Handbook of Autism and Pervasive Developmental Disorders, Assessment, Interventions, and Policy

By Fred R. Volkmar, Rhea Paul, Sally J. Rogers and Kevin A. Pelphrey

The latest and most comprehensive resource on autism and related disorders

Since the original edition was first published more than a quarter-century ago, The Handbook of Autism and Pervasive Developmental Disorders has been the most influential reference work in the field. Volume 2 of this comprehensive work includes a wealth of information from the experts in their respective specialities within the larger field of autism studies: Assessment, Interventions, and Social Policy Perspectives.

Within the three sections found in Volume 2, readers will find in-depth treatment of:

• Screening for autism in young children; diagnostic instruments in autism spectrum disorders (ASD); clinical evaluation in multidisciplinary settings; assessing communications in ASD; and behavioral assessment of individuals with autism, including current practice and future directions
• Interventions for infants and toddlers at risk; comprehensive treatment models for children and youth with ASD; targeted interventions for social communication symptoms in preschoolers with ASD; augmentative and alternative communication; interventions for challenging behaviors; supporting mainstream educational success; supporting inclusion education; promoting recreational engagement in children with ASD; social skills interventions; and employment and related services for adults with ASD
• Supporting adult independence in the community for individuals with high functioning ASD; supporting parents, siblings, and grandparents of people with ASD; and evidence-based psychosocial interventions for individuals with ASD
• Special topic coverage such as autism across cultures; autism in the courtroom; alternative treatments; teacher and professional training guidelines; economic aspects of autism; and consideration of alternative treatments

The new edition includes the relevant updates to help readers stay abreast of the state of this rapidly evolving field and gives them a guide to separate the wheat from the chaff as information about autism proliferates.

• Language: English
• Publisher: Wiley
• Release date: Feb 21, 2014
• ISBN: 9781118282205

Book preview

Table of Contents

Cover

Title Page

Copyright

Contributors

Preface

Section IV: Assessment

Chapter 24: Screening for Autism in Young Children

Characteristics of Autism in Young Children

Importance of Early Screening for Autism

The Screening Process

Review of Level 1 Screening Measures

Review of Level 2 Screening Measures

Conclusion and Future Directions

Cross-References

References

Chapter 25: Diagnostic Instruments in Autistic Spectrum Disorders

General Issues in Diagnosis of Autistic Spectrum Disorder

Issues in Selecting the Appropriate Focus and Level of Analysis

Implications of Information From Other Areas of Research for Diagnostic Instruments

Diagnostic Instruments and More Intellectually Able Individuals With ASD (Some of Whom Would Have Formerly Been Diagnosed as Having DSM-IV Asperger's Disorder)

Validity

Diagnostic Instruments for Autism

General Behavioral Measures that Include Core Features of ASD

Rating Scales

Diagnostic Interviews

Direct Observation Scales

Instruments for Asperger's Disorder

Conclusions

Cross-References

References

Chapter 26: Clinical Evaluation in Multidisciplinary Settings

Diagnostic Assessment

Specialized Assessments for Autism and Related Conditions

Integration of Findings

Summary

Cross-References

References

Chapter 27: Assessing Communication in Autism Spectrum Disorders

Assessing Prelinguistic Communication

Assessing Early Linguistic Communication

Assessing Communication in Children With Advanced Language

Conclusion

Cross-References

References

Chapter 28: Behavioral Assessment of Individuals With Autism: Current Practice and Future Directions

Characteristics of Behavioral Assessment: A Functional Ecological Approach

Domains of Behavioral Assessment

The Educational Relevance of Ecological Assessment

Limitations and Future Directions

Determination of Variables Controlling the Target Behavior

Development of a Treatment Plan

Evaluation of the Effects of Intervention

Cross-References

References

Section V: Interventions

Chapter 29: Interventions for Infants and Toddlers at Risk for Autism Spectrum Disorder

Introduction

Unique Features of Infant Toddler Interventions

Main Therapeutic Approaches in the Literature

Synthesis of Findings

In Closing

Cross-References

References

Chapter 30: Comprehensive Treatment Models for Children and Youth With Autism Spectrum Disorders

Operationalization

Implementation Measures

Replication

Efficacy Research

Individual Studies of CTM Features

Contemporary Influences on Development, Adoption, and Implementation of CTMs

Conclusion

Cross-References

References

Chapter 31: Targeted Interventions for Social Communication Symptoms in Preschoolers With Autism Spectrum Disorders

Introduction

Intervention

Five Promising Interventions: Effects on Generalized Characteristics

Links Between Intervention Components and Social Communication

Conclusion

Cross-References

References

Chapter 32: Augmentative and Alternative Communication

Introduction

AAC Assessment

AAC to Support Language Comprehension

AAC to Support Functional Communication/Language Production

AAC Instruction

Collateral Effects of AAC

Conclusion

Cross-References

References

Chapter 33: Interventions for Challenging Behaviors

What Are Challenging Behaviors?

Development of Behavior Analytic Function-Based Comprehensive Approaches for Challenging Behaviors

Components of Comprehensive Behavior Support Plans

Positive Behavioral Support and Persons With ASD

Identifying Evidence-Based Interventions for Specific Classes of Challenging Behaviors in ASD

Conclusion

Cross-References

References

Chapter 34: Supporting Mainstream Educational Success

Managing Challenging Behavior

Self-Management

Peer-Mediated Support

Visual Supports

Priming

Assistive Technology

Use of Paraprofessionals

Conclusions

Cross-References

References

Chapter 35: Supporting Inclusive Education

The Toddler and Preschool Years

Entering Elementary School

Moving on to Middle School and High School

College for Those Who Are Ready

Varied Outcomes of Inclusion

Educator and Family Perspectives

Conclusion

Cross-References

References

Chapter 36: Promoting Recreational Engagement in Children With Autism Spectrum Disorder

The Importance of Recreation

Patterns of Recreational Participation

Determinants of Recreational Participation

Supporting Recreational Engagement

Conclusion

Cross-References

References

Chapter 37: Social Skill Interventions

Social Characteristics of Youth on the Autism Spectrum

Assessment of Social Skills

Ingredients of Effective Social Skills Programs: Results of Meta-Analytical Studies

Evidence-Based Social Skill Interventions

Social Cognitive Interventions

Summary and Conclusions

Cross-References

References

Chapter 38: Employment and Related Services for Adults With ASD

Introduction

Discussion

Cross-References

References

Chapter 39: Beyond Academic Intelligence: Increasing College Success for Students on the Autism Spectrum

History

Making the Grade Academically and Socially for Students on the Spectrum

Conclusion

Cross-References

References

Chapter 40: Supporting Parents, Siblings, and Grandparents of Individuals With Autism Spectrum Disorders

Parents

Siblings

Grandparents

Conclusion

Cross-References

References

Chapter 41: Supporting Adult Independence in the Community for Individuals With High-Functioning Autism Spectrum Disorders

Challenges to Independence

Programming for Adult Independence

Conclusion

Cross-References

References

Chapter 42: Evidence-Based Psychosocial Interventions for Individuals With Autism Spectrum Disorders

Comprehensive Programs for Children With ASD

Focal Interventions

Evidence-Based Curricular Components

Summary and Closing Thoughts

Cross-References

References

Section VI: Social Policy Issues

References

Chapter 43: Autism Across Cultures: Perspectives From Non-Western Cultures and Implications for Research

Perspectives of Autism From Non-Western Cultures

The Impact of Culture on Autism Research

Summary and Conclusions

Cross-References

References

Chapter 44: Developing and Implementing Practice Guidelines

Introduction

Historical Background

Current Knowledge

Information for Discussion With Children, Young People, and Parents and Caregivers

Future Directions

Cross-References

References

Chapter 45: Autism in the Courtroom

Where Did All This Autism Litigation Come From?

