Skip to main content
. 2017 Jan 31;8:3. doi: 10.1186/s13229-017-0118-z

Table 3.

Unadjusted regression models of androgen levels and 12- and 36-month outcomes within each sex

Androgensa Infant sex Outcomes
12-month AOSI 36-month SRS
Beta 95% CI P Beta 95% CI P
ln(T) Female (n = 62) 0.20 (−0.12,0.51) 0.22 −0.06 (−0.30,0.19) 0.64
Male (n = 75) 0.20 (−0.13,0.52) 0.23 0.05 (−0.25,0.36) 0.73
ln(A4) Female (n = 62) 0.06 (−0.36,0.48) 0.78 −0.18 (−0.50,0.14) 0.26
Male (n = 75) 0.06 (−0.33,0.45) 0.76 −0.21 (−0.57,0.15) 0.24
ln(DHEA) Female (n = 62) 0.09 (−0.19,0.37) 0.51 −0.05 (−0.25,0.16) 0.67
Male (n = 75) 0.04 (−0.28,0.34) 0.82 −0.01 (−0.31,0.29) 0.94

ln natural log transformed

aSeparate robust regression models of loge-transformed testosterone (T), androstenedione (A4), and dehydroepiandrostenedione (DHEA) with total loge (AOSI + 1) and loge (SRS raw) scores. Outcome measures are 12-month Autism Observation Scales for Infants (AOSI) total score and 36-month Social Responsiveness Scale (SRS) total raw score