Bibliographic record
Abstract
AFTER AN ARTICLE THAT DISCOURagesroutineautismscreeningappearedonline inPediatrics, coauthor JanWillemGorter,MD,PhD,heard an array of heartfelt responses. “Some peoplewereupset,especiallyparentswith a child with autism,” said Gorter, an associate professor of pediatrics in the McMaster University Faculty of Health Sciences in Hamilton, Ontario, Canada. “I also got responses of parents who had a child mislabeled with autism initially, and that had a huge impact on the parents’ life and the child’s life.” The study struck an emotional chord by concluding that sound evidence to support routine screening is lacking (AlQabandi M et al. Pediatrics. 2011;128[1]: e211-e217). The authors say that currently available autism screening tools have not been evaluated in randomized controlled trials and that treatment is only modestly effective in certain subgroups of children. Although childhood screening, early detection, and treatmentareeffective for conditions such as congenital hypothyroidism or phenylketonuria, Gorter and his colleagues say that existing evidence doesnot showthat routine screening for autismdoesmoregoodthanharm.Infact, theysaythatmisdiagnosescanstigmatize children, expose them to unnecessary treatment, and trigger excessive costs. Gorter noted that the article focuses on routine screening that includes apparently healthy children, not clinical surveillance in which pediatricians evaluate children because they or the parents suspect a problem. The data were taken from a literature search designed to answer 7 questions concerning the appropriateness, feasibility, and value of screening for autism. In 2007, the American Academy of Pediatrics (AAP) published a clinical report that supports screening all children, regardless of risk factors, for autism spectrum disorders beginning at age 18 months (Johnson CP et al. Pediatrics. 2007;120[5]:1183-1215). For children without risk factors, the AAP report says appropriate evaluation methods include several screening tools that consist of parental interviews, questionnaires, or direct observation of the child. Among them are the Checklist for Autism in Toddlers (CHAT) and the Modified Checklist for Autism in Toddlers (M-CHAT). Both are available at no cost to primary care pediatricians. The reported specificity of CHAT is at least 98%, but sensitivity is between 18% and 38%. M-CHAT has 93% specificity and 85% sensitivity. Gorter and his colleagues say that M-CHAT, a 23item questionnaire that takes parents about 5 minutes to complete, is a “promising” tool, but that it misses 15 of every 100 children with autism. They say that none of the currently available screening tests “fulfill the properties of accuracy, namely high sensitivity, high specificity, and high predictive value” in population-based screening programs. Gorter said he and his colleagues did not intend for their study to split clinicians into “right” and “wrong” camps. “We took a scientific approach,” he said. “We wanted to consider not only what we know about testing itself, but also the impact of the screening program on public health and society at large.” Catherine Lord, PhD, director of the UniversityofMichiganAutismandCommunication Disorders Center in Ann Arbor, said the study lacks a necessary ingredient: a thoroughcost-benefit analysis. “We need to know what is the cost of screening and the benefit of screening, and what is the cost of treatment and the benefit of treatment,” said Lord, who was not involved in Gorter’s research. “It’s certainly not true that we can cure autism, but to say that nobody has ever shown a treatment has any effectiveness, any generalizability, that’s just not true.” Fred R. Volkmar, MD, director of the Child Study Center at the Yale University School of Medicine, agreed that interventions for autism do make a differ-
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.061 | 0.213 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.005 | 0.011 |
| Scholarly communication | 0.008 | 0.017 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.018 | 0.039 |
| Insufficient payload (model declined to judge) | 0.013 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".