In Conversation With… Hardeep Singh, MD, MPH
Editor's note: Hardeep Singh, MD, MPH, is Chief of the Health Policy, Quality and Informatics Program at the Houston Veterans Affairs (VA) Center for Innovations in Quality, Effectiveness and Safety and an Associate Professor of Medicine at the Michael E. DeBakey VA Medical Center and Baylor College of Medicine. He has conducted extensive multidisciplinary research supported by the VA, AHRQ, and NIH and is now a nationally recognized expert in electronic health record–related patient safety issues and diagnostic errors. We asked him to speak with us about becoming a patient safety researcher.
Dr. Robert Wachter, Editor, AHRQ WebM&M: I know you began your professional career as a full-time clinician. When you decided to embark upon a research career and leave what I assume was a fairly stable clinical job, what were your biggest worries?
Dr. Hardeep Singh: Salary support for researchers is always an issue, so my biggest worry was not getting funded through grants. Another disadvantage was my lack of a research fellowship and having to learn most of what I needed to know while on the job. In 2003–2004, I was already several years out of residency and that made me ineligible for certain types of research career development grants. Leaving an environment of stability and moving to a research career filled with uncertainty took me a while getting used to.
RW: Was there a moment where you realized you were going to make it?
HS: I must say it took a lot of perseverance. In the first 2 years of my transition I worked full-time as a clinician and applied for several small grants and career development awards; it's the latter type of grant that helps you transition from a full-time clinician to at least 60% to 75% research. But funding was tight for patient safety—it was just after the time period when AHRQ had funded many safety grants soon after the IOM report came out. It took about 12 unsuccessful grant submissions to various federal agencies and foundations and a lot of persistence, and I admit that I almost decided to remain a full-time clinician. But then one hit turned my professional life around and that was the K12 award, an institutional 2-year faculty development award given to junior faculty as part of NIH's K30 training program. When I received the news about the K12 award in 2005, I was told the committee not only liked my proposed project but also unanimously recognized the persistence and effort I had made and felt I deserved a chance. I then realized that my after-hour research endeavors might have paid off. There wasn't any looking back after that—I finally got a K23 grant [a larger 5-year faculty development grant] and have stayed in business ever since. But even now I think there's uncertainty. I am always thinking about what's next and how are we going to support future research. So it could be a revolving door, and sometimes it's challenging, but I love what I do and that keeps me going.
RW: Talk about mentorship, how important it's been, and what makes a good mentor for you.
HS: Mentorship is almost everything. I was fortunate to receive great mentorship especially given my unique early challenges. Laura Petersen, my primary mentor, was not only invested in the research area that I had chosen but she was really invested in my professional career—and I think those are the hallmarks of a good mentor. It's about more than just a research area, it's about a person, and she took interest in both. Despite all the grant rejections I received, she remained encouraging, and still is to this day. I also worked with Eric Thomas, Laura's husband, who works at the University of Texas. Having them both as mentors was like getting the best of both worlds. Mentorship is the key to success of any researcher, but especially for people who have not done a whole lot of research before and come from a non-fellowship background like I did. I would go out on a limb and say that without mentorship it's really hard to become an independent researcher.
RW: If you were advising a young researcher starting out, in addition to mentorship, what are the other things you'd tell them to look for as they consider different opportunities? Someone who is finishing a fellowship or comes to you at the end of a residency and says, "I want to be a researcher in safety or quality," and asks for your advice, "What should I be looking for in my training and what should I be looking for in a job?"
HS: It's important to find training programs that can support innovation, create the right mentorship environments, and are collaborative. Quality and safety touch upon so many fields in addition to medicine, such as human factors, informatics, cognitive science, and social sciences, and it takes a multidisciplinary approach to be successful. I would say that people interested in quality and safety should always try to surround themselves with people who have expertise in different fields both within and outside medicine and who collaborate well. I also think institutional leadership and commitment to safety is important in terms of any training program or job. Your mentor support is important, but if your program or institutional leaders are supportive of patient safety and quality research, it's an added bonus.
