Patient-reported outcome measures applied in chronic diseases relevant to asthma remission: a scoping review.

Autism is one of the most common neurodevelopmental conditions in society today and represents a broad population of individuals that are heterogeneous across multiple scales, from genotype to phenotype [], across development [], and across important clinical outcomes (e.g., responses to intervention) [–]. A top priority on the path towards precision medicine is the development of early intervention approaches that facilitate developmental gains and positive outcomes aligned with multiple different stakeholder perspectives [–]. The push for earlier diagnosis and intervention is also paramount [, ] and predicated on the idea that there is higher probability of facilitating more positive outcomes because of the greater neural plasticity offered during the earliest periods of neurodevelopment []. The top priorities/questions for early intervention research are quite clear. We seek to better understand 1) , 2) , and 3) []? Since heterogeneity in response to early intervention is a high priority topic, it is important to further unpack the ‘’ question. We need to better understand and early intervention is facilitating differential individualized outcomes. For these important questions, we have to go beyond examination of on-average group-differences due to intervention and dissect what are the factors or characteristics present in children before intervention begins that would help us predict the subsequent developmental path during a particular intervention [, , ]. Prior reviews and meta-analyses of autism early intervention research provide an initial starting point on the ‘’ and ‘’ questions. Such work helps us get a better sense of how early intervention may or may not have effects at a group-level on children’s functional outcomes. However, despite extensive culls through the literature on early intervention, results can seem mixed (e.g. [, –]). For example, without filtering or controlling for study quality/bias, there is some evidence that various types of early intervention (e.g., developmental and naturalistic developmental behavioral interventions; NDBI) can be effective on-average in changing a variety of outcomes including language, intellectual and social communication abilities [, , , ]. However, a somewhat different picture emerges when evaluating a smaller handful of high-quality randomized controlled trials (RCT). Using this particular restriction, earlier meta-analyses showed that several types of early behavioral interventions may be limited in effectiveness in changing cognitive, language, or core autism symptom domains [, , ]. However, a more recent updated meta-analysis suggests that some intervention types like naturalistic developmental behavioral interventions (NDBI) (e.g., Early Start Denver Model, ESDM) can promote positive outcomes in cognitive, language, adaptive functioning, and core diagnostic characteristics of autism []. While meta-analytic inferences are very important, there may also be some limitations for some research priorities/questions. First, meta-analytic work enables tests of whether replicable non-zero group-level effects exist in the literature. This goal is tailored to answer the ‘’ and ‘’ questions quite well. Nevertheless, these goals may be complicated when heterogeneity in response to intervention exists (i.e. answering the ‘’ question). Although evidence from high-quality RCTs suggests that group-level effects vary considerably across the literature [, , , ], this does not preclude the fact that the interventions may still have important effects on specific types of individuals. Thus, a different analytic approach may be complementary for answering the ‘’ question. Approaches such as individual participant data meta-analysis (IPD-MA) or ‘mega-analysis’ [] may be complementary to meta-analysis by allowing for deeper insight than what may be possible from meta-analysis alone.