Quality of Evidence for Prenatal Down Syndrome Screening: An Umbrella Review

Many systematic reviews (SRs) report the accuracy of prenatal Down syndrome screening, but their methodological quality has not been comprehensively evaluated. This umbrella review assessed the quality, compared non-invasive prenatal testing (NIPT) with traditional screening, and evaluated the findings' robustness.

We searched MEDLINE, Embase, and multiple other databases from inception to March 24, 2026, for SRs, including gray literature sources as specified in the methods. Methodological quality, risk of bias, and evidence certainty were assessed using AMSTAR 2, ROBIS, and GRADE. The study was registered with PROSPERO (CRD42024597291).

Of 42 included SRs, NIPT and combined screening were the most common. AMSTAR 2 ratings indicated that 88.1% (37/42) of SRs were of low or critically low confidence; 66.7% SRs were judged to have a high risk of bias; and high-quality evidence was scarce (9.6%). Despite this, NIPT's higher accuracy was maintained in sensitivity analyses, showing higher sensitivity (93.0%-100.0%) and specificity (96.0%-100.0%) than traditional methods, which showed wider performance variation (e.g., biochemical screening sensitivity: 28.0%-89.0%).

The methodological quality of SRs evaluating Down syndrome screening is limited. Nevertheless, NIPT consistently shows higher accuracy than traditional screening. NIPT implementation should be prioritized where feasible, while future primary research requires large, prospective studies.