An Overview of the Federal Statutory Protections

Just What Is a FAPE?

The Impact of Congress's Least Restrictive Environment Mandate

Bullying as a Recognized FAPE Deprivation

The Valuable Statutory Entitlement of Pendency

Which Party Has the Burden of Proof in IDEA Litigation?

How Good Must the Student's Private Program Be to Be Reimbursable or Otherwise Fundable?

The Threat of Criminal Proceedings

Housing Discrimination

The Future of Autism Litigation

Cross-References

References

Chapter 46: Alternative Treatments

Alternative Treatments for Autism

AAH Interventions

Challenges in Studying Alternative Treatments

The Appeal of Alternative Treatments

Addressing Alternative Treatments in Practice

Discussion

Cross-References

References

Chapter 47: Preparing Teachers and Professionals

Introduction

Assumptions

Components for Comprehensive Teacher Preparation and Continued Development

Teaching in Natural Contexts (Naturalistic Teaching Strategies)

Future Directions for Teacher Preparation for Working With Individuals With ASD

Cross-References

References

Chapter 48: Economic Aspects of Autism

Introduction

Efficiency and Equity

Economic Evaluation

Evidence on Costs

Evidence on Cost-Effectiveness

Links to Policy and Practice

Conclusion

Cross-References

References

Chapter 49: Translating Research Into Effective Social Policy

The Story of Joe

The Nature of ASD

Important Elements of Research

Mediating Elements

Important Elements of Policy

Other Considerations

Translating Research Into Policy and Outcomes: A Team Approach

Cross-References

References

Author Index

Subject Index

End User License Agreement

List of Illustrations

Figure 24.1

Figure 30.1

Figure 31.1

Figure 31.2

Figure 44.1

Figure 46.1

Figure 46.2

Figure 46.3

Figure 48.1

Figure 48.2

Figure 48.3

Figure 48.4

Figure 48.5

Figure 49.1

Figure 49.2

Figure 49.3

List of Tables

Table 24.1a

Table 24.1b

Table 24.2

Table 25.1

Table 25.2

Table 25.3

Table 27.1

Table 27.2

Table 27.3

Table 27.4

Table 27.5

Table 27.6

Table 27.7

Table 27.8

Table 27.9

Table 27.10

Table 27.11

Table 29.1

Table 29.2

Table 31.1

Table 31.2

Table 31.3

Table 33.1

Table 34.1

Table 36.1

Table 41.1

Table 44.1

Table 44.2

Table 44.3

Table 44.4

Table 46.1

Handbook of Autism and Pervasive Developmental Disorders

Volume 2

Assessment, Interventions, and Policy

Fourth Edition

Edited by

Fred R. Volkmar, Sally J. Rogers, Rhea Paul, and Kevin A. Pelphrey

Wiley Logo

Cover design: Wiley

This book is printed on acid-free paper.

Copyright © 2014 by John Wiley & Sons, Inc. All rights reserved

Published by John Wiley & Sons, Inc., Hoboken, New Jersey

Published simultaneously in Canada

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor the author shall be liable for damages arising herefrom.

For general information about our other products and services, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone], or fax [phone].

Wiley publishes in a variety of print and electronic formats and by print-on-demand. Some material included with standard print versions of this book may not be included in e-books or in print-on-demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com. For more information about Wiley products, visit www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Handbook of autism and pervasive developmental disorders / edited by Fred R. Volkmar, Sally J. Rogers, Rhea Paul, and Kevin A. Pelphrey.--Fourth edition.

Autism and pervasive developmental disorders

Includes bibliographical references and indexes.

ISBN 978-1-118-10702-7 (v. 1 : cloth : alk. paper)

ISBN 978-1-118-10703-4 (v. 2 : cloth : alk. paper)

ISBN 978-1-118-14068-0 (set : cloth : alk. paper)

ISBN 978-0-471-69442-7 (ebk.)

ISBN 978-1-118-28219-9 (ebk.)

I. Volkmar, Fred R., editor of compilation. II. Rogers, Sally J., editor of compilation. III. Paul, Rhea, editor of compilation. IV. Pelphrey, Kevin Archer, editor of compilation. V. Title: Autism and pervasive developmental disorders.

[DNLM: 1. Child Development Disorders, Pervasive. WS 350.8.P4]

RJ506.A9

618.92′85882---dc23

2013034363

Contributors

George M. Anderson, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Karla K. Ausderau, PhD

Kinesiology Department

University of Wisconsin–Madison

Madison, Wisconsin

Grace T. Baranek, PhD, OTR/L, FAOTA

Department of Allied Health Sciences

University of North Carolina at Chapel Hill

Chapel Hill, North Carolina

Erin E. Barton, PhD, BCBA-D

School of Education and Human Development

University of Colorado, Denver

Nirit Bauminger-Zviely, PhD

School of Education

Bar-Ilan University

Ramat-Gan, Israel

Scott Bellini, PhD

Social Skills Research Clinic

School Psychology Program

Indiana University

Bloomington, Indiana

Raphael A. Bernier, PhD

Center on Human Development and Disability

University of Washington

Seattle, Washington

Stefanie Bodison, OTD, OTR/L, C/NDT

Division of Occupational Science and Occupational Therapy

University of Southern California

Los Angeles, California

Leah Langford Booth, MS, CCC-SLP

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Kristen Bottema-Beutel, PhD

Department of Special Education

Vanderbilt University

Nashville, Tennessee

Brian A. Boyd, PhD

University of North Carolina at Chapel Hill

Chapel Hill, North Carolina

Jane Thierfeld Brown, EdD

University of Connecticut

School of Law

Hartford, Connecticut

Ariane Buescher, MSc

Personal Social Service Research Unit

London School of Economics and Political Science

London, United Kingdom

Alice S. Carter, PhD

Department of Psychology

University of Massachusetts, Boston

Boston, Massachusetts

Manuel F. Casanova, MD

Department of Psychiatry

University of Louisville

Louisville, Kentucky

Ya-Chih Chang, PhD

Center for Autism Research and Treatment

University of California

Los Angeles, California

Katarzyna Chawarska, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Frank Cicero, PhD, BCBA

