In 2019, someone asked 178 of North America's Parkinson's specialists a simple question: how many of your patients did you send for genetic testing last year? Forty-one percent answered none. Eighty-seven percent had ordered ten or fewer. The reasons they gave were not mysterious or philosophical. Insurance would not cover it, and the result would not change what they did on Monday morning. That survey, led by Roy Alcalay and colleagues and published online in Genetics in Medicine in November 2019, became the benchmark against which everything that happened next would be measured.
What happened next was that the testing got free. In August 2019 the Parkinson's Foundation launched PD GENEration (ClinicalTrials.gov NCT04057794 and NCT04994015), which offers people with Parkinson's a panel of the seven most commonly implicated genes through a CLIA-certified laboratory at no cost and with no reliance on insurance, and returns the results through a neurologist or genetic counselor. In 2024 it expanded again, adding whole genome sequencing and a further 21-gene panel to the protocol. The preprint of the new survey reports that more than 25,000 people with Parkinson's had enrolled in the study's first five years. So a team led by Jesse Wang and Anne-Marie Wills at Mass General Brigham went back and asked the same questions again. Between October 15, 2024 and January 6, 2025, 621 clinician members of the Parkinson Study Group were emailed three times; 119 responded and 113 finished. Their answers were published in Genetics in Medicine Open on March 10, 2026.
The headline result is not that doctors started ordering more tests. It is that they started ordering a different kind. The proportion of clinicians sending more than ten insurance-billed clinical tests a year did not rise significantly: 12.3 percent in 2019, against 18.1 percent of respondents at PD GENEration sites and 26.7 percent of those at unaffiliated sites in 2024. But the proportion making more than ten referrals a year to research testing that returns results to the participant, meaning PD GENEration, ROPAD or PPMI, went from 15.5 percent in 2019 to 23.3 percent at unaffiliated sites and 60.2 percent at PD GENEration recruiting and referral sites. Direct-to-consumer testing and DNA banking barely moved. Every perceived barrier the clinicians were asked to grade on a 0 to 100 scale came down significantly, and here is the part the authors themselves flag as a surprise: it came down just as much at sites with no connection to the program as at sites running it. Knowledge shifted too. Asked about a case scenario, 52.4 percent correctly identified GBA1-associated Parkinson's in 2019, against 76.8 percent at PD GENEration sites and 63.3 percent elsewhere in 2024.
There is a wrinkle that families will recognise faster than anyone. In 2024, 16.8 percent of clinicians reported more than ten patients declining a genetic test in the past year, compared with 3.8 percent in 2019. Offering a test to everyone means hearing no from more people, and a no is not a failure; it is a choice about information someone may not want. The two oldest objections also survived the whole exercise. Cost and lack of insurance coverage remained the highest-rated barrier, and so did the belief that the result would not change the clinician's approach to care. Free testing appears to have loosened the first without settling the second, even as the number of precision medicine trials the authors counted for genetically defined Parkinson's rose from one in the United States in 2019 to five in 2024.
For a family sitting in a clinic room, the practical meaning is narrow and real. Specialists at these sites now report caring for many more patients whose genetic status they actually know: those seeing more than ten patients with a known LRRK2 variant rose from 5.5 percent to 12.6 percent, and for GBA1 from 4.9 percent to 18.4 percent, with reported GBA1 heterozygotes climbing from 402 to 674. Roughly six in ten respondents, 59.3 percent, said they would involve a genetic counselor or recommend a referral before testing, essentially unchanged from 61.2 percent in 2019, which means the counseling workforce question has not been solved by making the test cheaper. Whether testing is right for any one person, what a variant would mean for their children, and whether it opens a trial door are conversations for their own neurologist and a genetic counselor, not for a survey.