AGRE Diagnostic Information

Autism Speaks AGRE

Diagnostic Information

Family Diagnosis and Characterization

During recruitment, families identify the diagnosis they received from their physician or autism specialist. Families with two or more children diagnosed with Autism, PDD-NOS or Asperger’s syndrome are accepted into the program, provided they have not participated in any other autism gene bank. Once enrolled, AGRE conducts an Autism Diagnostic Interview (ADI) and an Autism Diagnostic Observation Schedule (ADOS) to confirm the diagnosis and characterize the affected family members along the autism spectrum.

Currently, the ADI-R provides an empirically-derived algorithm only for the diagnosis of strict Autism. In order to assist researchers in determining whether or not to include data from affected individuals who did not meet the strict ADI-R criteria for Autism, AGRE provides alternate “affected status categories” of NQA (Not Quite Autism) and Broad Spectrum. For more information on these affected status categories, please go to AGRE Affected Status Descriptions.

AGRE uses experienced ADI-R and ADOS raters who travel to the family home to conduct the diagnostic evaluations. These raters have achieved research reliability with a certified trainer and are subsequently monitored on a regular basis by to maintain on-going reliability. The ADI-R and ADOS assessments are videotaped for quality control and for use in diagnostic case reviews. 

The ADI-R and ADOS assessments are videotaped for quality control and use in diagnostic case conferences. Only individuals whose diagnosis is confirmed with autism spectrum disorder are published in the AGRE pedigree catalog.

Other measures conducted during the diagnostic evaluation include the Vineland Adaptive Behavior Scales, the Raven Colored Progressive Matrices, the Peabody Picture Vocabulary Test, and the Stanford-Binet Intelligence Scales, Fifth Edition. Please note that not all data is available for every affected member. In addition to the clinical evaluation, an in-home medical exam, including physicals, neurological exams and detailed medical histories, have been conducted by a pediatric neurologist on a subset of AGRE families. All data collected from these evaluations are entered into the online AGRE phenotypic database.

For further information on AGRE Diagnostic Evaluations, go to:


AGRE Affected Status Categories 

AGRE provides the following three computer-scored affected status categories, which are based on the Autism Diagnostic Interview-Revised (ADI-R) domain scores:


  • Autism
  • Not Quite Autism (NQA)
  • Broad Spectrum


These categories are not “diagnoses” per se but are offered to further categorize subjects based on the ADI, given that the ADI-R provides a validated algorithm only for the diagnosis of strict autism. To meet the criteria for autism on the ADI-R, subjects must meet the required cutoff scores on the domains of social interaction, communication, repetitive behaviors, and age of onset. 

It should be understood that while AGRE uses ADI-R data to derive these affected status categories, the authors of the ADI-R (Rutter, Lord, LeCouteur, 1997) have not provided validated algorithms to identify individuals on the broader autism spectrum (e.g., PDD-NOS, Asperger's Disorder, or other broader phenotypes). As a result, many research groups utilize the ADI-R in combination with other diagnostic instruments, such as the ADOS, to derive clinical diagnoses of PDD-NOS and Asperger's. 

We provide these affected status categories not to limit analyses or impose them upon researchers, but to facilitate analysis by those who may not be comfortable interpreting the ADI-R data to formulate their own diagnoses. In recognizing that there are many different ways to approach diagnostic algorithms, AGRE has provided the full ADI-R data so that individual investigators can apply their own criteria. In addition, we have provided ADOS scores and other clinical data for many of our subjects, so that researchers may consider these data when clarifying diagnoses along the autism spectrum. 

Because AGRE understands that some researchers prefer to have affected status categories pre-assigned, AGRE has provided such classifications for the AGRE sample. In addition to full Autism, AGRE has assigned the affected status categories of NQA and Broad Spectrum so that researchers can easily identify family members on the spectrum who show potential value for inclusion in genetic or behavioral studies of autism. Please note that these latter two categories are not DSM-IV diagnoses, but rather, diagnostic classifications that would require further clinical review in order to establish a final clinical diagnosis.

The following is a summary of AGRE's current affected status categories: 

  1. “Autism” is identified using the well-validated ADI-R scoring algorithm (Rutter, Lord, LeCouteur, 1997). 
  2. “NQA (Not Quite Autism)” represents individuals who are no more than one point away from meeting the autism cutoffs on any or all of the three "content" domains (i.e. social, communication, and/or repetitive behaviors), plus meeting the autism cutoff on the “age of onset” domain; or, individuals who meet the autism cutoffs on all three "content" domains, but do not meet the autism cutoff on the "age of onset" domain.
  3. “Broad Spectrum” defines individuals who show symptoms along the spectrum of pervasive developmental disorders. This is a broad diagnostic category that encompasses individuals ranging from mildly- to severely-impaired. This category potentially includes such pervasive developmental disorders as PDD-NOS and Asperger's disorder, which are used in many genome scans; however, this classification is not based on any validated algorithms and may include individuals who do not meet full criteria for any DSM-IV pervasive developmental disorder. 

PLEASE NOTE: Because a number of the subjects included in the Broad Spectrum category have relatively low domain scores and may not meet the DSM-IV diagnostic criteria for an autism spectrum disorder, we recommend that researchers take a close look at the domain scores of each subject (along with the ADOS results, if available), before deciding whether or not to include them in their subject pool.