The Value of Implementing the First Year Inventory - Lite for Screening in a Healthcare

Background: Healthcare administrators are debating whether or not to mandate universal infant autism spectrum disorders (ASD) screening. The additive value of implementing a specific ASD screening tool above known risk factors such as preterm birth and parental concerns is unclear. The variance in typical social-communication development requires careful examination of salient predictive markers in low risk infants.

Objectives: To investigate the predictive validity of the First Year Inventory - Lite (brief version of the FYI) for identifying 12-month-old infants with poor social-communication development at 13- and/or 24- months.

Methods: Parents of 583 infants 12 months of age attending 16 well baby clinics (WBCs) in Israel completed the FYI-L. Eight percent had preterm births and for 2% parents reported social-communication concerns at 12 months. This questionnaire contained 24 items from the FYI and its risk status is norm-based. Ten infants who failed the FYI-L and a matched subset of infants who passed the FYI-L (n=12) were evaluated using the Autism Observation Scale for Infants (AOSI) and the Mullen Scales of Early Development (MSEL). Based on this evaluation six infants were referred for an evaluation in the tertiary care system. WBC records of 153 infants were reviewed for social-communication development problems. Records were reviewed only for infants with non-anonymous FYI-L and with WBC information ≥ 24 months.

Results: Relative to research referral for evaluation at 13 months, FYI -L risk status showed a positive predictive value (PPV) of 60% and a negative predictive value (NPV) of 100%. Relative to the presence of social-communication problems at 24 months on medical records, PPV was 60% and NPV 95%. A stepwise logistic regression to predict social-communication problems on WBC records from preterm status, parental social-communication concerns, and FYI-L sensory and social scores resulted in a final significant model including preterm status and FYI-L social-communication score (p=.004). A one unit increase in FYI-L social-communication score increased the odds of having problems on records by nine times. There were five FYI-L items that significantly (p< .05) differentiated infants who had social-communication problems on medical records versus not: imitation of sounds, imitation of body movements, following pointing, gaze avoidance, and prolonged staring at lights.

Conclusions:  Implementing a brief specific ASD screener contributes to the identification of infants who later on have social-communication problems. Preterm status and FYI-L social-communication markers were highly associated with social-communication problems on records during the third year of life. Parental concerns may play a stronger role at an older age. The false positive rate of the FYI-L should be weighed against the importance of early diagnosis and intervention. Clinicians may want to pay closer attention at 12 months to the highly differentiating markers identified in this study.