What Drives Detection and Diagnosis of Autism Spectrum Disorder? Looking Under the Hood of a Multi-stage Screening Process in Early Intervention

Abstract

U.S. guidelines for detecting autism emphasize screening and also incorporate clinical judgment. However, most research focusses on the former. Among 1,654 children participating in a multistage screening protocol for autism, we used mixed methods to evaluate: (1) the effectiveness of a clinical decision rule that encouraged further assessment based not only on positive screening results, but also on parent or provider concern, and (2) the influence of shared decision-making on screening administration. Referrals based on concern alone were cost-effective in the current study, and reported concerns were stronger predictors than positive screens of time-to-complete referrals. Qualitative analyses suggest a dynamic relationship between parents' concerns, providers' concerns, and screening results that is central to facilitating shared decision-making and influencing diagnostic assessment.

Multi-method process maps: An interdisciplinary approach to investigate ad hoc modifications in protocol-driven interventions

“…Drawing on our case study to illustrate this point, the causal mechanism of a multistage screening protocol might rely on the assumption that screening results will be assessed by the family and clinician as accurate and they will each follow the protocol in proceeding to the subsequent diagnostic assessment for ASD within the stated timeframe. However, findings in our current and prior work suggest that clinicians and families do not always agree with the tool's results and instead frequently place value on their own and one another's ASD concerns independent of screening results [19,38,40]. Such a finding suggests that the screening tool, itself, functions in concert with parents' and providers' concerns, knowledge, and beliefs about ASD rather than in isolation, requiring attention not only to the results of the tool but also the iterative process of shared decision-making and meaningmaking [19].…”

“…Our focus on ad hoc modifications highlights both the different "world view" of frontline providers, the discretion provided to those who implement protocols, and the potential for unanticipated modifications to intervention protocols. Frontline administrators in communitybased settings generally seek to serve a population rather than to test a theoretical model and thus may modify programs to address the perceived needs [19] or characteristics of their clients [20][21][22][23] and a host of other contextual factors (e.g., finances, agency capacities, funding requirements, or political climates) [24,25]. Implementation science studies increasingly recognize frontline administrators may deviate from an intervention protocol when knowledge exchange and alignment of values are inadequate [17,18].…”

“…Accordingly, we tested whether referrals based on concern alone were cost-effective compared to referrals based on screening results alone (with no reported concerns). Findings suggest that in the context of our screening process, reported concerns were, in fact, stronger predictors than positive screens alone in time-to-complete referrals and that referrals based on concern alone were cost-effective overall [19]. Based on these results, we intend to further optimize our screening protocol by developing implementation strategies (e.g., decision aids) to facilitate shared decision-making when there lacks alignment between the screening results, provider concern, and parent concern.…”

“…The Stage 1 questionnaire and the Stage 2 observational screenings were embedded into the routine clinical care provided by EI providers (i.e., frontline administrators), while implementation team members conducted a university-based diagnostic assessment for ASD. The complete description of the protocol-driven intervention is available in prior publications [19]. The present study primarily sought to identify whether and how frontline administrators modified the protocol-driven intervention to influence the primary outcome, time to diagnosis.…”

Understanding How Parents Make Meaning of Their Child’s Behaviors During Screening for Autism Spectrum Disorders: A Longitudinal Qualitative Investigation

“…Notably, our findings arrive from a community-based multi-stage screening protocol in a socially-disadvantaged community and specifically sought to reduce disparities in screening participation and outcomes (Sheldrick et al 2019;Mackie et al 2020). To examine whether disparities in screening participation and outcomes persisted in the parent study, our research team investigated the predictors of screening participation and outcomes at each stage of the process; demographic differences (race, language, public insurance) were observed only at first-stage screening and reflected higher participation for children of color and higher positive screens for publicly-insured children .…”