Eden II Programs

Staten Island, New York

Elaine E. Coonrod, PhD

TEACCH Autism Program

University of North Carolina at Chapel Hill

Chapel Hill, North Carolina

Christina Corsello, PhD

Autism Discovery Institute

San Diego, California

Naomi Ornstein Davis, PhD

Department of Psychiatry

Duke University Medical Center

Durham, North Carolina

Whitney J. Detar, PhD

Graduate School of Education

University of California, Santa Barbara

Goleta, California

Oana de Vinck-Baroody, DO

Developmental-Behavioral Pediatrics

Yale School of Medicine

New Haven, Connecticut

Peter Doehring, PhD

ASD Roadmap

Chadds Ford, Pennsylvania

Shaunessy M. Egan, MS Ed, BCBA

Center for Children with Special Needs

Glastonbury, Connecticut

Ruth Blennerhassett Eren, EdD

Professor of Special Education

Southern Connecticut State University

New Haven, Connecticut

Donia Fahim, PhD, Cert. MRCSLT

Hunter College

City University of New York

New York, New York

Kate E. Fiske, PhD, BCBA-D

Douglass Developmental Disabilities Center

Rutgers, The State University of New Jersey

New Brunswick, New Jersey

Eric Fombonne, MD

Department of Psychiatry

McGill University

Montreal, Quebec, Canada

Solandy Forte, MSW, BCBA

Center for Children with Special Needs

Glastonbury, Connecticut

Megan Freeth

Psychology Department

University of Sheffield

Western Bank, Sheffield, United Kingdom

Lauren Gardner, PhD

Boling Center for Developmental Disabilities

University of Tennessee Health Science Center

Memphis, Tennessee

Peter F. Gerhardt, EdD

Organization for Autism Research

Arlington, Virginia

Mark P. Groskreutz, PhD

Southern Connecticut State University

New Haven, Connecticut

Rebecca Grzadzinski

Teachers College

Columbia University

New York, New York

Abha R. Gupta, MD

Department of Pediatrics

Yale University School of Medicine

New Haven, Connecticut

Laura J. Hall, PhD

Department of Special Education

San Diego State University

San Diego, California

Antonia Hamilton, PhD

School of Psychology

University of Nottingham

Nottingham, United Kingdom

Jan S. Handleman (deceased)

Sandra L. Harris, PhD

Douglass Developmental Disabilities Center

Rutgers, The State University of New Jersey

New Brunswick, New Jersey

Irva Hertz-Picciotto, PhD

Division of Environmental and Occupational Health

and

MIND Institute

UC Davis Medical Center

University of California, Davis

Davis, California

Alison Presmanes Hill, MS, PhD

Department of Pediatrics

Oregon Health & Sciences University

Beaverton, Oregon

R. Peter Hobson, MD

Institute of Child Health

University College London

London, United Kingdom

Patricia Howlin, MSc, PhD

St. George's Hospital Medical School

University of London

London, United Kingdom

Kara A. Hume, PhD

FPG Child Development Institute

University of North Carolina at Chapel Hill

Carrboro, North Carolina

Lisa V. Ibañez

University of Washington Autism Center

Seattle, Washingon

Brooke Ingersoll, PhD

Department of Psychology

Michigan State University

East Lansing, Michigan

William R. Jenson, PhD

Department of Educational Psychology

University of Utah

Salt Lake City, Utah

Connie Kasari, PhD

Center for Autism Research and Treatment

University of California at Los Angeles

Los Angeles, California

So Hyun Kim, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Martin Knapp

London School of Economics and Political Science

King's College London

London, United Kingdom

Lynn Kern Koegel, PhD

Koegel Autism Center

University of California, Santa Barbara

Goleta, California

Robert L. Koegel, PhD

Koegel Autism Center

University of California, Santa Barbara

Goleta, California

Elizabeth Lanter, PhD, CCC-SLP

Department of Communication Sciences and Disorders

Radford University

Radford, Virginia

Jennifer Leung, MD

Department of Pediatrics

Yale University School of Medicine

New Haven, Connecticut

Lauren M. Little, PhD

Department of Allied Health Sciences

University of North Carolina at Chapel Hill

Chapel Hill, North Carolina

James W. Loomis, PhD

Center for Children with Special Needs

Glastonbury, Connecticut

Catherine Lord, PhD

Center for Autism and the Developing Brain

Weill Cornell Medical College

White Plains, New York

Kristen Lyall, ScD

MIND Institute

UC Davis Medical Center

University of California, Davis

Davis, California

Megan C. Lyons, MS, CCC-SLP

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Suzanne L. Macari, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

David S. Mandell, ScD

University of Pennsylvania School of Medicine

Philadelphia, Pennsylvania

Kimberly Markoff, MSEd

St. John's Pavilion

Springfield, Illinois

Andrés Martin, MD, MPH

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Megan P. Martins, PhD, BCBA-D

Center for Development & Disability

University of New Mexico Health Sciences Center

Albuquerque, New Mexico

Gary S. Mayerson, JD

Mayerson & Associates

New York, New York

Erik Mayville, PhD, BCBA-D

Institute for Educational Planning

Connecticut Center for Child Development

Milford, Connecticut

Carla A. Mazefsky, PhD

Department of Psychiatry

University of Pittsburgh

Pittsburgh, Pennsylvania

Iain McClure, MB, BS

University of Edinburgh

Edinburgh, United Kingdom

James C. McPartland, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Judith Meyers, PhD

The Child Health and Development Institute of Connecticut, Inc.

Farmington, Connecticut

Amber R. Miller, BA

Graduate School of Education

University of California, Santa Barbara

Goleta, California

Elizabeth Milne

Psychology Department

University of Sheffield

Sheffield, United Kingdom

Pat Mirenda, PhD

Centre for Interdisciplinary Research and Collaboration in Autism

The University of British Columbia

Vancouver, British Columbia

Stewart Mostofsky

Laboratory for Neurocognitive and Imaging Research

Kennedy Krieger Institute

Baltimore, Maryland

Elizabeth C. Nulty, MS, BCBA

Center for Children with Special Needs

Glastonbury, Connecticut

Leona Oakes, BA

Strong Center for Developmental Disabilities

University of Rochester Medical Center

Rochester, New York

Samuel L. Odom, PhD

Frank Porter Graham Child Development Institute

University of North Carolina

Chapel Hill, North Carolina

Robert E. O'Neill

Department of Special Education

University of Utah

Salt Lake City, Utah

Mark J. Palmieri, PsyD, BCBA-D

School Consultation Services

Center for Children with Special Needs

Glastonbury, Connecticut

L. Diane Parham, PhD

Occupational Therapy Graduate Program

University of New Mexico

Albuquerque, New Mexico

Rhea Paul, PhD, CCC-SLP

Department of Speech-Language Pathology

Sacred Heart University

Fairfield, Connecticut

Kevin A. Pelphrey, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Lauren Pepa, BA

Douglass Developmental Disabilities Center

Rutgers, The State University of New Jersey

New Brunswick, New Jersey

Marie-Christine Potvin, PhD, OTR, ATP

Center on Disability and Community Inclusion

University of Vermont

Burlington, Vermont

Kelly Powell, MA

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Michael D. Powers, PsyD

Center for Children with Special Needs

Glastonbury, Connecticut

and

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Patricia A. Prelock, PhD

College of Nursing and Health Sciences

University of Vermont

Burlington, Vermont

Keith C. Radley, III, PhD

Department of Psychology

University of Southern Mississippi

Hattiesburg, Mississippi

Rajani Ramachandran, PhD

University of Calicut

Kerala, India

Brian Reichow, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Sally J. Rogers, PhD

