The much-anticipated advent of "precision medicine" promises tailored technologies that not only assist individuals but also have the potential to reduce healthcare costs. Recent research highlights how prenatal screening can achieve these dual objectives, while also emphasizing that precision medicine must be well-suited to patients to truly save money.
The study focuses on cfDNA screenings, a type of blood test that can detect conditions related to chromosomal variations, such as Down syndrome. For many pregnant women, cfDNA screenings can serve as an alternative to invasive procedures like amniocentesis or chorionic villus sampling (CVS), which carry a risk of miscarriage.
The research examined the optimal use of cfDNA tests and reached a notable conclusion. "We find that the highest value of cfDNA testing is for individuals at high risk, but not at an extraordinarily high risk," says Amy Finkelstein, an economist at MIT and co-author of a recently published paper detailing the study.
The paper, "Targeting Precision Medicine: Evidence from Prenatal Screening," appears in the Journal of Political Economy. Co-authors include Peter Conner, associate professor and senior consultant at Karolinska University Hospital in Sweden; Liran Einav, professor of economics at Stanford University; Finkelstein, the John and Jennie S. MacDonald Professor of Economics at MIT; and Petra Persson, assistant professor of economics at Stanford University.
"Precision medicine holds a lot of promise," says Persson. "We can do many new things and tailor healthcare treatments to patients, which is very promising. In this paper, we highlight that while all of this is true, personalizing medicine also incurs significant costs. As a society, we might want to consider how to use these technologies while keeping an eye on healthcare costs."
Assessing the Benefit for ‘Medium-Risk’ Patients
To conduct the study, the research team analyzed the introduction of cfDNA screening in Sweden from 2011 to 2019, using data from over 230,000 pregnancies. There were also regional differences in how cfDNA screenings were covered by Swedish healthcare for patients who had not yet committed to invasive testing. Some regions broadly covered cfDNA tests for all patients with a "moderate" or higher assessed risk, while others limited coverage to a subset of high-risk patients within this group. This variation provided a basis for the researchers’ analysis.
With the most generous coverage of cfDNA tests, the procedure was used by 86 percent of patients; with more targeted coverage, this figure dropped to about 33 percent. In both cases, the number of invasive tests, including amniocentesis, significantly decreased to about 5 percent. (While cfDNA screenings are highly informative, they are not as conclusive as invasive tests, leading some pregnant women to opt for follow-up procedures.)
Both approaches resulted in similar reductions in invasive testing rates. However, due to the cost of cfDNA tests, the economic implications differ significantly. The study estimates that introducing broad coverage of cfDNA tests would increase overall medical costs by about $250 per pregnancy. In contrast, introducing cfDNA with more targeted coverage results in a reduction of about $89 per patient.
Ultimately, the broader dynamic is clear. Pregnant women at the highest risk of having children with chromosomal disorders are likely to opt for invasive tests like amniocentesis. Those at virtually no risk may not undergo cfDNA testing at all. For an intermediate group, cfDNA tests have substantial medical value, eliminating the need for invasive testing. Reducing the group of patients undergoing cfDNA testing lowers the overall cost.
"People at very high risk often resort to the invasive test, which is definitive, whether or not they have had cfDNA screening," explains Finkelstein. "But for those at medium risk, cfDNA coverage leads to a significant increase in cfDNA testing, resulting in a substantial decrease in the rates of more risky and costly invasive tests."
How Precise?
The study’s findings raise a broader point. Precision medicine, in nearly all its forms, will increase medical spending. Therefore, it is crucial to develop precision in determining who receives it.
"The appeal of precision medicine is targeting those who need it, so we don’t perform costly and potentially unpleasant tests or treatments on people who don’t need them," says Finkelstein. "That sounds great, but it kicks the can down the road. We still need to determine who is a candidate for what type of precision medicine."
Thus, in medicine, rather than simply tackling the problem with technology, we might prefer to aim carefully where the evidence justifies it. Overall, this means that good precision medicine relies on sound policy analysis, not just good technology.
"Sometimes, when we think about the impact of medical technology, we simply ask whether the technology increases or decreases healthcare costs, or whether it improves patient health," observes Persson. "An important aspect of our work, I think, is that the answers are not just about technology. It’s about combining technology and policy, as policy will influence the impact of technology on healthcare and patient outcomes. We see this clearly in our study."
In this case, finding comparable outcomes for patients with more restricted cfDNA screenings suggests a way to target diagnostic procedures. In many possible medical situations, identifying the subset of people for whom a technology is most likely to provide new and actionable information seems a promising goal.
"The benefit is not just an inherent feature of the tests," explains Finkelstein. "With diagnostic technologies, the value of information is greater when you are neither clearly suitable nor unsuitable for the next treatment. It’s really the non-monotonic value of information that’s interesting."
The study was partially funded by the U.S. National Science Foundation.
