Rare Disease Study Gets A Boost From Crowdsourced Patient Reported Data
A conversation between Jen A. Levitt, MD, and Clinical Leader Executive Editor Abby Proch
Academic medical research has long been constrained by slow, resource‑intensive publication processes, particularly in rare diseases where patient recruitment and data access pose significant barriers.
In this interview, Jen A. Levitt, MD, Department of Dermatology at Emek Medical Center in Israel, talks about her team’s recent Hailey‑Hailey disease study, exploring how an AI‑native, patient‑centric RWD platform helped her team reduce research timelines by 76% and opened new possibilities for scientific discovery.
Clinical Leader: Dr. Levitt, your recent work highlights persistent challenges in academic publishing. Why does medical research, particularly retrospective research, take so long to reach publication?
Jen A. Levitt, MD: The core issue is structural inefficiency. Traditional retrospective research requires navigating multiple layers of bureaucracy before analysis can even begin. Obtaining Institutional review board (IRB) approval and securing access to large healthcare databases can take months. After that, researchers must clean and normalize data and then outsource statistical analysis because many institutions lack dedicated biostatistical support.
In practice, this means that intellectual work - hypothesis generation, critical thinking, and interpretation - gets overshadowed by administrative and tactical tasks. It is not uncommon for retrospective studies to take one to two years from concept to manuscript submission.
Your Hailey‑Hailey disease study seems to have broken that pattern. What made this project different?
Hailey‑Hailey disease (HHD) is a rare genetic blistering disorder, and historically, research has been limited to case reports or small case series. Even the largest registries include only a few hundred patients.
Our study used a different starting point. We sourced data already IRB approved, collected and normalized by the StuffThatWorks platform, which eliminated the most time-consuming phases of conventional retrospective research. We could focus directly on the clinical questions: what do patients actually experience, what triggers their disease, and what helps them. For a rare condition like HHD, where assembling even a modest cohort through conventional means takes years, this approach made a study of this scale genuinely feasible.
Can you elaborate on how patient‑reported data enhanced your understanding of Hailey‑Hailey disease?
Patient-reported data may capture aspects of the disease that physician-assessed outcomes sometimes miss. Because responses included open-ended questions, patients described not just their symptoms but how the disease affected their daily functioning, their work, and their psychological well-being in ways that go beyond what a structured clinical assessment would elicit.
For example, we found that nearly 30% had completely stopped physical activity due to their condition - something that would rarely be documented in a clinical encounter. Dietary triggers also emerged as a meaningful pattern, not part of any standard HHD assessment but clearly present in patients’ responses. Perhaps most unexpectedly, a novel safety signal emerged that we had not anticipated at study outset, and that warrants prospective evaluation.
Your team compressed a traditionally two‑year process into just 24 weeks. How was that possible?
Compressing the process was possible because the platform removed many of the steps that normally slow retrospective research. Literature review and hypothesis generation, which usually require about three months, were completed in roughly two to three weeks because AI‑driven exploration accelerated the process. Study design, which traditionally takes about a month, was shortened to one to two weeks because the data structure was already known. The IRB approval and data‑extraction phase- typically three to six months- was eliminated entirely because the environment was already pre‑approved. Data collection, cleaning, and normalization, which normally add another three to six months, were also removed from the timeline because the dataset was already pre‑collected and pre‑normalized.
Statistical analysis, which often takes two to three months, was completed in two to three weeks, carried out by the platform’s built‑in tools. Preparing raw tables and figures, a process that usually requires one to two months, was reduced to two to three weeks through automated generation. The platform includes additional capabilities we did not use in this study, such as AI-powered pattern recognition and automated table and graph generation, which could compress future timelines even further.
Taken together, these changes allowed the team to move from a traditional 12-24‑month timeline to a 24‑week timeline, demonstrating how an AI‑enabled, research‑ready environment can fundamentally streamline retrospective clinical research.
What broader implications does this model have for rare disease research?
Rare diseases present a fundamental research paradox: The populations most in need of evidence are the hardest to study by conventional means. What this model offers is a way around that paradox- accessing patients where they already are, in communities they have built around their shared experience. For HHD, we assembled the largest patient-reported cohort to date without a single clinic visit or database access request.
The limitations must be acknowledged honestly: unverified diagnoses, unvalidated instruments, and inherent selection, recall, and reporting biases mean findings should be interpreted as hypothesis-generating rather than definitive. This approach does not replace traditional prospective research; it identifies the questions worth asking in a prospective trial. That is a meaningful contribution for a disease where the evidence base has long been scarce.
Beyond rare diseases, what impact could this approach have on academic medicine more broadly?
Academic medicine stands to benefit from this new capability in research methodology. Clinicians can bring the patient’s own perspective directly into the scientific record, moving beyond the traditional clinician‑centric lens that has shaped medical literature for decades. This approach also aligns with the growing emphasis on real‑world evidence, which journals and regulators increasingly view as essential for understanding how diseases and treatments play out in everyday life. By dramatically shortening the research cycle, it enables faster translation of insights into clinical practice, allowing clinicians to act on new knowledge much sooner. Importantly, it also democratizes research, giving clinicians at smaller or resource‑limited institutions access to tools and datasets that once existed only within major academic centers. And with these efficiencies, the model has the potential to accelerate academic careers, making it feasible for researchers who previously published one paper a year to produce several meaningful contributions within the same timeframe.
Finally, why is this shift so important now?
Because the tools exist now in a way they simply did not before. And while rare diseases are perhaps the most obvious beneficiary, the implications extend beyond them. The ability to detect patterns and flag signals automatically across large patient databases, regardless of disease, represents something genuinely new. A safety signal, a subgroup that responds differently, a trigger no clinician had thought to ask about — these can now surface from the data rather than waiting to be hypothesized. We are at the very beginning of what this means for medicine.
About The Expert:
Jen A. Levitt, MD, is a dermatology resident in the Department of Dermatology at Emek Medical Center, Israel. She holds an MD from the Sackler School of Medicine, Tel Aviv University, and completed a research-based MA in psychobiology at Tel Aviv University, as part of her studies in the Adi Lautman Interdisciplinary Program for Outstanding Students.
Dr. Levitt's research spans clinical and epidemiological dermatology, with particular focus on hidradenitis suppurativa. She has authored numerous peer-reviewed publications in journals including Acta Dermato-Venereologica and Clinical and Experimental Dermatology and has presented her work at national and international conferences, including the European Academy of Dermatology and Venereology (EADV). She is also a peer reviewer for the Journal of the American Academy of Dermatology (JAAD), contributing to the broader scientific review process in her field.