Crandi

Finding Patients With Rare Diseases Blue Matter Consulting

Leo Migdal
-
finding patients with rare diseases blue matter consulting

Part 1 of this series provided an overview of rare diseases and the special challenges they present to biopharmaceutical companies. In this installment, we explore one of the most basic—and most confounding—of those challenges: Finding patients. In the European Union, a disease is considered rare when it affects fewer than five in 10,000 people. In the United States, a disease is defined as rare when it affects fewer than seven in 10,000 people.1 Some diseases are far rarer. Huntington’s disease, for example, only affects 3 to 7 per 100,000 people of European ancestry (and is even less common in other populations).2 And, compared to some other ultra-rare diseases, Huntington’s disease could seem... As one would expect, rare diseases mean rare—and typically geographically dispersed—patients.

That makes them hard to identify, and this presents significant problems for three key stakeholder groups: The patients themselves, the physicians who treat them, and the pharmaceutical companies that are working to develop new... The “typical” rare disease patient faces a range of obstacles before ever getting a proper diagnosis. Often, primary care physicians have limited (or no) familiarity with the patient’s condition, and refer the patient to a specialist after being unable to determine the cause of the patient’s symptoms. Unfortunately, the patient’s journey usually does not end with the specialist. Depending on the disease, awareness among specialists can be similarly limited, and patients will often see multiple doctors on their journey to a diagnosis. It can take numerous doctors and multiple misdiagnoses and hence, a long period of time before the correct rare disease is identified.3

Rare Disease Consulting specialises in navigating the unique challenges presented by rare diseases for the purpose of driving innovation and of improving patient outcomes. By unlocking insights respecting the landscape of rare diseases, our company is aiming at bringing clarity to complexity by utilising cutting-edge strategies in health economics and real-world evidence. From uncovering hidden value to shaping policies, we are dedicated to empowering stakeholders and making meaningful impact on patient care. Our clients include Greek and International organisations working in the healthcare industry, ranging from commercial companies to academic institutions, professional organisations as well as patient advocacy groups. Rare Disease Consulting (RDC) has emerged from the professional experience accumulated by our senior team in the area of Health Informatics, Health Economics, and Real World Evidence, since the early 1990s. Forty years after the initial instatement of the US Orphan Drug Act, awareness of rare diseases, as well as the number of available treatments has increased significantly.

Despite this positive trend, however, over 90% of rare diseases still lack effective treatments, a situation exacerbated by the challenges drug developers face when trying to bring novel treatments to rare disease patients1. A significant challenge within the rare disease field is finding patients who could benefit from a specific treatment. While it might be assumed that the unmet needs of a rare disease would automatically drive demand for a novel therapeutic—whether for participation in clinical trial or for access to newly approved drugs—the reality... In fact, matching patients with treatments can become a literal search for a needle in a haystack for drug developers. The time to diagnosis is for rare diseases typically long, averaging 4.7 years2. Time to diagnosis can be even longer for many patients, with physicians often struggling to piece together the complex puzzle of signs and symptoms that indicate a rare disease.

This can result in low diagnostic accuracy, with around three-thirds of patients being misdiagnosed at least once2. Rare diseases exist on a spectrum, ranging from the easily detectable to the elusive (Figure 1). Rare diseases that are relatively easy to recognise, such as Heamophilia3, typically have high disease awareness and a well-understood pathophysiology that allows for clear symptom description. For monogenic disorders, family history and distinct phenotypic features can often aid in the recognition of the disease. In the best-case scenario, such rare diseases also have clear diagnostic tests, making them identifiable. These tests can be based on genetic markers, such as testing for CAG repeats in the HTT gene for Huntington’s disease4 or SMN1 gene deletion for spinal muscular atrophy5, or on laboratory biomarkers, such...

Accurately identifying patients in rare disease markets is critical, both for improving patient outcomes and achieving commercial success. On the clinical side, limited disease awareness often leads to delayed diagnoses or misdiagnoses, preventing patients from receiving timely, appropriate care. Commercially, the challenge lies in pinpointing small, dispersed patient populations and the healthcare providers who treat them. Without this visibility, it can be very challenging to get important therapies to the right patients. Enhancing patient identification capabilities not only accelerates the path to diagnosis and treatment but also supports more targeted and effective engagement strategies and resource allocation. Claims data remains a widely used source for identifying patients, particularly in rare disease markets.

However, several challenges can limit its effectiveness. These include incomplete data capture, inconsistent patient stability, and the underutilization of approved ICD-10 codes. This brief case study explores how to address the latter issue by using a predictive methodology that leverages the clinical characteristics of patients already diagnosed with the appropriate ICD-10 code. This can help identify individuals who may be misdiagnosed or remain undiagnosed, ultimately expanding visibility into the true patient population. Machine learning has become one of the most powerful tools for identifying undiagnosed or misdiagnosed patients in rare diseases – areas where traditional brute force approaches often fall short. These models excel at detecting complex patterns across large volumes of data.

