Erratum to: Sex differences in the association between infant markers and later autistic traits.

Andrew Pickles,Tony Charman,Rachael Bedford,Mark H. Johnson,Emily J. H. Jones,Teodora Gliga

doi:10.1186/s13229-016-0094-8

Andrew Pickles, Tony Charman

Open Access

https://doi.org/10.1186/s13229-016-0094-8

Copy DOI

Abstract

Erratum After publication of the article [1], the authors noted that an error was observed in the coding of one variable in the analysis. Three children from the low-risk control group were coded as girls in the published version but were actually boys. After re-running all the statistical analyses, the effects we reported remain substantially similar, and therefore the conclusions of the manuscript and our summary in the abstract remain unchanged. Detailed corrected analyses are presented below.

Highlights

Erratum After publication of the article [1], the authors noted that an error was observed in the coding of one variable in the analysis
A linear regression showed a significant relationship between Autism Observational Scale for Infants (AOSI) and ADOS score (β = 0.54, p < 0.001) and a significant sex*AOSI interaction (β = -0.47, p = 0.003)
When this was broken down by sex, AOSI was a significant predictor of ADOS in males (β = 0.58, p < 0.001) but Bedford et al Molecular Autism (2016) 7:33

Summary

Introduction

Erratum After publication of the article [1], the authors noted that an error was observed in the coding of one variable in the analysis. Different numbers of infants contributed data to the three early autism markers, as such: Autism Observational Scale for Infants (AOSI); (see Gammer et al, 2015): 53 high-risk (21 male) and 48 low-risk (20 male) infants completed the AOSI assessment (see Table 1).

Results

Conclusion