UC Davis Medical Center

University of California, Davis

and

MIND Institute

Sacramento, California

Jessica L. Rohrer, MS, BCBA

Center for Children with Special Needs

Glastonbury, Connecticut

Justin Rowberry, Major, USAF

Developmental and Behavioral Pediatrics

Mike O'Callaghan Federal Medical Center

Nellis AFB, Nevada

Michael Rutter, CBE, MD, FRCP, FRCPsych, FRS

Social, Genetic and Developmental Psychiatry Centre

Institute of Psychiatry

King's College, London

London, United Kingdom

Maura G. Sabatos-DeVito, MS

Department of Psychology, Developmental Program

University of North Carolina at Chapel Hill

Chapel Hill, North Carolina

Micheal P. Sandbank, MEd

Department of Special Education

Vanderbilt University

Nashville, Tennessee

Liliane Beaudoin Savard, PT, DPT, PCS, PLLC

Zippy Life Physical Therapy

Montpelier, Vermont

Lawrence Scahill, MSN, PhD

School of Nursing

and

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Rebecca J. Schmidt, PhD, MS

Department of Public Health Sciences

MIND Institute

UC Davis Medical Center

University of California, Davis

Davis, California

Elizabeth Schoen Simmons, MS, CCC-SLP

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Katelyn Selver, BA

Strong Center for Developmental Disabilities

Department of Pediatrics

University of Rochester Medical Center

Rochester, New York

Elizabeth Sheppard, PhD

Psychology Department

University of Nottingham Malaysia Campus

Selangot Darul Ehsan, Malaysia

Frederick Shic, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Tristram Smith, PhD

Strong Center for Developmental Disabilities

Department of Pediatrics

University of Rochester Medical Center

Rochester, New York

Laurie Snider, PhD, OTR(C)

School of Physical and Occupational Therapy

McGill University

Montreal, Canada

Wendy L. Stone, PhD

UW Autism Center

University of Washington

Seattle, Washington

Helen Tager-Flusberg, PhD

Department of Anatomy and Neurobiology

Boston University School of Medicine

Boston, Massachusetts

Anita Thapar, MBBCh, PhD, FRCPsych, FMedSci

MRC Centre for Neuropsychiatric Genetics and Genomics

and

Institute of Psychological Medicine and Clinical Neurosciences

Cardiff University School of Medicine

Cardiff, Wales, United Kingdom

Caitlin S. Tillberg

Frank H. Netter School of Medicine

Quinnipiac University

North Haven, Connecticut

Rachael M. Tillman, BA

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Katherine D. Tsatsanis, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Nita Vaswani, DO

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Laurie Vismara, PhD

Psychiatry and Behavioral Sciences

University of California, Davis

and

MIND Institute

Sacramento, California

Giacomo Vivanti, PhD

Department of Psychology

Olga Tennisson Autism Research Centre

La Trobe University

Melbourne, Australia

Fred R. Volkmar, MD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Allison Wainer, MA

Department of Psychology

Michigan State University

East Lansing, Michigan

Christine Wenzel, BA, MA

Center for Students with Disabilities

University of Connecticut

Storrs, Connecticut

Alexander Westphal, MD

Department of Psychiatry

Yale University School of Medicine

New Haven, Connecticut

Susan W. White, PhD

Virginia Tech Autism Center

Virginia Tech

Blacksburg, Virginia

Lisa A. Wiesner, MD

Pediatrics and Adolescent Medicine

Orange, Connecticut

Tiffany Woynaroski, MS, SLP

Vanderbilt Kennedy Center

Nashville, Tennessee

Daniel Y.-J. Yang, PhD

Child Study Center

Yale University School of Medicine

New Haven, Connecticut

Paul Yoder, PhD

Vanderbilt Kennedy Center

Nashville, Tennessee

Katharine E. Zuckerman, MD, MPH

Division of General Pediatrics and Child and Adolescent Health Measurement Initiative

Oregon Health and Sciences University

Portland, Oregon

Preface

The pace of autism research has increased dramatically since the previous edition of this Handbook appeared. In that year, 2005, there were approximately 800 peer-reviewed scientific papers on autism, while in 2012 this number had increased to over 2,600. This marked increase in research productivity poses important challenges for editors of a comprehensive handbook devoted to autism. Inevitably, some difficult choices have to be made in balancing coverage of research, intervention, theory, and social policy.

In the 70 years since Kanner's initial description of autism, the condition has attracted interest from clinicians and researchers alike. As a disorder that impacts core aspects of socialization, it has posed important challenges for theories of developmental psychology and neurobiology as well as for clinical practice in diagnosis and intervention, and studies of diagnostic validity and treatment. Essentially every theory relating to child development—cognitive, social, behavioral, affective, neurobiological—has been applied to understanding this enigmatic condition. Autism has served as a paradigmatic disorder for research on the essential preconditions for normal social-cognitive maturation—expression and recognition of emotions, intersubjectivity, sharing the focus of interest with other people, the meaning and uses of language, forming attachments, and relating empathetically to others.

In developing this new edition, we have been mindful of the considerable progress made in the field as well as areas where knowledge remains limited. Great advances have been made, for example, in understanding the social brain, in genetics, and in basic aspects of neurobiology. Other advances have also been made in the areas of intervention and there is a new and growing convergence between research findings and evidence-based practice. On the other hand, there are many areas where knowledge remains limited—for example, work on aging in autism is almost nonexistent.

As with other areas of science, we believe that autism scholarship and service will advance when we adopt, as much as possible, rigorous standards of scientific research. Our aim with this fourth edition is to provide a comprehensive account of current work in the field. In many instances, authors have kindly revised earlier contributions in light of current research; in other cases, we have solicited new contributors and chapters. Our goal for these volumes is to provide timely overviews in key areas that can help researchers, clinicians, and policy makers.

We are acutely aware that investigators and clinicians, working alongside families and advocates, have learned so much, often with limited resources. The knowledge summarized in these volumes speaks to the commitment of these individuals in understanding and caring for children with autism. We hope that these volumes document their achievements and inspire their future efforts.

We thank a number of colleagues who have critiqued early versions of chapters or who helped us select chapter authors or focus chapter topics. These include Brian Reichow, Roger Jou, William Nordhaus, Peter Doehring, Abha Ghupta, Carlisle Runge, Iain McClure, Christopher McDougle, Linda Mayes, George Anderson, and Dean Sutherland. We also thank a number of individuals for secretarial and administrative support: at the Child Study Center Lori Klein, Emily Hau, and Rosemary Serra, and from UC Davis MIND Institute, we would like to thank Diane Larzelere. We are also grateful to our editor at Wiley, Patricia Rossi, who has helped us consistently strive for excellence.

Section IV

Assessment

The assessment of individuals with autism spectrum disorder (ASD) calls on the expertise of a variety of disciplines, including pediatrics, neurology, psychiatry, psychology, special education, speech-language pathology, and physical rehabilitation, to name a few. In this section, we ask expert clinicians from these disciplines to discuss the assessment process from their point of view. Our authors address:

Major goals of assessment, including:

screening to determine the need for further assessment;

diagnostic evaluation that establishes eligibility for services;

differential diagnosis that distinguishes ASD from other neuropsychiatric syndromes;

identifying baseline function in a range of developmental areas, against which progress in intervention can be measured;

characterization of strengths and needs of the individual, in order to guide the development of intervention objectives and procedures; and

detailed description of functioning across a range of developmental areas, in order to characterize the range of phenotypic expression within the syndrome.

