ASD Screening in Primary Care: 10 Years of the M-CHAT Program in Spain

Background: Great efforts have been put into developing methods for early identification of toddlers with autism spectrum disorder (ASD) (Zwaigenbaum et al 2015). Despite the recommendation from the American Academy of Pediatrics (AAP) to do population-based screening (first-level) for ASD (Johnson & Myers, 2007) applied specifically at least in two occasions to all children before two years of age, it has hardly been approached from the public health system (PHS) perspective. Moreover there is a need to validate the usage of standardized instruments as a cost-efficient strategy in the PHS.

Objectives: The main goal of this study is to evaluate the “M-CHAT and M-CHAT-R ASD Early Screening Program” in community settings, after 10 years ongoing in two regions of the North of Spain (Salamanca and Zamora), in terms of feasibility, reliability and costs with the purpose of extending the program at regional and national levels.

Methods: Parents of 18 months and/or 24 months aged children residing in the selected geographical area during the study period (October 2005-October 2015), were asked to fill the Spanish version of the M-CHAT (updated version M-CHAT/R since April 2014) at the outpatient health services (compulsory vaccination programme and well-child check-up programme respectively) by a professional working at any of the total 54 pediatric teams (nurses and pediatricians) of the PHS that received training and agreed to participate (every team in the area). Scripted phone follow-ups were performed for positive (failed) screens and, differential diagnosis was done following a standardized protocol, using Vineland, Merril-P and ADOS-G. A well stablished coordination with the ASD early intervention centres and the Hospital diagnosis units in the area served as surveillance for tracking false negatives of both provinces.

Results: A total of 17332 M-CHAT were administered considering both M-CHAT and M-CHAT/R Spanish versions. Out of the total, 1600 (9.2%) resulted as ASD high-risk, of which 156 (9.7%) were confirmed as positive with the follow-up interview. Out of these, 54 children received a diagnosis of any ASD and 88 other disorders of childhood onset according to DSM-IV . The 74% of the participant professionals felt the program was totally feasible and 22% felt it was viable but with reservations (n = 54). Lastly, false negatives surveillance system identified 8 ASD cases out of the total of screened children. These results show a sensitivity of 0.87, a specificity of 0.99 and a PPV of 0.38 for ASD cases in this screening program (PPV of 0.94 when considering other DD).

Conclusions: This study has been able to show for the first time in Spain, the feasibility of a long lasting ASD screening program within the PHS. The current findings suggest that training on social and communicative development and dissemination of ASD early signs among pediatric teams, besides the use of a standardized tool, is essential for progress in the early detection of these disorders. The feasibility of this program should be considered for future strategies regarding ASD early detection by Health Policy makers at international level.