White-Boarding The Future Of IVF

From the introduction to “Democratizing IVF: Using Science, Medicine, and Engineering (and Venture Capital) to Help Build Families”

Wanted: a parallel in vitro fertilization industry built alongside the existing one, but designed for much higher numbers, mechanized for quality control, and capable of aggregating, analyzing and acting upon the enormous amounts of data generated but not routinely collected in IVF.

One of my favorite venture companies begins its pitch by predicting that almost two hundred million in vitro fertilization (IVF) babies will have been born by the end of the twenty-first century.

Sound implausible?

I think so. The number is way too low.

Let’s mess around with the data a bit. Worldwide, there are 130 million babies born per year, which is about the size of the population of Mexico or Japan. The prevalence of infertility is usually pegged at 15%, but let’s err on the conservative side and assume it’s 10%, which means that there are 13 million people worldwide who are trying unsuccessfully to start or expand a family. Let’s be even more conservative in our model and assume that three million of the ten ultimately conceive on their own. That leaves ten million worldwide medically unable to start or expand a family.

Assuming zero population growth for the rest of the century, over the next eighty years there will be 800 million desired pregnancies that will not occur without intervention.

Sound high? Judged by the number of people we currently treat, definitely. Somewhere around 600,000 babies worldwide are born every year from IVF. If we continue to do exactly that for the rest of the century we’ll have 48 million more IVF babies by 2099. But cycle volumes increase by 3–8% per year (according to CDC data, volumes increased 7.6% from 2017 to 2018, the most recent numbers.) If we slow worldwide growth in IVF to 3% every year for the rest of the century, and assume that the procedure stops its forty year history of innovation and improved productivity, we would help families have 193,000,000 children between now and 2099. Adding in the nine million IVF people from 1979 to now brings us to over 200 million by the end of the century, very close to the number we cited at the beginning.

And if the growth is higher than that?

If we change the rate of growth to 5% per year, we arrive at over 580 million babies between now and 2099. That number seems really high, but still a ways to go before we reach the 800 million that our conservative 10% infertility population prevalence model predicted.

And that 580 million starts with a base limited only to heterosexual couples with infertility.

It ignores the underserved LGBTQ population, an incremental 5–20%.

It ignores the largely unserved population who, with proactive egg freezing and delayed IVF can preserve their fertility before sterilizing cancer treatment. There are 45,000 newly diagnosed cancers per year in the 0–28 age group.

It does not include the up to 6% of fertile couples who may choose to eliminate the risk of inherited disease in their children by using IVF with preimplantation genetic testing.

And it does not account for the large and growing (20+% yearly) population of women who choose egg freezing to mitigate the unpredictable risk of future loss of ovarian reserve.

At the 2019 American Society for Reproductive Medicine meeting, I modeled a United States IVF market that included these populations, which in total increased the addressable market by over 30%. Adding in these factors brings the number of babies needing IVF in our population prevalence model to over 1.2 billion for the rest of the century.

So maybe the 5% growth number isn’t all that unrealistic.

Regardless of the size of the error bars that we use to bracket our estimates, even the most conservative numbers suggest that our current IVF infrastructure will be overwhelmed without some serious system-wide preparation. Our current capacity and volumes resulted in two major laboratory failures in 2018. Like any engineered system, accelerations in throughput expose vulnerabilities in design; unsustainable volume is sometimes only obvious after the system has broken down. This concern is particularly acute in 2021. Much of the IVF world shut down for a long stretch of 2020 due to Covid-19 issues. Those patients did not disappear, and clinics will likely run at higher operating capacity to accommodate them on top of the normal cycle volume.

Longer term, the challenge of building an IVF infrastructure capable of consistently high pregnancy rates while ensuring patient and specimen (sperm, egg and embryo) safety transcends the speeding up the assembly line of patient care clinic to clinic. To accommodate an industry with 3x or 5x the current volumes, we need to account not only for specialized/premium labor and facilities constraints, but also much higher throughput data-management systems, supply chains and work flows, and, importantly, quality control and reporting surrounding patient and specimen safety. IVF in 2021 is US long distance trucking in 1956. We need to stop widening the existing local roads and start building an Interstate system alongside them. 65 MPH meant new design and different engineering: on and off ramps, three to six lanes and no stop signs or traffic lights — uninterrupted high-speed movement for hours at a time, designed to accommodate future traffic needs, far greater than the number of cars and trucks in the fifties.

Focusing again on IVF, millions of additional cycles per year means…what?

