International Journal of Environmental Research and Public Health | An Interview with the Author—Dr. Liam O’Neill

Name: Dr. Liam O’Neill
Affiliations: Department of Rehabilitation and Health Services, College of Health and Public Service, University of North Texas, Denton, TX 76203, USA
Interests: mental health and law enforcement encounters; health disparities in vulnerable populations

“The Role of Mental Health, Recent Trauma, and Suicidal Behavior in Officer-Involved Shootings: A Public Health Perspective”
by Liam O’Neill
Int. J. Environ. Res. Public Health 2025, 22(6), 945; https://doi.org/10.3390/ijerph22060945
Available online: https://www.mdpi.com/1660-4601/22/6/945

The following is a short interview with Dr. Liam O’Neill:

1. Congratulations on your recent publication. Could you briefly introduce yourself and your current research focus?

My name is Liam O’Neill, and I am an associate professor at the University of North Texas. When the pandemic arrived in 2020, I began working on COVID-19 and its impact on children. I recently published a paper on the impact of COVID-19 on the homeless population.  However, now there is not as much emphasis on COVID-19, and the number of publications in this area has declined rapidly. And so, I began to research the urgent problem of gun violence in the US from a public health perspective. A huge obstacle for researchers is the lack of quality data about gun violence, especially the type of “big data” that is needed for this research. That is a theme of the current paper, including how hospital databases can fill this gap.  Gun violence is a huge topic, so I narrowed it down to focus on officer-involved shootings. This is my focus right now, and there are many interesting follow-up questions from this paper, such as the alarming increase in the use of the drug “crystal meth” and officer-involved shootings. This data set is unique, as it contains detailed data on both mental health and substance use. So, I approached this problem from a public health perspective, as opposed to law enforcement or psychology.

2. Your recent studies tackle complex public health emergencies. What role do you see for real-time big data analytics in predicting and mitigating future health crises—whether infectious diseases or mental health emergencies?

I am very optimistic about it! I have taught health information systems since the 1990s.  For as long as I can remember, there has been a lot of hype surrounding the use of Electronic Health Records. For the longest time, the hype has outpaced the reality, which has led to a lot of frustration along the way.  With a lot of technology, it takes a long time to develop. Starting around 2012, there was a huge investment in health information technology. Around 2016, we finally made the transition from ICD9 to ICD10.  That was a game-changer, and it has been quite helpful to researchers. So, there has been a lot of improvement on the data side, and the quality of the data has been getting a lot better.  And it is only going to improve.  So it is really on the verge of becoming a disruptive innovation.

However, there are still definitely problems in terms of getting access to the data or certain kinds of data, such as the E-codes that pertain to social determinants. But I think that it is very promising for a whole bunch of things, such as predicting accidents and even the health effects of climate change. With the officer-involved shooting, it is similar to a plane crash because a plane crash is extremely rare. It is not a question of one thing going wrong, but, say, five things going wrong at once.  If four out of five go wrong, you are OK. But if that fifth thing goes wrong, then you have the accident.  In the vast majority of police encounters, there is no violence.  And even if there is violence, normally it is not going to escalate to the point of a shooting. There are also less lethal means of restraint.

As far as real-time data, there is about a one-year lag in data availability, as I am just now getting the data from 20204.

I am very optimistic. The quality and quantity of data have improved significantly over the years, and the pace of change is accelerating. Artificial intelligence is also a huge game-changer.

For years, the bottleneck has not been the data itself, but the shortage of people with the analytical skills to derive meaningful insights from the data. For many years, much of the top talent in data analytics has gone to big tech firms, such as Google or Amazon. Some of my former students went to work for Epic in Wisconsin, and they have performed quite well. More recently, health care has begun to attract some of the top talent in IT. One of my former students is now a Chief Information Officer at a large health care system. There are many interesting problems in health care to work on right now.

3. Did you encounter any significant challenges while you were carrying out this research? And if so, how did you overcome them?

