Healthcare Intelligence is a content series dedicated to exploring the latest trends, news and research across the entire spectrum of health and wellness.

Today, the healthcare industry generates approximately 30 percent of the world’s data volume, but according to the World Economic Forum, 97 percent of it is not being used. Much of the data collected is unstructured (documentation, notes, photos) and does not easily fit into spreadsheets or databases. That doesn’t mean, however, that there aren’t opportunities to identify underutilized structured data sets that can be used to improve population health outcomes.

The true power of information does not come from merely possessing the data. It comes from creating context and opportunity from the story the numbers tell. In fact, with the right tools, even information collected on something as simple as a PDF could be a rich data source that can be mined and tracked, trended, and run against other data sets.

Maternal and Child Underutilized Data Set

Recently, Telligen helped a state client organize and analyze a maternal and child health data set that had primarily been used for eligibility and authorization of services. Collected electronically via PDF, this data had been previously loaded into separate systems for its intended purpose of establishing eligibility for the child’s prior authorization for the inpatient stay for the mother, and not looked at for its value as a whole population health data set.

A wealth of information such as birthweight, gestational age, Apgar scores at 1 and 5 minutes, the delivery type, and other data points, was being collected on the form and could be incredibly important by assisting the state in determining how to improve their maternal/child health outcomes.

Data Extraction Process

To streamline the data processing of these PDF forms, our developers implemented an automated solution which efficiently extracted information from over 9,000 PDF forms. By automating the extraction of text fields, checkboxes, and other relevant data points, we significantly reduced the manual effort and potential errors associated with traditional data entry methods.

Further, to enhance data analysis capabilities, we enriched the extracted data by mapping the information to the county level, which provided valuable context, enabling researchers to delve deeper into demographic trends, healthcare patterns, and other relevant factors within specific geographic regions.

Getting the Full Picture

Data was analyzed by demographics of mother and infant, gestational age, birthweight, and delivery type. The exercise quickly keyed into some concerning trends and provided the client with actionable information to inform potential improvement activities and share with other stakeholders to affect change.

By overlaying national data to this state level analysis regarding access to care, delivery method, birthweight, and gestational age, the client has a snapshot of where they ranked nationally.

Data-Driven Action

This information was then presented to the state in easy-to-digest graphs and a dashboard. Because this is recent data and can be updated at regular intervals, as desired, the dashboard allows this data to be trended and compared over time. Armed with this new analysis of maternal and infant data in the state, the Agency can make informed decisions about what they would like to focus on, improving and designing policies and programs to support those activities. They can also share information and identify groups that might be best positioned to partner to work toward common goals.

No matter how they choose to use the information, it has the power to impact healthcare outcomes and change lives for mothers and their babies; a transformational opportunity only possible when we look at collected data not for what it is now, but what it could be.