P0030 - Searches for Web Tools to Screen for Autism Produce Few Results

Background: What happens when a parent is worried about a child’s risk for Autism Spectrum Disorder (ASD) and he/she searches online for help?

Objective: To simulate and characterize the results of a web search by a parent who is concerned about their child’s risk of ASD.

Methods: ASD screening tools were defined as a) existing on the internet; b) user inputs data; c) assigning a risk category to a child; d) intended for parental use; e) screening for ASD; f) can stand alone; g) designed for children < 9 years. A set of 42 relevant search terms was developed iteratively from Google Search auto-fill responses. An 18-member parent panel suggested their own terms and then chose their preferred terms out of the set of 42. For a 10-term search set, similar terms among top-ranked parent choices were combined and age-specific terms were added in order to represent all major themes from parent suggestions. Predominant browsers, search engines, and countries with large Anglophone internet markets (United States, United Kingdom, India, Nigeria) were identified based on demographic and market statistics. Australia and Canada were added due to known relevant research there. Country-specific search results were accessed using VPN (Virtual Private Network) software although Nigeria was unavailable. Each of the 5 resulting countries was searched using Google with all 10 terms to total 50 searches. The 3 most productive search terms were used to validate comparability of Bing with an additional 15 searches. The first page of each set of search results were recorded and tools meeting criteria were identified.

Results: Sixty-five searches yielded 1,475 results (mean 23 per search), 275 (19%) of which were sponsored links. Bing searches were comparable to Google. 147 (10%) results led to websites with content that met criteria for an ASD screening tool. Sixty-six (45%) of these required more than one click to arrive at the tool from the search result. Sixteen (1%) unique websites containing tools were ultimately identified, three of which were labeled M-CHAT (Modified Checklist for Autism in Toddlers) or M-CHAT-R (M-CHAT Revised). Four terms were most likely to identify any tools (tools per search): “autism quiz” (7.8), “autism screening tool” (4), “does my child have autism” (3.6), and “autism toddler” (2.6). Study searches yielded no new tools after these four terms were used in the United States and the United Kingdom.

Conclusion: This simulated search for ASD screening tools revealed few screening tool websites. Low yield rates illustrate the challenge for parents in finding useful tools online to help them assess their child’s risk for an ASD. To help parents connect to needed resources more efficiently, advocates may consider curated resources that are optimized for common online search engines.