At this point I pause to take a question. The border of one of the zoom rectangles turns yellow.

“Do you think the world really cares enough about this to do all this work?”

Sigh. Time to put the numbers aside and tell stories. Not my core competence, but here goes.

There are very few things that I know that most of my venture capital, private equity, buy side and sell side colleagues don’t. But unless they have experienced infertility first-hand, they underestimate the enormous vacuum that it creates in the center of the lives of those suffering from it. It is a cruel affliction, diminishing life’s joys and canceling out other achievements and triumphs. These defeated feelings don’t go away just because the solution to the problem is too costly or too far away or logistically impossible due to the rest of life taking up most of the hours in the week. The void created by the inability to start or grow a family resides on the lowest portion of Abraham Maslow’s hierarchy, right alongside hunger and thirst and physical pain and extreme fatigue. I treated IVF patients for fifteen years, and, believe me, there ain’t nothing elective about this area of medicine.

There is an ongoing but pointless debate about whether or not infertility is a “disease.” One side points to the very real pain and suffering that it creates, the randomness with which it strikes and the collateral harm that it causes to mental well-being, productivity and strained relationships, as well as the presence, in some degree or another, of organ-level pathology, like pelvic scarring, hormonal disruptions that interfere with ovulation or disorders of sperm production. The other side argues that we are bio-engineered to get pregnant easily in our late teens and early twenties, and expecting the reproductive system to work at young adult efficiency levels a decade or two later is unrealistic. Besides, no one dies from infertility. Disease or not, it is not progressive, or at least progressively worse. We all end up infertile eventually.

I stopped engaging on this debate years ago. Semantic gymnastics aside, infertility is hurtful and harmful and left alone reduces life to the pursuit of mitigated unhappiness. In the United States alone, we have over seven million people affected by this. And our one really good treatment is only producing 80,000 or so good outcomes, just a little more than 1%. Now you can look at these numbers as a business opportunity, a healthcare crisis, a policy challenge, a moral failure, or a lack of imagination on the part of the insurance industry. Regardless of how you choose to describe, label, or otherwise characterize the problem, pain and suffering, compounded by a second problem: that we have an effective solution that is out of reach for most of those who are suffering.

So two problems, infertility and IVF access. And the better IVF gets at solving infertility, the more intolerable the lack of access to IVF becomes.

Improving access means more than insurance coverage, although we need to help the insurance industry apply its risk management strategies to IVF. Simply put, we need to make it easy to plug reliable numbers into actuarial tables, that will then generate utilization predictions accurate enough to model coverage. A given risk, whether it is the risk of a needing IVF to conceive versus the costs of paying for it spread out over a given population (all reproductive aged people in the state of Washington, or all 22–45 year olds working at Microsoft, or everyone in the United States covered by state or federal health plans or the VA) or the risk of auto liability for 17 year olds, can be rationally spread out and its costs underwritten rationally given accurate enough data to assess risk and assign costs. In a way, insurance companies and casinos do the same thing (casinos have more precise mathematical models since there are finite outcomes from throwing dice, summing the total of three cards or a given five card poker hand) agnostic to what exactly they are insuring.

So, for example, how do we increase insurance coverage? Yes, we continue to lobby for mandates, and we make sure that the industry keeps looking at infertility and IVF, again and again, each time the procedure improves and it gets incrementally easier to predict and model. AND — we work on technology to make it easier to do that modeling, by expanding data collection well beyond the pregnancy outcomes reports that the CDC mandates. Lots more on that later.

Ultimately, the battle to improve IVF access — democratizing family building — will be fought on many fronts: public, private, academic, using tools developed by science, engineering, human resources, banking and government.

The IVF procedure is moving from an “each lab is its own artisanal kitchen” model towards standardization and process optimization. States are adding insurance mandates. The VA is covering infertility. One specialty benefits manager alone increased the number of people with IVF insurance in the US by 3 million over 3 years. Applying our 10% prevalence of infertility means 300,000 people, many of whom could not have afforded IVF before.

Given that we perform approximately 300,000 cycles a year total, mainly to those who can afford the procedure without insurance coverage, this represent a massive increase in not only the addressable population, but the population actively seeking treatment.

Together, each of these observations suggests that we are mobilizing large groups of unserved and underserved people, and inviting them into the IVF world. Do I think the world really cares enough about this to do all this work?

Yeah. The world’s doing the work already. The patients are coming. Are we going to be ready for them?