One of the biggest challenges was the amount of data required. I needed almost 20 million hospital records. So that is why I had to use 7.5 years of data in order to get a sufficient sample size.

If I only have two years of data, I have only got about 20 officer-involved shootings. That is just not enough observations, in terms of statistical power.

Also, writing the computer code takes a lot of work. To operationalize the variables requires thousands and thousands of lines of code. And each step must be validated. For each variable, there could be 15 - 20 (ICD10-CM) codes associated with it.   

In terms of methods, there is a lot that goes on behind the scenes. You are limited in the methods section because it can be somewhat boring for the reader. The important thing is to include enough detail so that others can replicate your findings. What the reader wants to know is: “What are the findings?” and “Why does it matter?”.

4. Could you share what your vision is for the future of your research and the kind of contributions you aspire to make in your field in the future?

There is a lot of exciting work being done on the boundaries between disciplines. For example, there is much potential to apply the tools of health economics to other areas, such as law enforcement and psychology. However, this approach requires a lot of data. The same basic methodology that has been used to identify the risk factors for COVID-19 in children can be used to identify the risk factors for officer-involved shootings. The methods are quite well-developed. It can also be used to address other questions, such as: What are the key differences between persons who are obsessed with suicide, also called “suicidal ideation”, as compared to those who take the next step and attempt suicide? These are hugely important questions! Moreover, there is an extensive literature in psychology about these issues. Yet there is not much research from the perspective of behavioral economics.

The importance of negative results is often overlooked, especially from a publication perspective. For example, at the onset of the pandemic, childhood asthma was thought to be a risk factor for developing COVID-19. Many parents kept their kids out of school based on these early guidelines. Further research found that children with asthma were not at greater risk for COVID-19. However, this was only clear in hindsight. With regards to officer-involved shootings, it is important to know what NOT to focus on. For example, the risk of an OIS drops significantly after age 60. Whereas dementia may be a huge problem for society, it does not appear to be a significant issue with regard to OIS. So, it is important to focus on those few causes that are associated with the majority of OIS incidents.

Many of these officer-involved shootings could be prevented. The evidence supports that most events were precipitated by a mental health crisis, such as the loss of a family member. If people had access to mental health services and access to medications to manage schizophrenia and addiction, many of these incidents could be prevented. By the time someone goes to Walmart and starts waving a knife around, by then it may be too late. However, it is still possible that crisis intervention teams could de-escalate the situation. What the evidence shows is that, in many cases, the subject is having a psychotic episode.  

5. What motivated you to choose IJERPH for your submission, and how would you characterize your experience during the publication journey?

I have heard that this is a good journal, and I have read articles in this journal before; it has a good reputation, it has a good impact factor, and the fact that it is international is good.  They also do not get many articles about officer-involved shootings, because in most countries, that is not a problem.

The United States is unique in having so many officer-involved shootings. But I also mentioned in the paper that there were also firearm injuries to police officers in the database, because guns shoot in both directions. And so, I tried hard to be objective and unbiased. It is a very political issue. So, I just presented the evidence, based on what the data showed, and let the reader decide.

For example, homeless people accounted for more than three percent of the OIS incidents.  In my view, this is a huge problem for society, and I think we can do much better. Some may disagree, and that is fine. All I can do is bring attention to the problem and shine a light on what is happening.

My experience was quite good in terms of the journal. I thought the reviews were constructive and helpful.  I gave a thorough response to the reviews. I felt that they got it, like, in terms of my vision for the paper.  The publication process was timely, which is always a good thing.  I was also impressed with the proofreading stage they went through, line by line, to identify any potential errors. So, I appreciate that, because I am sure that it is going to end up improving the quality of the paper at the end of the day.

We sincerely appreciate Dr. O’Neill for sharing his valuable research on applying big data analytics to address critical public health challenges. His work provides important insights into data-driven approaches for improving health care outcomes and community safety. The journal wishes him continued success in his important contributions to advancing evidence-based solutions in these vital research areas.