That's not real research

Digital health and informatics sit at the intersection of humans, healthcare, information, and technology, and working in this interdisciplinary space can result in misunderstanding and confusion. This is evident when talking about research. Clinical and engineering colleagues have told me that our research is not real research. A friend and PhD student in biochemistry applied visualization techniques to metabolic networks for metabolic profiling. His biochemistry monitor loved the project, but his computer science (visualization) mentor was reluctant and embarrassed to put his name on a poster publication, because--you guessed it--applying known techniques to a new data source isn't real research.

I currently serve on an NIH Advisory Committee to the Director Working Group on Novel Alternative Methods (complementary approaches to the use of animals in research). The majority of what the group talks about is over my head, and I’ve wondered how I could contribute to the recommendations we will make to the NIH for investment in this area. A theme kept recurring in our discussion in Washington, DC last month (see picture of NIH Building 1): Biologists are using a lot of data to train new computational techniques, but the data do not harmonize with data from other labs. Also, they are building organoids and microphysiological systems that are not easily used by other groups. These biologists should make sure their research can be used by others. My contribution finally became evident:

Just as biologists would not trust me to ensure the biology was right in the computational models, we should not expect biologists to know how to make their data interoperable and their complex models as usable as an iPhone out of the box. There are people whose research focuses on data management and standards and human factors, and we should respect the need for their expertise like we respect the need for the expertise of a biologist or a clinician or an engineer.

So what is the type of research we do in digital health and clinical informatics?

In general, I’d say that we do applied research, which

focuses on developing practical solutions to real-world problems. It involves the use of existing knowledge, theories, and techniques to address specific problems…and may involve the development of new products, technologies, or processes.

Applied research still generates new knowledge, but it’s knowledge about how to improve the efficiency and effectiveness of those products, technologies, or processes.

Two journals in our field illustrate very different approaches to applied research. The Journal of Biomedical Informatics

publishes research on new methodologies and techniques that have general applicability and form the basis for the evolving science of biomedical informatics. Papers should focus on a real-world biomedical or clinical problem, develop a novel approach to address the problem, and evaluate its appropriateness in comparison to the current state-of-the-art methods.

Applied Clinical Informatics publishes on

clinical information systems, information technology development, deployment, and evaluation, socio-technical aspects of information technology and health IT training.

So you tell me: Is it real research to come up with a new algorithm for determining whether a concept in clinical notes is negated or to apply an existing method like process mining to identify gaps in care or to determine the best way to implement a new program for pharmacist-led telehealth medication management program for veterans receiving oral antineoplastic therapies?