Study designs of evaluations included in the review
Clinical or experimental cross-sectional studies evaluating the diagnostic characteristics of at least two tests were eligible for inclusion. Diagnostic case-control studies, case series, case reports, narrative reviews, editorials and letters were excluded from the review. The included studies ranged from 6 in sample size to a multicentre trial of 2,253.
Specific interventions included in the review
Studies examining the use of Southern blot, polymerase chain reaction (PCR) or cytogenetic techniques were eligible for inclusion. The included studies examined cytogenetic tests (with diagnostic thresholds of less than 4%, 3%, 4%, or greater than or equal to 4%) and PCR.
Reference standard test against which the new test was compared
No inclusion criteria relating to the reference standard were reported in the review. The included studies used PCR and/or Southern blot as the reference standards.
Participants included in the review
Studies examining humans or human tissues were eligible for inclusion; animal studies were excluded. The included studies investigated prenatal samples, family members considered at risk, individuals with clinical characteristics of fragile X, individuals known to have fragile X, individuals previously assessed for fragile X, or mentally retarded children and adults. The studies were conducted in France, Sweden, Brazil, Mexico, Russia, Australia and the United States.
Outcomes assessed in the review
Studies reporting any outcome that addressed the diagnostic characteristics of fragile X were eligible for inclusion. No trials comparing clinical outcomes in individuals subjected to diagnostic testing with untested individuals were identified. The outcome measures used in the review were the sensitivity, specificity, and positive and negative likelihood ratios.
How were decisions on the relevance of primary studies made?
Three independent reviewers decided whether to accept or reject studies. Any discrepancies were discussed and resolved by consensus.