Methods available for rigorous assessment of core symptoms associated with ASD, including:

standardized instruments,

observational protocols,

criterion-referenced probes, and

caretaker and teacher questionnaires and interviews.

The application of evidence-based practice to assessment.

These issues not only impact clinical practice in ASD, but also affect the conduct of research. Until recently, there have been few well-standardized, validated, and reliable measures that provide diagnostic and assessment information about individuals with ASD, making both creditable diagnoses and replicable research problematic. In the area of screening and diagnostic assessment, great strides have been made since the previous edition of this handbook in the development and validation of measures designed specifically to determine the need for assessment as well as the diagnosis of ASD. This progress has come as a result of intensive research efforts to assess the sensitivity, specificity, and validity of screening and diagnostic measures. Still, issues and controversies remain regarding the construction, implementation, and interpretation of these measures, many of which are discussed in the chapters in this section.

Differential diagnosis within the autism spectrum and among ASD and other disorders remains stubbornly difficult, particularly for the youngest children, for whom early identification is so crucial to optimal outcomes. Recent changes in DSM-5 highlight this issue, and heated debate continues about the propriety of moving from older systems identifying subgroups within the autism spectrum to a system of treating it as an undifferentiated continuum. As chapters in this section demonstrate, assessment contributes to this debate by providing the opportunity to answer questions on the basis of data rather than theory or anecdotal experience. An important goal of assessment is to move beyond global descriptions to more fine-grained, precise documentation of functioning across domains, including cognitive, linguistic, communicative, social, motor, and adaptive behaviors. These fine-grained descriptions of individual patterns of behavior and ability contribute in important ways to the identification of an individual's baseline function for gauging progress in intervention and of the profile of strengths and needs that guide the development of educational goals. In addition, though, they provide the opportunity to explore patterns across individuals. The discovery of consistent patterns of behavioral and cognitive functions provides an empirical basis for testing hypotheses about the existence and validity of subgroups within the broad autism spectrum, and the possibility of finding biological correlates, including those at genetic, neuroanatomical, neurophysiological, or neurochemical levels. The identification of subgroups with such biomarkers holds out the promise that specific medical or pharmacological interventions may one day be devised to address specific elements of the syndrome within specific subgroups. The fulfillment of this promise depends to a great degree on the precision of assessment information collected, and on the accumulation of this information in research based in psychometrically sound clinical instruments and structured observation.

An additional message that emerges from this section is the importance of seeing ASD within a developmental framework. Despite the unique symptoms and uneven skills often seen in individuals with ASD, many strands of their development nonetheless follow the normative sequence. This normative aspect of development is a crucial consideration in determining the needs of individuals with ASD, in terms of focusing on developmentally appropriate social, academic, self-help, and motor skills. It is incumbent upon clinicians not to be diverted by the many atypical behaviors of people with ASD into ignoring the aspects of their function that will allow them to take advantage of interactions with peers and other members of their community. Detailed assessment information collected within a developmental framework, including well-constructed instruments designed specifically for individuals with ASD as well as psychometrically sound instruments that take a broader developmental spectrum into account, is essential for optimizing these opportunities. Many chapters in this section provide guidelines and methods for achieving these developmentally situated evaluations.

Another point made clear in this section is the degree of experience, expertise, and teamwork needed to accomplish the kind of multidimensional assessment that will both deeply characterize an individual's diagnostic and developmental status and provide the fine-grained data that will inform and advance research on ASD. Individuals with ASD can show unusual preferences for reinforcement, attentional, and motivational characteristics, and uneven profiles across domains. These differences can make it challenging for them to participate in assessment activities, and to demonstrate their optimal level of competence. Clinicians performing assessments need more than knowledge of their instruments; they need understanding of the particular challenges individuals with ASD face in the assessment process, insight and empathy with their struggles, and patience and flexibility to elicit the best performance. They need, too, to work closely with colleagues from other disciplines and to think creatively about planning the overall assessment experience to maximize the client's opportunities for success. No one discipline has a monopoly on the diagnostic process, and the most thorough and effective assessments will involve interprofessional collaboration in planning, administering, and interpreting assessment data.

Finally, the results of assessments must be placed within the context of the opportunities the individual has had for social relations, academic achievement, recreation, and self-advocacy. We can anticipate that children who had consistent, intensive educational programming since early childhood will appear quite different on assessment than those with more restricted experiences. Moreover, contextually based assessments can help to locate ways in which the environment can be engineered to enhance appropriate experiences that will enable the acquisition of a range of academic and practical skills. Chapters in this section provide examples of the ways in which this kind of ecological assessment can round out the picture of the skills and needs of individuals with ASD.

The results of careful and comprehensive assessment are a function of the individual's congenital biological endowment, maturation, personal experiences, and community opportunities. Used appropriately, objective, rigorous assessment provides both the best guide to comprehensive, effective intervention and the most direct path to advancing research on this complex syndrome.

Chapter 24

Screening for Autism in Young Children

Lisa V. Ibañez, Wendy L. Stone, and Elaine E. Coonrod

Characteristics of Autism in Young Children

Importance of Early Screening for Autism

The Screening Process

Dimensions of Screening Measures

Evaluating Screening Measures

Review of Level 1 Screening Measures

Broad-Based Measures

Autism-Specific Screening Measures

Level 1 Summary

Review of Level 2 Screening Measures

Screening Tool for Autism in Toddlers (STAT)

Childhood Autism Rating Scale (CARS/CARS2)

Gilliam Autism Rating Scale (GARS/GARS-2)

Social Communication Questionnaire (SCQ)

Other Promising Measures

Level 2 Summary

Conclusion and Future Directions

Cross-References

References

Evidence from early intervention research clearly indicates that participation in specialized behavioral intervention programs at young ages can optimize the social-communicative and cognitive outcomes of children with autism and autism spectrum disorders (ASDs; Dawson et al., 2010; Ingersoll, 2010; Kasari, Gulsrud, Wong, Kwon, & Locke, 2010; Landa, Holman, O'Neill, & Stuart, 2010), and may even normalize patterns of brain activity (Dawson et al., 2012). Although parental concerns about their child's development are often present by 17−19 months (Coonrod & Stone, 2004; De Giacomo & Fombonne, 1998), the average age of an ASD diagnosis in the United States is 4.5 years (Centers for Disease Control [CDC], 2012) and the median age is 5.7 years (Shattuck et al., 2009). As a result, many children miss the opportunity to benefit from early intervention services. Screening for autism in young children has the potential to promote earlier diagnosis and more widespread, systematic referrals to appropriately specialized intervention programs. This chapter addresses several topics related to the screening of young children for autism and ASD, including the early characteristics of autism, early screening practices and models, and the current state of the science regarding early screening measures for autism.

