Generating Evidence to Streamline the Clinical Pathway in Autism Spectrum Disorder Using Simulation Models: Costeffectiveness Comparisons of Screening and Genetic Testing Strategies

摘要

The clinical pathway to diagnosis in autism spectrum disorder (ASD) is complex. The objective of this thesis was to estimate the health and monetary impact of changes in the provision of select ASD screening and diagnostic health services. First, a meta-analysis estimated that the Modified Checklist for Autism in Toddlers, a commonly used ASD screening tool, performed at a low-to-moderate accuracy among children with developmental delay (pooled sensitivity: 0.83, 95% credible interval [CrI] 0.75, 0.90; specificity: 0.51, 95% CrI: 0.41, 0.61; positive predictive values [PPV] in high-risk children: 0.55, 95% CrI: 0.45, 0.66; PPV in low-risk children: 0.07, 95% CrI: <0.01, 0.16) and its performance changed with patient characteristics. The second study evaluated the cost-effectiveness of universal or high-risk screening compared to standard care, surveillance monitoring, in ASD using discrete event simulation. Results demonstrated that universal screening would greatly burden the healthcare system by heightening demand for diagnostic services and increasing healthcare expenditure. High-risk screening, on the other hand, could be a cost-effective strategy, yielding incremental cost-effectiveness ratios (ICERs) of $1100-1900/child initiated treatment or diagnosed earlier. The last study compared the cost-effectiveness of genome (GS) or exome sequencing (ES) to chromosomal microarray (CMA) using a microsimulation model. The use of ES in children with syndromic features after a negative CMA could be cost-effective compared to CMA alone (ICERs $5800-6000/child with pathogenic variant). If CMA was to be replaced by sequencing, GS would be the more cost-effective option. Findings from this thesis indicate that strategic resource allocation is crucial in ASD. Given the network of health, psychosocial and educational services required by individuals with ASD, changes in one component can have a large impact on the wait time, resource use and expenditure on downstream services that can affect children without ASD and extend beyond the healthcare system.