Microtia is a congenital abnormality, in which a child is born with an underdeveloped or malformed external ear. Based on observations in campus healthcare facilities as well as their own clinic, these clinicians formed a hypothesis and wanted to apply Compass data to its analysis. The Compass team was tasked with developing a dataset with supplemental elevation and altitude data along with electronic medical record (EMR) data.
The Solution: Elevation and altitude data seamlessly integrates with patient records
Health Data Compass’ staff includes teams in engineering and data delivery, which is then broadly divided into healthcare analysts and developers. Analysts were instrumental in defining the microtia research variables, identifying a data source that would be needed to meaningfully answer the research questions.
“It’s the need on campus that drives us to seek out additional data sources,” added Patton. “In Melissa, we have found broad expertise supported by fine-grained data tools that allow us to integrate datasets of priority value to research at hand.”
Melissa supplied additional information on elevation and altitude data, garnered from the company’s multi-source public and proprietary datasets and its deep expertise in global address validation and enhancement. Melissa’s data quality and enrichment tools pair robust consumer data with a spectrum of third-party datasets, empowering researchers with identity cross-matching that closes gaps in data profiles and delivers a full, single view of the customer.
Variables spanned a set of inclusion and exclusion criteria for the patient cohort, focusing on factors such as children below the age of two, born in certain geographic regions of Colorado, the presence of underlying co-morbidities, and gender. Coupled with elevation and altitude data, raw data was synthesized through an ETL (extract, transform, load) integration process and then ingested into the Compass data warehouse. The Compass team developed SQL code to pull all the required data, formatted as a patient demographics table. This structured subset was delivered to the principal investigator, supported by a statistician’s comparative analysis of the varying elevations and altitudes.
Melissa’s Global Address Database (GAD) provides comprehensive master address information and geographic data like latitude/longitude coordinates, elevation and altitude. The above image shows a bird’s eye view of four addresses available from the 469,067 address points for the county of Denver. The second image shows the elevation profile for each of the four addresses.
The Benefit: Better data breathes life into deeper research
By including elevation and altitude data provided by Melissa, Compass can query patient data based on the elevation of selected clinics, or the altitude of a patient’s residence. Data can be broken up into much smaller strata of one altitude versus another, efficiently guiding the principal investigator to compare the prevalence of microtia in different altitude groups.
“The microtia study is a good example of how Melissa’s data enhancements were critical to completing the research. Their data allows us to fulfill research projects in ways that we haven't previously been able,” added Patton. “We’ve also found that Melissa’s tools and services increase the overall quality of our data – correcting, completing, and standardizing electronic health records that often feature missing information or variations based on how clinicians in the field enter patient data in real-time.”
Through collaboration, data from Melissa connects seamlessly with Compass data. Data is shared with a ‘finder file,’ which includes a list of pieces of identifiable information used to ensure records line up with existing formatting. These identifiers may be fields such as a patient record or member number, phone number, or social security number. And while identifiers are the smartest way to make sure patient data matches record by record, they also illustrate the sensitive nature of healthcare data. Melissa meets the security requirements that support protected health information – a critical factor in meeting compliance regulations even as data is enhanced for ideal research value. Melissa follows all data security standards outlined in HIPAA and HITECH legislation, protecting data security and privacy as required by data use and sharing agreements with Compass.
Melissa is also supporting Compass with additional data enhancements, for example, in the Vision program validating vaccine efficacy in partnership with the Centers for Disease Control (CDC). Compass is one of ten sites selected to build a dashboard reflecting vaccine effectiveness, sharing data with the CDC team bi-weekly. Melissa is providing Census tract data, geographic information, and patient demographics as part of this project.