Numbers, Robots, and other Critical Truths About Rare Disease

by Daniel Zaksas, PhD

The narrative of how rare diseases fit in the general healthcare landscape often goes something like this: by definition, rare diseases are highly infrequent. Despite this, physicians should do a better job recognizing them. The costs of new treatments, while high, are merely a drop in the bucket for the system. This narrative is logical, but it is increasingly misleading and needs to be corrected in a few important ways.

It’s Time to Stop Thinking of Rare as Rare
When we as a society and as an industry demand attention for a rare disease, it is often because the experience of patients and their families is akin to wandering a desert. Bewildered and exasperated in their isolation, they suffer without answers, without resources, and too often without hope. And so we draw upon a collective empathy to help those who are other, who are diagnostically orphaned, who are medical minorities.

This presents a significant conundrum. The medical community’s Hippocratic responsibility is appropriately consistent with Mr. Spock’s famous paraphrase of utilitarianism: “The needs of the many outweigh the needs of the few.” We have to achieve the greatest possible good with limited resources and time. But of course this principle is completely at odds with the motivation to invest in diagnostic and therapeutic solutions needed by the few. As we often see and hear in market research, physicians struggle mightily with this notion. Their sympathies may be with the few, but their practice habits serve the many.

So how do we escape this utilitarian dilemma and convince the medical community that a focus on rare is not only honorable, but also undeniably deserving of dramatic investment? First and foremost, we need a paradigm shift in the way rare diseases are considered. Stop focusing on the specific infrequency of a single diagnosis, and start thinking of the entirety of the rare disease space as a coalition of sorts. The truth is that rare diseases are incredibly common!

In the United States alone, the number of people living with a rare disease may be at least 11.4–19.3 million.1 Even this conservative estimate comprises substantially more people than Alzheimer’s and Parkinson’s diseases combined. The high end of the estimate even exceeds the number of people who suffer from depression. In this light, the need to address the vast, underserved rare disease community becomes immediately clear.

It’s Time for Doctors to Be Replaced by Robots
Considering rare diseases as a bloc may be simple, but it entirely belies the complexity of diagnosing and treating each one of the more than 6000 that have been cataloged by OrphaNet.1 To accomplish this, it’s high time we started to replace doctors with robots.

Ok, I exaggerate. But let’s be honest for a second about what our industry is increasingly asking of human physicians. Every time a new pharmacotherapy is developed for a rare disease, we rush to heighten HCP awareness. We remind them of a few facts about the disease that they may or may not have learned in medical school, and we arm them with lists of red flags, diagnostic algorithms, and treatment considerations, often on the odd chance that once or twice in their career they might encounter the appropriate patient. Don’t get me wrong, this is by far the best we can do to help patients and advance clinical practice.

Still, a casual perusal of ongoing studies tells us that over the next decade we are likely to start losing count of the rare diseases for which diagnostic and therapeutic options are available. Do we honestly expect that physicians and nurses, all their remarkable skills and knowledge notwithstanding, can retain the information we will be delivering to them? In the coming years they will need lots of help connecting dots, fishing needles out of haystacks, spotting zebras, or doing whatever other cliché connotes the difficulty of rare diagnoses.

Luckily, this type of work is a perfect opportunity for artificial intelligence (AI) to shine. Machine learning algorithms have already shown that AI can at least match humans in the analysis of medical imaging, and even learn to achieve incredible diagnostic accuracy in autism.2,3 In the near future of rare disease, AI will be at the forefront out of necessity and it will quickly earn its keep. Once we show the advanced tech its place in cutting meanders from the patient journey, we can let physicians refocus their energy where it’s needed most—compassionate, quality patient care.

It’s Time for a Reckoning in Our Healthcare System
Having reconciled that rare diseases are actually common, and that the rates of diagnosis and treatments are likely to skyrocket, we come to the third and most difficult challenge: who’s going to pay for it all? At the moment, commercial viability of rare disease products requires high prices relative to mass market therapies. The math is simple; the solution, far from it. Yes, technology will play a major role in reducing costs through efficiencies in drug discovery and manufacturing, as well as in genetic screening, diagnosis, and therapeutic compliance.

But this will not be enough. Ultimately, we as a broad society will need to acknowledge that our national priorities must include our own health and wellbeing, and that the only humane and ethical choice is to facilitate a transparent but equitable system of healthcare that can fairly help, among others, the millions who suffer from increasingly treatable rare diseases.


  1. Nguengang Wakap S, Lambert DM, Olry A, et al. Estimating cumulative point prevalence of rare diseases: analysis of the Orphanet database. Eur J Hum Genet. 2020;28(2):165-173.
  2. Liu X, Faes L, Kale AU, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digital Health. 2019;1(6):E271-E297.
  3. Cognoa. Available at: Accessed June 9, 2020.

Planning your next project?

Work with us

Download our fact sheet

Click here