Open Data Changes the Way We Look at Community Health
In 2014, the City of Pittsburgh, in partnership with the Allegheny County and the University of Pittsburgh, began to make City data open and accessible to everyone. Our regional open data platform, Western Pennsylvania Regional Data Center, currently houses over 250 data sets. As part of the Open Data effort, we’ve reached out to different Pittsburgh residents, community groups, non-profits, and researchers to see how they have been using this data to improve Pittsburgh.
Zan Dodson is a postdoctoral researcher at the University of Pittsburgh’s School of Public Health. His research focuses on how human interactions with the environment can influence health outcomes.
Tell me about yourself and what research you focus on?
I am a geographer by training and so I have always been interested in spatial relationships. Everything in life is governed by space and the way we interact with the environment. How do the spaces we live in act as a barrier or facilitator for humans? How do humans manipulate the environment and what impact does this have on their health and well-being? How does the built environment facilitate the spread of some diseases?
What other spatial relationships affect health?
In a city, the urban heat island effect can have negative health outcomes. This is when urban area or metropolitan area is significantly warmer than its surrounding rural areas due to human activities thereby increasing pollution levels. This spurred a new trend for rebuilding urban environments to mitigate effects of pollution and emissions. That’s why we focus on having green spaces, both to reduce the effect of emissions as well as create a welcoming space to relax and be social. “Forest bathing” or spending time in green spaces can be incredibly beneficial to mental health.
How do you measure human-environment interactions?
One of the big things we look for are spatial correlations. The first law of geography states “everything is related to everything else, but near things are more related than distant things.” Certain groups may have a greater risk for infectious disease to due to their environment. I look more specifically at spatiotemporal dynamics and examine changes across space and time for a region or population. I have developed a set of methods that I can apply to several different situations from vector born disease in Columbia and HIV/AIDS in Africa to the opioid epidemic here in the United States.
Which project had the clearest use of city data?
I used public safety police data on my most recent project. I gathered opioid-related arrests (drug possession and drug trafficking) from 2010 to 2015 and made a spatiotemporal model to see how this epidemic has moved across Pittsburgh. We noted the chronic hot spots but as we ran this time analysis, we were able to detect the acute hot spots that lasted for short periods of time. Noticing these trends allows us to target interventions more effectively. We can supply these areas with clean needle exchanges, Narcan kits, and readily available medical aid.
Why are you interested in acute hot spots?
A hot spot indicates this area is statistically more likely to have these opioid-related events occur than other areas. These acute hot spots would have been missed through normal spatial modeling. They give us a better understanding of the progression of the epidemic. These spots can be as small as street corners and easily targeted with prevention measures.
Why might there by a time component to the epidemic?
Shipping or trafficking schedules, seasonal trends, school year. Using this information, we can iterate towards a predictive model to determine where our resources should be targeted.
What was your take-away?
I used a non-traditional approach to public health. Instead of waiting for health data which takes a long time to release and has a lot of bureaucratic red-tape, I was able to use city-collected public safety data to see trends in health. Also, we need to move to spatiotemporal modeling to examine population trends. Spatial modeling does not give us the same level of detail.
Did you share these results?
I shared them with the police and Department of Health Services at the city and county level. I am also working with the Medical Director of Emergency Medical Services (EMS) in Pittsburgh to create a “Leave Behind” program where the first responders can provide Narcan kits for those who refuse transport to the emergency department.
What are the next steps in your research?
To figure out why this is happening in these communities. What are the factors in your daily life that may influence your health outcomes? I plan to use pollution data, public safety data, and building data from the City to get a better understanding of the social influencers of health in Pittsburgh.
How was your experience working with the City’ s open data?
Not many other cities release data like this. All the data sets the City has released talk to each other and are valuable assets for researchers across all fields.
This is a profile from our Open Data Progress Report. Check back next week to see the full report! In the meantime, you can take a look at the City’s open data on Burgh’s Eye View — and see other datasets on the regional data center!