Characteristics of Autism in Young Children

Autism is a neurodevelopmental disorder that emerges early in life. The past 15 years have seen a dramatic increase in research focused on identifying the earliest signs and symptoms of the emerging disorder. Underlying the push for earlier diagnosis is the hope of preventing or mitigating some of the symptoms of autism by providing targeted interventions during a period of rapid brain growth and development that occurs in infancy and toddlerhood. Behavioral studies have revealed that the social-emotional and social-communicative impairments that are well established in preschool-aged children with autism are also present by the second year of life in children later diagnosed with autism. For example, behaviors such as socially directed gaze, motor imitation, social smiling, response to adult social bids and expressions of distress, and initiation of joint attention by pointing to and showing objects differentiate infants with and without autism using both retrospective (Adrien et al., 1993; Baranek, 1999; Osterling, Dawson, & Munson, 2002; Werner, Dawson, Osterling, & Dinno, 2000) and prospective research designs (Charman et al., 1997; Hutman et al., 2010; Landa, Holman, & Garrett-Mayer, 2007; Ozonoff et al., 2010; Rozga et al., 2011; Yoder, Stone, Walden, & Malesa, 2009). In a similar vein, repetitive motor behaviors and atypical object use have also been found in children with autism by the second year of life (Ozonoff, Macari, Goldring, Thompson, & Rogers, 2008; Watt, Wetherby, Barber, & Morgan, 2008).

Behavioral findings such as these have informed the development and refinement of many screening tools currently used for the early detection of autism risk. However, the identification of this pattern of symptom expression in a young child can be challenging for several reasons. First, social-communication impairments, often considered to be the core feature of autism, represent negative symptoms, or the absence or reduced frequency of behaviors expected for a child's developmental level (Filipek et al., 1999). It is very difficult to interpret the absence of a behavior in young children, whose moment-to-moment behaviors are more vulnerable to the influence of internal state and setting factors. For example, the failure to observe social smiling in an 18-month-old during a clinic visit may represent a red flag for autism, but may also reflect fatigue, hunger, or a host of other contributing causes. Second, social-communicative behavior is not an all-or-none phenomenon. It is rarely the case that young children with autism never make eye contact, or never imitate the actions of others. Rather, the differences between children with and without autism tend to be in the consistency with which these behaviors are exhibited and the effort required to elicit them (Baranek, 1999), both of which can be difficult to measure or assess. Third, there are no established norms or milestones for social behaviors in the same way they exist for motor or language development. What percentage of time is a child expected to look at a parent who calls his name, and how does one quantify social reciprocity during everyday interactions? There is much greater ambiguity in the definitions and expectations for social behaviors than for other developmental milestones.

Importance of Early Screening for Autism

For over a decade, several professional groups and consensus panels, including the American Academy of Neurology (Filipek et al., 1999; Filipek et al., 2000) and the American Academy of Pediatrics (AAP Committee on Children with Disabilities, 2001), have advocated for early and regular screening for autism. Specific recommendations from the American Academy of Pediatrics in 2007 include providing surveillance at every well-child visit, with special attention to subtle red flags and the infant siblings of children with autism (high-risk siblings); conducting autism-specific screening for all children at 18 and 24 months of age; scheduling a targeted clinic visit when parent or physician concerns remain after a negative screen; and acting on a positive screen, or on the presence of two or more autism risk factors, by referring for an autism evaluation and/or early intervention (Johnson & Myers, 2007). Although these guidelines provide concrete steps and recommended tools for autism-specific screening, compliance with the guidelines is by no means universal. A recent survey conducted across six states revealed that 60% of pediatricians conduct formal screening for autism at 18 months and 50% at 24 months (Arunyanart et al., 2012). While these rates are substantially higher than the 8% reported in an earlier study of pediatrician practices (Dosreis, Weiner, Johnson, & Newschaffer, 2006), there is much room for improvement.

Several recent studies have provided additional support for the importance of early formal screening for autism. For example, one study conducted in a large, community-based pediatric practice found that the use of formal autism screening tools was more effective in identifying ASD than was pediatrician's clinical judgment alone (Miller et al., 2011). Other studies have found that training health care workers in the use of autism screening instruments is not only associated with increased knowledge about early social and communication markers of autism, but also contributes to significant practice change (Charman et al., 2001; Oosterling et al., 2010; Swanson et al., in press; Warren, Stone, & Humberd, 2009). Perhaps the most compelling support, however, comes from an early detection program conducted in the Netherlands (Oosterling et al., 2010), in which health care providers in a certain region were trained in the use of a formal autism screening tool, the Early Screening of Autistic Traits Questionnaire (ESAT; Dietz, Swinkels, van Daalen, van Engeland, & Buitelaar, 2006; Swinkels et al., 2006). The mean age of diagnosis in the region participating in the program dropped from 82.9 months to 63.5 months, whereas no change was observed in a region not participating in the program. In addition, children in the targeted region were over nine times more likely to be diagnosed before 36 months relative to those in the comparison region. Thus, the routine use of formal screening tools has been shown to increase providers' knowledge about early red flags, reduce bias in referrals for further clinical assessment, and lower the age of autism diagnosis.

The Screening Process

Screening is designed to be a brief assessment for identifying children in need of a more comprehensive diagnostic evaluation due to risk of delay or disability (Meisels, 1985). As such, screening is the first of a multistep process that may also include rescreening, referral to a diagnostic center for further assessment, and referral to early intervention programs (Aylward, 1997). Screening measures differ from diagnostic measures in that they typically require less time, training, and experience to administer, and the results of screening measures indicate levels of risk for disability rather than provide a diagnosis.

Dimensions of Screening Measures

Screening measures can vary across several different dimensions, which impact their suitability for use in different practice settings (Zwaigenbaum & Stone, 2006). Level 1 screening (i.e., universal screening) is conducted for the purpose of identifying children at risk for developmental disorders from the general population of unselected, presumably low-risk children. Accordingly, Level 1 screening measures are used most commonly in pediatric or other primary health-care settings, where they can be administered routinely to all children during well-child visits, regardless of whether developmental concerns are present. For use in these settings, Level 1 screeners need to be quick and easy to administer, score, and interpret. In contrast, the purpose of Level 2 screening is to identify children's risk for a specific disorder after a developmental concern has already been identified. Thus, Level 2 screening measures can help differentiate children at risk for autism from those at risk for other developmental disorders, such as global developmental delay or language impairment. Level 2 screening measures are used more often in community settings such as child-find agencies, early intervention programs, or evaluation centers serving children with a variety of developmental challenges. Many of these settings do not have the same types of time pressures as pediatric practices and can accommodate somewhat more involved screening measures and approaches.

Another dimension on which screening measures differ is their format, which can be informant report, observational, or interactive. Each format is associated with specific strengths and liabilities. Informant-report measures, which often involve parental report, have the advantage of capitalizing on caregivers' knowledge about multiple areas of the child's behavior over time and across diverse situations and contexts, but may be compromised by reporting bias. In contrast, observational measures that employ ratings of children's behaviors in the immediate context can benefit from the rater's training and expertise regarding age-specific norms and expectations, but are restricted in the breadth of information that can be acquired at a single point in time. As well, screening measures involving structured interactions with the child provide a more direct experience of his or her social-communicative behaviors and interaction style, but tend to be more time and training intensive. The selection of screening measures thus needs to be guided by considerations such as the type of information desired, the demands of the screening context, and the background and training of the user. Given the availability of different types and levels of screeners, it is feasible that using combinations of screening tools and multilevel models of screening could be more effective than using a single screening measure at a single point in time. In fact, several recent studies exploring screening models have successfully used combinations of screening measures as well as serial approaches to screening (Miller et al., 2011; Roux et al., 2012) for young children with autism.

