Linking healthcare data, public data, and place

This body of research comes out of healthcare data, public health, epidemiology, and applied statistics. But the work was never only about health. It was about how you get closer to what is actually happening when the information is split across systems, stored in different forms, and uneven in what it can show.

Integrating electronic health record, cancer registry, and geospatial data to study lung cancer in Asian American, Native Hawaiian, and Pacific Islander ethnic groups dealt with that directly. No single source could do the job alone. Electronic health records captured part of the picture. Cancer registry data added another part. Geospatial data brought in place. Linking them made it possible to study populations that are often lumped together and, because of that, poorly understood.

Multilevel social factors and NICU quality of care in California asked a different question, but the underlying problem was similar. Quality of care is not produced inside a hospital and nowhere else. It is shaped by the social conditions around it, the institutions people move through, and the uneven realities attached to different places. Looking at those layers together gave a more honest view of why care differs across settings.

An analysis of lung cancer screening beliefs and practice patterns for community providers compared to academic providers focused on variation across provider environments. Community and academic settings do not run the same way. They work under different pressures. They have different routines, different assumptions, and different constraints. That shows up in screening patterns, and it matters.

What ties this research together is the work behind it. Data had to be gathered, stored, queried, cleaned, linked, and interpreted. Some of it came from organizations’ own systems, including registry records, clinical data, and research datasets. That information did not arrive in a clean, ready form. It had to be handled on its own terms first. Other data came from public sources such as the Census and the American Community Survey, which added demographic, neighborhood, and social context. Geospatial analysis helped show how those layers lined up on the ground.

That matters because a single dataset usually tells a partial story at best. Internal data often has to be pulled together before it can say anything at all. It has to be structured, checked, and understood for what it is and what it is not. Public data can then add the surrounding conditions that help explain why a pattern looks one way in one neighborhood, another way in another, and differently again across institutions. Maps and statistical analysis can make that visible. Interpretation is what keeps it from becoming dead information.

This is part of why the work belongs here. The subject was healthcare, but the structure of the work carries into schools, nonprofits, and small governments without much strain. The same problem shows up over and over. Information lives in too many places. It is incomplete, inconsistently gathered, hard to compare, or too detached from context to mean much on its own. The work is to gather it, make it usable, connect it to other relevant information, study the pattern carefully, and turn the result into something people can actually use.

It also shows why cartography and interpretation matter alongside analysis. A dataset can be correct and still not tell you much. The work is in asking the right question, looking closely, showing the pattern clearly, and tying it back to the lived and institutional reality behind the numbers. More information is not the point. Information that can actually be understood and used is the point.

Related publications

Integrating electronic health record, cancer registry, and geospatial data to study lung cancer in Asian American, Native Hawaiian, and Pacific Islander ethnic groups
https://pubmed.ncbi.nlm.nih.gov/34001502/

Multilevel social factors and NICU quality of care in California
https://pubmed.ncbi.nlm.nih.gov/32157221/

An analysis of lung cancer screening beliefs and practice patterns for community providers compared to academic providers
https://pubmed.ncbi.nlm.nih.gov/30375235/