As part of the firearms special issue, author Brianna Mills has contributed this blog post that articulates with her brief report Firearm assault injuries by residence and injury occurrence location.

Brianna Mills is a Research Scientist at the Harborview Injury Prevention and Research Center at the University of Washington in the United States. Dr. Mills has a PhD in Epidemiology and a MA in anthropology. In addition to firearm injury, her work focuses on efficient, interdisciplinary use of administrative data to address a wide range of injury prevention topics. She is sporadically active on Twitter @BriannaMMills

“Location, location, location” isn’t just the first rule of buying a house. Knowing where something happens is an important part of accurately describing injury circumstances, conducting place-based research on injury risk factors, and ultimately designing effective interventions to prevent further injuries.

Given the importance of complete, accurate, and precise injury location data it may surprise people how often researchers use an injured person’s residential address as a proxy for where his or her injury occurred. Depending on a person’s daily habits and the nature of their injury, where they live might be very distinct, both in distance and in character, from where they were injured. In our Brief report in this issue, my co-authors and I demonstrate how patient age in particular is associated with greater distance between residential and injury location.

I believe researchers use residential address in place of injury location because while we have near-complete capture of where people live we are frequently missing data on where they were injured. The reasons for this level of missingness as I see them are:

• Medical records are a primary source of data on injuries, particularly firearm injuries.
• Whether for patient contact or billing purposes, medical records almost always have a mailing address for patients.
• Compared to mailing address, injury location is much more haphazardly documented. It is rarely a mandatory field, and most often found in free text narratives of the patient’s history or abstracted from the address where emergency medical services picked up the patient for transport to hospital.
• Even when there is an attempt to capture injury location, patients may be unable or unwilling to give enough specific details to generate an address.

Injury location isn’t the only information researchers would like to be more completely and precisely captured in the medical record. Enhanced data collection by frontline medical staff and trauma units has been proposed in order to get better data on injury circumstances and police involved shootings, as well as injury-related health disparities. [Full disclosure: I am involved in the latter project]. The close working relationship between public health researchers and clinicians in many fields can be leveraged for better data, leading to better science/interventions.

But before we ask for more data it’s important to understand why the data aren’t available already. Data that are vital to future prevention efforts are often extraneous to treatment, and asking frontline medical personnel to gather it can present a conflict. Consider the issue of mandatory reporting of domestic violence injuries, and how individuals’ fear of arrest can influence cooperation with medical treatment. Researchers need to acknowledge that research priorities, treatment priorities, and the priorities of injured patients and their families can be in conflict.

So how can we work to resolve these conflicts? We could accept that we work with imperfect data and improve our analytic methods to deal with that uncertainty. We could look beyond our traditional partnerships with hospital-based medical professionals; expand our data sources and work more closely with EMS or police data that place a higher emphasis on where an incident occurred. We can also work more closely with health informatics and programmers to simplify our requests and reduce the burden of data collection.

Lastly, we can work harder to convince people that injury research and prevention are worth the effort that we ask our partners (medical professionals, law enforcement, IT, and patients) to put forth. Do work that is valuable, and ensure that we communicate that value widely.