The flexible use of screening tools may also serve the ultimate goal of improving access to early intervention for young children with autism. In many communities, the traditional diagnosis-intervention model can pose some serious challenges in access to early intervention services, particularly for young children with autism. This model (see Figure 24.1a) requires that children receive a formal diagnosis of autism before they can obtain autism-specialized services. However, there are several common roadblocks that occur between the time of initial concerns and the time that specialized intervention is initiated. First, primary care providers may be reluctant to make early referrals for diagnostic evaluation for reasons including a lack of familiarity with the early signs of autism, lack of knowledge regarding the use of autism screening tools, and lack of comfort speaking with families about autism concerns (Dosreis & Weiner, 2006; Tomlin, Koch, Raches, Minshawi, & Sweizy, 2013). Once the referral is made, families may often discover extensive waiting lists for diagnostic evaluations, resulting in long delays in obtaining a confirmatory diagnosis of autism (Barton, Dumont-Mathieu, & Fein, 2012; Young, Brewer, & Pattison, 2003). Finally, after the child receives an autism diagnosis, families may be faced with a shortage of autism-specialized service providers (Tomlin et al., 2013; Wise, Little, Holliman, Wise, & Wang, 2010). In the meantime, many children have aged out of eligibility for early intervention services. In contrast, a risk-prevention model may facilitate earlier access to specialized intervention by using interactive Level 2 screening measures to confirm autism risk as well as identify targeted treatment goals and activities that can be initiated immediately (see Figure 24.1b). In this way, specialized intervention can begin earlier, prior to the formal diagnosis, thus capitalizing on early brain plasticity and the possibility of mitigating symptoms and improving outcomes (Dawson et al., 2010; Dawson et al., 2012; Kasari et al., 2010; Landa et al., 2010).

c24f001

Figure 24.1 Potential role of screening in expediting autism-specialized intervention. (a) Traditional diagnosis-intervention model. A formal autism diagnosis is required before specialized intervention is provided. (b) Risk-prevention model. A Level 2 interactive screen provides information about risk status and child-specific behavioral needs so that specialized intervention can begin while waiting for the formal diagnosis.

Evaluating Screening Measures

Four psychometric properties are typically considered when evaluating screening measures: sensitivity, specificity, positive predictive value, and negative predictive value. Users of screening tools should have a clear understanding not only of the general meaning of these terms but also of how these values are affected by the specific setting in which screening occurs. Sensitivity refers to the proportion of children with developmental disorders who are identified as being at risk by the screening measure, while specificity refers to the proportion of children without developmental disorders who are identified as being not at risk (Aylward, 1997). The sensitivity and specificity of a screening measure are determined by comparing the results of the screener (i.e., risk or no risk) with the diagnostic gold standard for a disorder (Riegelman & Hirsch, 1989). Sensitivity and specificity can range in value from 0.0 to 1.0, with higher values indicating greater probability that those with and without the disorder will be correctly identified by the screening measure. In developing a screening measure, the goal is to identify a cutoff score in which both sensitivity and specificity are maximized.

Sensitivity levels of .80 or higher are generally recommended (Glascoe, 1991; Squires, 2000), meaning that at least 80% of children who truly have developmental disorders should be identified by their scores on the screening measure. Recommended specificity levels range from .80 to .90 (Glascoe, 1991; Squires, 2000), meaning that 80% to 90% of children without a developmental disorder should be identified as being not at risk.

Sensitivity and specificity are interrelated such that increasing one by changing the measure's cutoff score to increase sensitivity will often decrease its specificity. For example, a measure's sensitivity can often be improved by lowering the cutoff score so that the likelihood of detecting those with a disorder is increased. However, at this lower threshold, it is also easier for those without a disorder to be misidentified as being at risk, resulting in a lower specificity (Aylward, 1997; Frankenburg, 1974). In the most extreme example, a screening measure's sensitivity can be raised to 100% by simply changing the cutoff to the lowest possible score, such that all scores would indicate risk for disorder. However, the specificity would then be zero, and the screening measure would obviously be useless (Frankenburg, 1974).

Sensitivity and specificity indicate the proportion of children with and without the disorder who are correctly identified by the screening measure. However, it is also important to consider the proportion of children identified as being at risk (or not at risk) by the screening measure who actually have (or do not have) the disorder (Riegelman & Hirsch, 1989). Positive predictive value (PPV) refers to the proportion of children identified as being at risk who actually have the disorder, while negative predictive value (NPV) is the proportion of children identified as not being at risk who do not have the disorder (Aylward, 1997). Like sensitivity and specificity, PPV and NPV are proportions with values ranging from 0.0 to 1.0, with higher values indicating greater probability that the screening result is accurate.

It is important to note that PPV and NPV vary according to the prevalence, or base rate, of the disorder; as the base rate increases, PPV will increase and NPV will decrease. Conversely, as the base rate of the disorder decreases, PPV will decrease and NPV will increase (Riegelman & Hirsch, 1989). These changes in PPV and NPV can be illustrated using a sample of 1,000 children and a hypothetical screener with a sensitivity of .80 and specificity of .80. If 1,000 children are screened for autism from the general population, which has a base rate of roughly 1%, then the screener will identify 206 individuals as at risk for the disorder (see Table 24.1a). Because only 8 of those 206 children actually have the disorder, the resulting PPV is 8/206, or approximately .04. This measure will identify 794 children as being not at risk for the disorder. Because 792 of those 794 children do not have the disorder, the resulting NPV is 792/794, or approximately .99. In contrast, if 1,000 children are screened for autism in a developmental evaluation center, which has a base rate of roughly 40%, the same screener will identify 440 children as at risk for the disorder (see Table 24.2b). Because 320 of those children actually have the disorder, the resulting PPV is 320/440, or .73. In this sample, the measure will identify 560 children as not at risk. Because 480 of those children do not have the disorder, the resulting NPV is 480/560, or approximately .86. While the two PPVs are on opposite ends of the spectrum (.04 vs. .73), Glascoe (2005) reports that PPVs ranging between .30 and .50 are not uncommon.

Table 24.1a Positive and Negative Predictive Values for a Screening Measure With Sensitivity and Specificity of .80 in a Sample of 1,000 With an Autism Prevalence Rate of 1%

TP = True positives; FP = False positives; FN = False negatives; TN = True negatives; Sensitivity = .80 (8/10); Specificity = .80 (792/990); Positive predictive value = .04 (8/206); Negative predictive value = .99 (792/794)

Table 24.2 Positive and Negative Predictive Values for a Screening Measure With Sensitivity and Specificity of .80 in a Sample of 1,000 With an Autism Prevalence Rate of 40%

TP = True positives; FP = False positives; FN = False negatives; TN = True negatives; Sensitivity = .80 (320/400); Specificity = .80 (480/600); Positive predictive value = .73 (320/440); Negative predictive value = .86 (480/560).