I would also add that you cannot do it alone. You need a team structure that includes support staff such as research assistants, coordinators, and other team members who support you in your work. The other advice I generally give to most aspiring researchers now is that research is generally not a 9-to-5 job. Patient safety is a dynamically evolving field, and it takes a lot to advance the science while also managing your other professional and personal duties. I think most researchers should expect to put in at least some additional hours on weekends or evenings to get the work done and to create a productive environment for themselves, especially in this competitive funding environment. Many of my good ideas came to me when I wasn't officially at work.
RW: You talked about choosing the research direction versus both your prior clinical work and operational work. But I guess one of the questions in quality and safety research relates to the fact that clinical delivery systems are trying to figure out how to improve quality and safety. That means that there's going to be some tension between the acts of research in improvement and the operational work to make improvement happen. How do you strike that balance?
HS: You're right—in patient safety there's a very fine line between research and operations. Clearly, delivery system operations want fixes and solutions, and sometimes we just don't have clear or timely answers. At times it's hard to push for research that seems to interfere or slow down clinical operations. In our work, we have tried to leverage this tension as much as possible. I try to take on opportunities that would be most relevant to our work and see how I could use our past research to create an impact on what the operational partner wants to do, or I try to see whether we can strike collaborations between our research group and the operational partner that will lead to future success and products, including potential publications or grants.
As an example, I have been asked to serve on many national VA committees over the years and participate in initiatives related to the safe management of diagnostic test results. I regularly engage with VA operations partners in other safety-related work. This partnership has made a tremendous difference in the way I think about my research now. I have learned from them, and at the same time I've been able to contribute to their mission to improve safety, and they have been very supportive of our research on missed test results and diagnostic errors. I even have some papers with VA operational partners as co-authors. Some time ago, VA health services leadership announced that they are introducing more partnership research into their research grant portfolios. The fact that our research shop had done this kind of research in the past instantly became a huge advantage for us.
Our work has been instrumental for some of the VA initiatives on improving communication of test results. While serving on a task force on improving safety in this area, we helped develop guidance documents for the field—a research informed toolkit to help people on the frontlines manage test results related safety issues at the facility level. I also served as co-chair of a committee to revise VA's national policy in this area. So we're talking about 150 facilities of the VA, impacting about 300,000 clinicians nationwide, and probably impacting millions of patients. Now, that is some real impact! We've been working together very closely for the last year in trying to make sure we produce guidance that is useful, practical, evidence-based, research-driven to the extent possible, but also realistic so we don't overburden the field.
RW: That's great. And it does highlight one of the advantages of straddling the research and operational line, which is if you can do this work well, it opens up another source of funding for your work.
HS: Absolutely, we are always hopeful of those funding sources. Federal funding for patient safety is very limited.
RW: As I look through your publications, it seems like your research has been fairly broad within the safety field—diagnostic errors, handoff errors, safety issues around technology, electronic medical records, issues around trainees. How do you think about that common question among researchers—the benefits and risks of focus versus breadth?
HS: In the beginning I was fairly focused on diagnostic errors in primary care. But as I studied the problem further, I realized that diagnostic error sat at the conceptual intersection of several fields. So my work started evolving more into informatics, human factors, cognitive psychology, and information technology. Some other tangential opportunities just happened because you need to write a lot of papers early in your career. So while I was a junior investigator, if there was an opportunity to write a first author paper on a related topic, I would take it if my mentors thought it was useful and if the general area was patient safety related. Over time though, I have found common threads between the breadth of topics I studied, and I often learned something new from a seemingly tangential topic and applied it into my primary focus area. I think as long as you can use the breadth of the topics to improve your productivity and creativity, it can work. While I would say for junior researchers it's good to be focused, the types of problems that we're trying to solve in quality and safety are so broad that it's okay to evolve over time. It took me a lot of effort to try to connect the dots between information technology, informatics, and diagnostic errors, but I surrounded myself with people who are experts in those fields. One expert is Dean Sittig, who I work with on many informatics and electronic health record–related patient safety issues. I have learned a lot of useful concepts for diagnostic error from this collaboration.
