No, NIFS Does Not Have '97%+ Accuracy for All Clinically Relevant Variants' — The Real Picture Is Far More Complicated
“The NIFS test has over 97% accuracy for clinically relevant variants”
The argument in brief
The claim that the NIFS prenatal screening test has over 97% accuracy for clinically relevant variants is unverifiable and misleading. While accuracy for common conditions like Down syndrome (trisomy 21) can exceed 99%, for rarer variants like microdeletions, the chance a positive result is correct can fall below 50%. No peer-reviewed study or regulatory body supports a single blanket accuracy figure across all clinically relevant variants.
Why it spread
Pregnancy is an anxious time, and a single high accuracy number offers real comfort. It is also easy for that number to travel: a clinician mentions the 99% figure for Down syndrome, a patient hears '99% accurate,' and it gets repeated as applying to everything the test screens for. Test manufacturers have commercial incentives to lead with their strongest performance data, and most people — including healthcare providers — do not have time to read the fine print on which specific variants those numbers actually cover.
The claim is that the NIFS test — a type of non-invasive prenatal screening using cell-free DNA from a blood sample — achieves over 97% accuracy for clinically relevant variants. This is not supported by the evidence. Accuracy varies so dramatically by condition that a single number is essentially meaningless, and in some cases, dangerously misleading.
For the most common chromosomal conditions, the test genuinely performs well. Illumina's clinical validation data and a 2015 meta-analysis by Benn et al. in Prenatal Diagnosis both show sensitivity above 99% for trisomy 21 (Down syndrome), trisomy 18, and trisomy 13. If the claim were limited to these conditions, it would be roughly defensible. But that is not what 'clinically relevant variants' means in practice.
The problem is what happens beyond those common trisomies. A 2019 study by Zhang et al. in PLOS ONE found that for microdeletion syndromes — a category routinely included in expanded NIFS panels — positive predictive values range from just 3% to 48%. That means for some conditions, fewer than 1 in 20 positive results are real. A joint statement from SMFM and ACMG (Gregg et al., Genetics in Medicine, 2016) confirmed that NIPT performance for microdeletions and copy number variants is simply not well-established. ACOG's Practice Bulletin No. 226 explicitly warns that applying a blanket accuracy figure across all conditions is misleading.
The FDA adds another layer of caution: NIFS-style tests are not approved as diagnostic tools, and no regulatory document endorses a universal 97% accuracy standard across all variants. A positive result still requires confirmatory diagnostic testing before any clinical decision should be made.
This claim spreads because it takes a real, impressive number — the high accuracy for trisomy 21 — and quietly expands it to cover the whole test. Test marketers and even well-meaning clinicians can inadvertently generalize the best-case figure to the full panel. Expectant parents, understandably anxious, are also drawn to a single reassuring number over a nuanced probability breakdown. Watch out for any prenatal screening claim that gives one accuracy figure without specifying exactly which conditions it applies to — that is almost always a sign the number is being stretched beyond what the evidence supports.
Sources
- Illumina NIPT/NIPS Clinical Validation Studies
Illumina's VeriSeq NIPT solution reports sensitivity and specificity above 99% for common trisomies (13, 18, 21), but performance metrics vary significantly by condition and fetal fraction, and 'clinically relevant variants' is a broad, undefined category.
- ACOG Practice Bulletin on Cell-Free DNA Screening (No. 226)
ACOG notes that cfDNA screening performance varies by condition: sensitivity for trisomy 21 exceeds 99%, but for sex chromosome aneuploidies and microdeletions, false positive and false negative rates are substantially higher, making a blanket '97% accuracy' claim misleading.
- Gregg et al., Genetics in Medicine (2016) - SMFM/ACMG Statement
A joint statement from SMFM and ACMG found that NIPT performance for microdeletions and copy number variants is not well-established, with positive predictive values sometimes below 50% depending on the variant and population prevalence.
- Benn et al., Prenatal Diagnosis (2015) - NIPT Performance Review
Meta-analyses of NIPT studies show high sensitivity (>99%) for trisomy 21 but considerably lower performance for rarer chromosomal conditions, meaning aggregate 'accuracy' figures depend heavily on which variants are included in the claim.
- FDA 510(k) Clearance Documents for NIPT Devices
The FDA has noted that NIPT tests are not approved as diagnostic tests and that accuracy claims vary by manufacturer and specific analyte; no universal '97% accuracy for clinically relevant variants' standard exists in regulatory documentation.
- Zhang et al., PLOS ONE (2019) - NIPT Accuracy for Microdeletions
Studies on NIPT for microdeletion syndromes report positive predictive values ranging from 3% to 48% depending on the syndrome, far below a 97% accuracy threshold for this subset of 'clinically relevant variants.'
Related debunks
- UnverifiableCan't Confirm or Deny: Bill Ritter's Claim About His Cancer Treatment Is Unverifiable — Not False
- UnverifiableNo Evidence Bill Ritter Has Been Diagnosed With Alzheimer's — The Claim Is Unverified
- UnverifiableYes, Whey, Soy, and Pea Proteins Really Can Leave a Bad Aftertaste — And Science Explains Why