Characterising 'other Developmental Concerns (ODC)' at 36 Months in Infants at Familial High-Risk for Autism Spectrum Disorder: A Bsrc Study

Background:  Approximately 20% of HR infant siblings are diagnosed with ASD at 36 months of age and a further 20% show other sub-optimal developmental profiles (Messinger et al., 2013). However, the pattern of these ODC and whether they are associated with early development or sex has not been determined.

Objectives:  N/A

Methods:  1,931 infant siblings (1,330 HR; 601 LR) from 15 different BSRC sites were recruited. The N=330 siblings with an ASD diagnosis at 36m were excluded (321 HR, 9 LR) leaving 1,601 siblings (1,009 HR, 592 LR). 36 month outcomes were defined as: Broader Autism Phenotype (BAP): ADOS severity score > 4 regardless of Mullen scores; Subthreshold ASD (Sub): ADOS severity score = 3 no Mullen subtest t score < 30 and at most 1 Mullen subtest t score < 35; Developmental Delay (DD): ADOS severity score < 3 combined with 2 or more Mullen subtest t scores < 35 or 1 or more subtest t scores < 30; or Typical Development (TD).

Results:

Multinomial logistic regression was used to test whether risk status (HR vs. LR) and sex (male vs. female) was associated with ODC outcome group (Table 1). Compared to TD children those categorised as BAP were more likely to be HR (relative risk ratio (RRR): 1.84 (95% CIs: 1.33 to 2.53), p<.001) and male (RRR: 1.52 (1.13 to 2.04), p<.01). Those who were categorised as Sub were more likely to be HR (RRR: 2.52 (1.51 to 4.18), p<.001) and those who were DD were more likely to be HR (RRR: 3.03 (1.55 to 5.90), p=.001) and male (RRR: 3.43 (1.86 to 6.32), p<.001).

In order to examine whether developmental abilities at earlier timepoints (6, 12, 18, 24 months) was associated with ODC outcomes the Mullen ELC was added to the model (Figure 1). Lower ELC at 6m was marginally associated with an increased likelihood of being in the BAP group (RRR: 0.99 (0.97 to 1.00), p=.06). Lower ELC at 12m was significantly associated with an increased likelihood of being in the DD group (RRR: 0.97 (0.95 to 0.99), p<.01). Lower ELC at 18m was associated with an increased likelihood of being in the BAP group (RRR: 0.98 (0.96 to 1.00, p<.05) and the DD outcome category (RRR: 0.96 (0.93 to 0.99), p<.01). Lower 24m ELC was associated with being in the DD group (RRR: 0.93 (0.91 to 0.95), p<.001).

Conclusions:  One quarter of HR siblings fell into one of the ODC groups. Each of these outcomes was also seen in the LR siblings, albeit at around half the rate. Boys were more likely than girls to be in these ODC groups. Earlier developmental abilities only weakly predicted outcomes, with the exception of DD where Mullen scores were low from 6 months onwards. Further analysis will explore the clinical utility of these outcome groupings to help establish if they should be used when reporting HR infant sibling studies.