RW: How opportunistic do you need to be as a researcher? Let's say you weren't fundamentally interested in IT-related safety issues, but you see a huge federal initiative in that area, or you're reading the literature, and it seems to be getting hot. Does that move you in one direction or the other, or is that a mistake to move in a direction that's not exactly what your core interest was simply because there's national interest or funding?
HS: For me, passion for my primary interest area, diagnostic error, drove me the most. I have tried not to be very opportunistic for topics just because the current public or federal interests are focused there. We were doing work on EHR-related safety issues including missed test results for many years before electronic health records became hot. And some of my ventures into IT-type safety issues were built nicely on our previous sociotechnical approaches to understanding and improving follow-up of abnormal test results in EHRs. Being opportunistic might be good for people who really need funding to get started or are looking for ideas to get their career going. But if you already have great ideas on your own and work on topics that are important to improve patient care, it's generally not a good idea to try to venture into every hot topic that comes along. That might make you unfocused and unbalanced. If you have passion to solve a specific quality or safety problem, that's what you should do. That is what helps push the science forward.
RW: Certainly you were interested in diagnostic errors before anyone was paying any attention to that at all.
HS: Absolutely, and that goes back to the time that I was in practice. Many senior researchers used to caution me saying, "Are you sure you want to study this? It seems too hard and we're not really sure if this is an important area." And my answer was always the same, "Well one day it will be." I was so pleased to see the AMA Patient Safety Report in Ambulatory Care, which came out in 2011. It really tried to bring this problem to the limelight. Diagnostic error is in the media quite often now. I was also a guest editor for a recent BMJ Quality and Safety supplement on diagnostic error. I think the field is slowly coming together, even though we have miles to go.
RW: Some of this is just having the confidence that if you care, it's an important issue and the world will eventually catch up. It sounds like you've done that with diagnostic errors, you've done that with IT-related safety hazards. You got into this about a decade ago and you had certain hopes and dreams. What are the things that keep you going? What are the joys of this as a career and what sustains you?
HS: As part of my work, I hear of many, many patient stories about misdiagnosis, and we have much to do in reducing diagnostic error and improving patient care in this area. Advancing and contributing to a research agenda for a topic so fundamentally important is one factor, but another is using science to make an impact on policy and practice. I study errors in primary care and find that there are many current pressures on primary care clinicians that would also need to be addressed. Some of our research has pushed this area into national recognition, and it has been great to watch as we have gained much-needed momentum. Ultimately it's all about making a difference in patient outcomes and changing the system where we clinicians practice so we can give the best possible care. It's nice to be recognized by the scientific community too. In 2012, I received the Alice S. Hersh New Investigator Award from AcademyHealth for innovative and high impact research on diagnostic error, which was very gratifying.
RW: Where do you think diagnostic errors will be in 5 to 10 years? It seems like the field is still so much in its infancy.
HS: Yes, we are still defining the "basic science" of diagnostic error but beginning to make some headway. I am the principal investigator for AHRQ's Diagnostic Error in Medicine conference and one of the themes for our recent 2013 conference was measurement. I think we really need to get the measurement right first, because what I call a diagnostic error might not be what others in the field call a diagnostic error, and measurement is often very setting-specific. We need to get the measurement science right. Some areas like missed test results are more advanced conceptually, and we are constantly exploring how IT can be best used to improve diagnosis. Several patient empowerment strategies are gaining traction. But it's going to take some time for all of this to come to fruition so I hope to be busy for years to come.
Previous interviews can be heard by subscribing to the Podcast