Exercise: Nuchal translucency

This website is for students following the M.Sc. programme in Health Sciences at the University of York.

The following is the abstract of a paper:

Objective The aim of this study was to assess the repeatability of nuchal translucency measurement by well-trained sonographers either experienced or Inexperienced with the measurement.

Methods The nuchal translucency was measured three times by two out of nine operators in 161 pregnant women attending two teaching hospitals. Intra-operator and inter-operator variation was assessed by calculating the standard deviation (SD). Intraclass Correlation Coefficients (ICC) with a 95% lower confidence limit (95% CL) were calculated for each operator and for pairs of operators. Intra-and inter-operator kappa statistics were calculated after dichotomising the results into a nuchal translucency smaller or larger than 3 mm.

Results The SD of intra-operator difference was 0.35 mm. The intra-operator repeatability, showed ICCs varying from 0.83 to 0.95 for the experienced operators and from 0.47 to 0.83 for the inexperienced. The SD of inter-operator differences was 0.56 mm, 0.46 mm and 0.44 mm, based on one, two and three measurements, respectively. The inter-operator repeatability showed a variation in ICC from 0.74 to 0.95 in pairs of experienced sonographers to 0.51 in one pair inexperienced with the measurement. The kappa value expressing the intra and inter-operator repeatability as being > 3 mm or < 3 mm was 0.88 and 0.85, respectively.

Conclusion This study supports the finding that the nuchal translucency measurement is reproducible, as long as it is performed by sonographers well trained in the technique of this measurement.

(Source: Pajkrt E, Mol BWJ, Boer K, Drogtrop AP, Bossuyt PMM, Bilardo CM. Intra- and interoperator repeatability of the nuchal translucency measurement. Ultrasound in Obstetrics & Gynecology 2000; 15: 297-301.)

Questions about this abstract:

1. The standard deviation of intra-operator difference was 0.35 mm. What exactly is this standard deviation? How does it differ from the within-subject standard deviation?
   Check suggested answer.
2. The standard deviation of inter-operator differences was 0.56 mm, 0.46 mm, and 0.44 mm, based on using single measurements by each operator and the averages of two and of three measurements, respectively. Why did the standard deviations get smaller as the number of observations increased? The drops in standard deviation are not very large. Why is this?
   Check suggested answer.
3. Intra-operator ICCs varied from 0.83 to 0.95 for the experienced operators and from 0.47 to 0.83 for the inexperienced. Why might we expect more experienced operators to have higher ICCs?
   Check suggested answer.
4. The inter-operator repeatability, showed a variation in ICC from 0.74 to 0.95 in pairs of experienced sonographers to 0.51 in one pair inexperienced with the measurement. Why is the inter-operator ICC less than the intra-operator ICC, irrespective of the experience of the operators?
   Check suggested answer.

Back to Measurement in Health and Disease index.

To Martin Bland's home page.

This page maintained by Martin Bland.
Last updated: 23 February, 2007.

Back to top.