In addition, they can process data with a high dimensionality of features (e.g., such as diagnosis and procedure codes and claims data in general, which includes a large number of discrete patient characteristics across... However, machine learning is not a one-size-fits-all solution. The choice of model depends heavily on the type and quality of available data. Models typically fall into one of three categories: In a recent project, a newly approved ICD-10 code for a rare disease enabled us to identify a subset of confirmed patients. However, due to limited physician adoption of the new code, the broader claims dataset contained a large volume of patients who were either misdiagnosed, undiagnosed, or unrelated to the disease – creating a mix...

Given this data structure, a semi-supervised learning approach was most appropriate and allowed us to train a model that could extend learnings from the known cohort to identify likely undiagnosed patients across the dataset. Common semi-supervised approaches for us to choose from included: While we have seen a significant increase in the number of drugs for rare diseases, accurate diagnosis rates of rare diseases continue to trail behind those of more prevalent diseases. The average time to diagnosis for a patient with a rare disease is 5 to 7 years. The delay is driven by: In response to this delay, companies have invested in disease awareness and diagnosis programs, including artificial intelligence (AI)-driven patient-finding initiatives.

While heralded as the solution to address the diagnosis lag, results from these programs have been mixed and are the source of several challenges, including: At 81qd, we take a unique approach by leveraging AI to identify specific physician practices where undiagnosed patients with the highest probability of having a rare condition are currently being managed for other conditions. Over the years, we have partnered with many companies to help them overcome the challenges with their patient-finding initiatives. To guide our clients forward, we offer a simple framework based on the popular icebreaker game: Be prepared to be surprised. Undiagnosed patients identified through AI-driven approaches will frequently not look like diagnosed patients.

If they did, and if they were that simple to find, why would we need AI? Because most marketers are guilty of confirmation bias, they often look for attributes that confirm their preexisting knowledge about the market or a disease. When they see something contrary to their experience, they naturally refute it. AI should be designed to identify patients that may not look like “typical patients.” However, the inherent heterogeneity of the patient journey experienced by those with rare diseases means there are no “typical patients.”... On 26 October 2021, Breakfast Club members participated in a “live consulting session”, facilitated by Blue Matter, that addressed a key topic: Patient Identification in Rare Diseases. During two hours of breakouts and discussions, members pooled their knowledge to examine this topic and answer difficult questions rare disease companies often face.

Initially, participants examined the drivers and barriers to patient identification. Next, they discussed potential strategies and tactics to overcome barriers. In this paper, we summarize the group’s insights. Dirk brings extensive industry experience to Blue Matter. His background spans 25+ years, with senior commercial roles in leading biopharmaceutical companies such as Shire, Roche / Genentech, and Amgen. His specialties include commercial strategy development, lifecycle management, and marketing, with a focus on rare diseases.

He has a degree in biochemistry, as well as a PhD in Molecular Immunology from the University of Basel. ©2025 Blue Matter Consulting. All Rights Reserved. Design by Hinge.

People Also Search

Part 1 Of This Series Provided An Overview Of Rare

Part 1 of this series provided an overview of rare diseases and the special challenges they present to biopharmaceutical companies. In this installment, we explore one of the most basic—and most confounding—of those challenges: Finding patients. In the European Union, a disease is considered rare when it affects fewer than five in 10,000 people. In the United States, a disease is defined as rare w...

That Makes Them Hard To Identify, And This Presents Significant

That makes them hard to identify, and this presents significant problems for three key stakeholder groups: The patients themselves, the physicians who treat them, and the pharmaceutical companies that are working to develop new... The “typical” rare disease patient faces a range of obstacles before ever getting a proper diagnosis. Often, primary care physicians have limited (or no) familiarity wit...

Rare Disease Consulting Specialises In Navigating The Unique Challenges Presented

Rare Disease Consulting specialises in navigating the unique challenges presented by rare diseases for the purpose of driving innovation and of improving patient outcomes. By unlocking insights respecting the landscape of rare diseases, our company is aiming at bringing clarity to complexity by utilising cutting-edge strategies in health economics and real-world evidence. From uncovering hidden va...

Despite This Positive Trend, However, Over 90% Of Rare Diseases

Despite this positive trend, however, over 90% of rare diseases still lack effective treatments, a situation exacerbated by the challenges drug developers face when trying to bring novel treatments to rare disease patients1. A significant challenge within the rare disease field is finding patients who could benefit from a specific treatment. While it might be assumed that the unmet needs of a rare...

This Can Result In Low Diagnostic Accuracy, With Around Three-thirds

This can result in low diagnostic accuracy, with around three-thirds of patients being misdiagnosed at least once2. Rare diseases exist on a spectrum, ranging from the easily detectable to the elusive (Figure 1). Rare diseases that are relatively easy to recognise, such as Heamophilia3, typically have high disease awareness and a well-understood pathophysiology that allows for clear symptom descri...