The practical implication of the two previous examples is that screening measures with the same sensitivity and specificity can be expected to have higher PPVs when used in a setting characterized by a higher prevalence of the disorder, such as a developmental clinic, compared to one characterized by a lower prevalence, such as a general pediatric practice. The PPV of .04 obtained in the general population, for example, indicates that among children who screen positive, only 4% will have autism; thus the screener has overidentified 96% of the children detected as at risk. This rate of overidentification is in stark contrast to the PPV of .73 obtained for the developmental evaluation center, where 27% of the children who screened as at risk were false positives. The relatively low prevalence of autism in the general population, as compared to a clinic referral sample, makes the direct comparison of Level 1 and Level 2 screeners' psychometrics difficult because the two types of screeners are linked to the separate settings. In fact, Clark and Harrington (1999) note that for a screening measure with a sensitivity and specificity of .80, it takes a prevalence of about 25% to obtain a PPV above .50 and correctly identify children as being at risk for the disorder a little better than half of the time.

Although these general guidelines can be helpful when evaluating a screening measure, they are not the only considerations. The relative cost of incorrectly identifying a child as being at risk should be compared to the cost of failing to identify a child who is at risk for a disorder. Given the documented benefits of early intervention for children with autism, it may be more beneficial to overrefer children for further evaluation than to underrefer and potentially delay identification and treatment. In this case, an emphasis on higher sensitivity relative to specificity might be desirable. In fact, in the case of autism screening, there is evidence that the majority of the overreferred children have some type of developmental disorder other than autism that would justify their identification (Robins & Dumont-Mathieu, 2006).

Another important consideration is that it is not possible to obtain accurate values for sensitivity, specificity, or NPV if children who screen negative do not receive a confirmatory diagnostic evaluation. This situation poses a particular problem for Level 1 measures, which may be administered to thousands of children in the course of a large population screening study. Not only would it be challenging to obtain high rates of compliance with a follow-up evaluation from parents of children who pass the screening, but the expense involved in following up such potentially large samples would be prohibitive. An alternative strategy is to select a random subset of screen-negative children for further evaluation.

Review of Level 1 Screening Measures

There are two strategies for identifying young children at risk for autism within the general population. One approach is to use a Level 1 screening measure designed to identify children at risk for a wide range of developmental problems (i.e., broad-based screening), with the expectation that some children with autism will be identified because of their cognitive, social, or language delays. The other approach is to use a Level 1 screening tool that specifically targets the symptoms of autism (i.e., autism-specific screening). Screening measures of both types of are described in the following sections; only those measures with peer-reviewed publications are included in this review.

Broad-Based Measures

Several broad-based screening measures are available for identifying young children with general developmental delays. These measures may target a range of developmental areas, including cognitive, communication and language, motor, social, self-help, and behavioral domains. Two of the most commonly used tools are the Parents' Evaluation of Developmental Status (PEDS; Glascoe, 1998; Glascoe, 2003) and the Ages and Stages Questionnaire, Third Edition (ASQ-3; Bricker & Squires, 1999; Squires & Bricker, 2009). The ASQ-3 is a 30-item parent report measure for children between 1 and 66 months of age that examines five domains: communication, fine motor, gross, motor, personal-social, and problem solving. The ASQ-3 has a high test-retest reliability (.92) and interrater reliability (.93) for risk classification, with sensitivity ranging from .83 to .89 and specificity ranging from .80 to .92 across ages. The PEDS is a 10-item parent report measure suitable for children between 1 and 95 months of age. Parental concerns about development in five domains (global/cognitive, expressive language, receptive language, social-emotional, and other) are assessed. Across the age range, sensitivity is .74 to .79 and specificity is .70 to .80. While one would not expect either of these measures alone to identify all cases of autism, their use in combination with autism-specific screeners may prove beneficial. One project (Pinto-Martin et al., 2008) that used the PEDS along with an autism-specific screener, the Modified Checklist for Autism in Toddlers (M-CHAT; Robins, Fein, Barton, & Green, 2001) found that the PEDS failed to identify the majority of the children who screened positive for autism on the M-CHAT. However, another program designed to reach underserved populations reported success using the PEDS in combination with the M-CHAT (Roux et al., 2012).

The Infant/Toddler Checklist (ITC; Wetherby & Prizant, 2002) is a population-based screener designed to identify children's risk for language and communication impairments. The ITC is a 24-item parent report questionnaire that is a component of the Communication and Symbolic Behavior Scales Developmental Profile (CSBS DP; Wetherby & Prizant, 2002; Wetherby et al., 2004). It yields composite scores in three domains: social (e.g., use of communicative gestures), speech (e.g., use of sounds), and symbolic (e.g., use of objects), as well as a total score. The ITC was normed on a sample of over 2,000 children and provides cutoff scores indicating risk for communication delay.

Though not designed as an autism-specific screen, two studies conducted by Wetherby and colleagues have examined the screening properties of the ITC for children with ASD. In the first study, the ITC was completed for a sample of 3,021 infants under 24 months who were recruited from health care, child care, and public settings (Wetherby et al., 2004). Children scoring below the 10th percentile on the ITC were invited for a subsequent evaluation with the CSBS DP Behavior Sample (Wetherby et al., 2004) and a subgroup scoring below the 10th percentile on the CSBS DP Behavior Sample received a diagnostic evaluation. ITC results were then examined for three groups: 18 with ASD, 18 with developmental delay, and 18 with typical development. ITC screening properties for the ASD group were sensitivity .94, specificity .89, PPV .90, and NPV .94. However, it is important to note that children in the ASD group were a highly selected sample, in that they (1) represented the small fraction of the original ITC sample who also failed subsequent screening using the Behavior Sample; and (2) were more likely than the other groups to have received the ITC from health care providers, suggesting that early concerns for this sample may have been present.

A second study included a cumulative sample of 5,385 children from earlier studies who received multiple screenings with the ITC between 6 and 24 months of age (Wetherby, Bronsan-Maddox, Peace, & Newton, 2008). Sixty children in the sample were diagnosed with ASD at 4 years of age. For children between 9 and 24 months of age, sensitivity (based on a positive score on at least one administration of the ITC) ranged from .77 to 1.00; for 6- to 8-month-olds, sensitivity was .20. These results indicate that children with ASD are among those who are identified by the ITC, which would be expected given the frequently observed language and communication delays in young children with ASD. Another component of the CSBS DP, the Systematic Observation of Red Flags (SORF; Wetherby et al., 2004), has been used in combination with the ITC and is discussed in a later section.

The ITC was also examined by two independent research groups. One group conducted a large population study of children screened during their pediatric well-child visit at 12 months of age (Pierce et al., 2011). A total of 1,318 out of 10,479 infants screened positive on the ITC. Of this group, diagnostic evaluation results were available for 184 children: 32 had ASD, 65 had language delay or developmental delay, 36 had other diagnosed disorders, and 46 had no diagnosed condition. The PPV for the ITC was .75 for the combined diagnostic groups and .17 for the ASD group alone. These results suggest the utility of the ITC as a Level 1 screener for a wide range of developmental disorders, but not for distinguishing autism specifically. Another group used the ITC and the M-CHAT in a large, community-based pediatric practice and found that neither screener identified all cases of ASD, and both screeners failed to identify some cases (Miller et al., 2011).

Autism-Specific Screening Measures

Several autism-specific tools have been developed as Level 1 screeners, designed to identify risk for ASD in general population samples. Psychometric and design properties of these measures are summarized in Table 24.2.

Table 24.2 Summary of Psychometric Properties for Autism Screening

Note: * = different cutoff scores are included within these ranges; P = Population-based sample; C = Clinic-based or community-based sample; P/C