Understanding Your Results

A CATI Total Score ≥ 134 indicates an elevated likelihood that an individual may be autistic.

In simple terms, a score of 134 or more on the CATI suggests that you are experiencing autistic traits at a higher rate than would be expected for a non-autistic individual. This score was chosen because it does a pretty good job at correctly identifying people who are autistic (sensitivity) and those who are not (specificity). However, some autistic people obtain total scores < 134 on the CATI, and some non-autistic people get scores ≥ 134. The CATI was not designed as a standalone tool for diagnosing autism, but it can help clarify if you experience certain traits at levels that are more similar to autistic or non-autistic adults.

The Comprehensive Autistic Trait Inventory (CATI) uses a cutoff score. This score is used to differentiate between people who might be autistic and those who are not. A score below the cutoff score indicates a lower likelihood that a person is autistic, and a cutoff score equal to or above the cutoff indicates an increased likelihood that a person is autistic.

The cutoff score was determined using a mathematical equation that aimed to minimize the amount of specific kinds of errors made when using the results of the test to predict whether an individual is autistic or not. Now, let's talk about sensitivity and specificity, which are just fancy terms for how accurate the CATI is.

Sensitivity (82.71%): This is about how well the CATI can correctly identify people who are autistic. Imagine that 100 people who are actually autistic take this test. If the CATI has a sensitivity of 82.71%, it means that about 83 of these 100 people will get a score above 134, correctly indicating that they have a significant number of autistic traits. However, that also means 17 people who actually have autism would be missed (i.e., incorrectly classified as allistic).

Specificity (79.00%): This is about how well the CATI can correctly identify people who are not autistic. Let's say another group of 100 people take the test, none of whom are autistic. With a specificity of 79%, about 79 out of these 100 people will score below 134, correctly showing that they have fewer autistic traits than the average autistic adult. However, that also means that about 21 people who do not actually have autism would be incorrectly categorized as being autistic.

So, in simple terms, a score of 134 or more on the CATI suggests the presence of autistic traits at higher rates than are typically seen in non-autistic individuals. This score was chosen because it does a pretty good job at correctly identifying people who are autistic (sensitivity) and those who are not (specificity). Remember, the CATI is not for diagnosing autism, but it helps in understanding where you might stand in terms of autistic traits.

For the CAT-Q, scores in the following ranges suggest an increased likelihood that a person might be autistic:

• A CAT-Q Total score ≥ 110

• A Compensation score ≥ 35

• An Assimilation score ≥ 40

Masking scores tend to be similar for both autistic and non-autistic individuals, so a higher or lower score on this scale does not mean you are more or less likely to be autistic.

Generally speaking, autistic individuals report higher scores on the CAT-Q Total core, Compensation, and Assimilation scales than non-autistic individuals, with female-identifying autistic individuals reporting the highest average scores in all areas. However, it is important to know that CAT-Q is a relatively new measure and that information about the diagnostic accuracy of the CAT-Q is not yet available. Due to this limitation, it is impossible to know what percentage of people are incorrectly categorized as autistic or non-autistic based solely on their CAT-Q scores. Additionally, social camouflaging is not unique to autism, and high rates of social camouflaging are also reported by non-autistic individuals with a variety of conditions (e.g., social anxiety, ADHD, etc.). The cut-off scores provided above are only general recommendations, and higher scores on the Total score, Compensation scale, or Assimilation score do not necessarily mean you are autistic, but they may merit further evaluation.

To interpret your CAT-Q scores, compare your score for each scale with the average score for the autistic and non-autistic groups for the gender you more closely align with. Non-binary, transgender, or otherwise gender expansive folx should compare their scores to the combined group as it includes male, female, and non-binary participants. While the study in which the CAT-Q was developed did include some non-binary individuals, there were not enough non-binary participants to yield reliable data for the non-binary group alone.

Not all autistic people engage in elevated rates of social camouflaging. However, those who do often go undetected by traditional autism assessments because their compensation strategies are so effective that their autistic traits are missed if social camouflaging isn’t directly assessed. Individuals with a high CAT-Q Total score report using camouflaging strategies to a higher degree than other individuals. This is often associated with lower scores (e.g., scores outside of the autistic range) on other tests for autism. Each subscale measures a different aspect of social-camouflaging.

Compensation refers to the use of strategies to compensate for social and communication difficulties.

Assimilation represents attempts to blend-in to social situations in which the individual is uncomfortable, without letting others see this discomfort.

Masking refers to strategies used to hide autistic characteristics or to portray a non-autistic persona.

In the initial study in which the CAT-Q was developed, there were significant differences between autistic and non-autistic individuals on the Compensation and Assimilation scales. The Masking scale demonstrated the smallest difference between autistic and non-autistic samples, suggesting masking may be less specific to autism than the other components of camouflaging, and may reflect more general self-presentation or impression-management strategies applied to autistic characteristics in both autistic and non-autistic individuals.

Limitations. The CAT-Q is a newer measure, and comprehensive information about its diagnostic accuracy (e.g., specificity, sensitivity, etc.) has not yet been publicly released. Due to this, it should be assumed that many people are incorrectly categorized as autistic or non-autistic based solely on their CAT-Q scores. Additionally, social camouflaging is not unique to autism, and high rates of social camouflaging are also reported by non-autistic individuals with a variety of conditions (e.g., social anxiety, ADHD, etc.). As such, the cut-off scores provided above should only be considered as general recommendations. Higher scores on the Total score, Compensation scale, or Assimilation score do not necessarily mean you are autistic, but they may merit further evaluation. Similarly, not all autistic individuals report elevated rates of social camouflaging, so low scores on the CAT-Q do not rule out a diagnosis of autism spectrum disorder.

An ASRS Total score of 14 or higher is suggestive of ADHD.

The ASRS uses a cutoff score. This score is used to differentiate between people who might have ADHD and those who likely do not. A score below the cutoff score indicates a lower likelihood that a person has ADHD, and a cutoff score equal to or above the cutoff indicates an increased likelihood that a person has ADHD.

The cutoff score was determined using a mathematical equation that aimed to minimize the amount of specific kinds of errors made when using the results of the test to predict whether an individual has ADHD or not. Now, let's talk about sensitivity and specificity, which are just fancy terms for how accurate the ASRS is.

Sensitivity (91.4%): This is about how well the ASRS can correctly identify people who have ADHD. Imagine that 100 people who have a confirmed diagnosis of ADHD take this test. If the ASRS has a sensitivity of 91.4%, it means that about 91 of these 100 people will get a score above 13, correctly indicating that they likely have ADHD. However, that also means about 9 people who actually have ADHD would be missed (i.e., incorrectly classified as not having ADHD).

Specificity (96%): This is about how well the ASRS can correctly identify people who do not have ADHD. Let's say another group of 100 people take the test, none of whom have ADHD. With a specificity of 96%, about 96 out of these 100 people will score below 14, correctly showing that they have fewer ADHD symptoms than the average adult with ADHD. However, that also means that about 4 people who do not actually have ADHD would be incorrectly categorized as having ADHD.