Practice-based Research (PBR) is a type of research that asks questions that originate from day-to-day clinical practices and collects/analyzes data generated during the course of practice -- for the purposes of understanding and improving effectiveness, efficiency and quality. PBR conducts pragmatic studies tied to the real (“messy”) world of practice, instead of highly structured controlled randomized trials, the results of which are often challenging to apply back to the real world.
The average number of questions that arise during a day in practice is upwards of 15-20. Compared to 20 years ago, many of these questions are now more answerable to a reasonable approximation from the current evidence base and literature. This is because that knowledge has been made more accessible to the clinician by advances like highly cross-linked and regularly updated “electronic texts” like “Up-to-Date”, and the more recent AI/LLM resources like “Open-Evidence” that can synthesize an answer to a complicated natural language clinical question. Despite these tools, there are still many unanswered questions that PBR can help answer.
As with all research, PBR has required steps to optimize its practicality and impact. The first is defining the clinical question carefully. These arise directly from clinicians’ observations in practice about many aspects of care: what is the best way to screen for, diagnose or manage a given problem, how widespread is the challenge among all the patients in the practice or in the community, are there systemic approaches to preventing or affecting the outcomes for a particular challenge, etc. Such ideas of how to improve practice are often called Quality Improvement (QI).
The next step is to decide what data elements are needed to answer the question and planning data collection methods that least disturb day-to-day practice. The data are often collected with a mixed methods approach, that is they include both quantitative data directly from the EMR and qualitative elements obtained from patient reported outcomes or experiences often via questionnaires. These data reflect the actual real-world events going on with patients in the course of their care.
A major requirement of PBR is that one must have the ability to extract data from the EMR, but also to be able to innovate the data capture in the EMR on the front end. This expertise is sometimes hard to come by and is not cheap. Some data collection in PBR can be done more simply and inexpensively, but if an increased scale is desired EMR-based data capabilities are required.
The final step is to analyze the data and design interventions to improve clinical practice. This is usually done over a short time frame, and the process can be repeated until one achieves an acceptable level of improvement. This process can be described as the “plan-do-study-act” or PDSA cycle.
PBR has several advantages: 1) it is designed to be a “byproduct” of the care being delivered so it can be done in parallel with ongoing practice; 2) it is a good beginning form of research because clinicians are motivated by their own practice needs and clinical curiosity, which helps them be and remain engaged; 3) the outcome is pragmatically useful as it can often actually improve practice in the short run.
Interested in partnering with UC San Diego clinician-scientists to work on practice-based research projects? With support from the Krupp Endowed Fund, the UC San Diego Centers for Integrative Health uses PBR to evaluate the impact of our integrative medicine clinical programs and services on a range of health conditions. These studies help generate evidence that directly informs patient care, advances best practices, and improves outcomes across our community. Pilot grants for in-kind research support are available each year. Learn more at cih.ucsd.edu/research/practice-based-research/