Descriptive Analysis of the B Codes: Understanding Restrictive and Repetitive Behaviors in a Clinical Population

Background:  A primary change to diagnostic criteria in DSM-5 involves requirements for restrictive and repetitive behaviors (RRBs).  While DSM-IV-TR permitted diagnosis for individuals who did not meet any RRB symptom criteria, DSM-5 criteria require that an individual meet at least two of four RRB symptoms.  While proponents argue that the change improves specificity, some research suggests the change may impact diagnoses for individuals who previously qualified for ASD under DSM-IV-TR.

Objectives:  A current challenge for clinicians completing diagnostic evaluations is to better understand RRB diagnostic codes and to properly classify behavioral symptoms.  By investigating the specific symptoms coded under each of the four codes, we investigate patterns of RRBs across individuals who do and do not qualify for ASD diagnosis under DSM-5 criteria.

Methods:  We evaluated 254 individuals presenting for psychological evaluation at an ASD specialty center.  The comprehensive assessment included developmental history, cognitive testing, diagnostic (ADOS2) and adaptive assessment (Vineland-II caregiver interview).  Licensed psychologists completed the evaluations.  Following assessment, psychologist(s) completed a rating form indicating which specific DSM-IV-TR and DSM-5 diagnostic codes the individual met, including (for a subset n = 137) the specific RRB symptoms endorsed under each of the four DSM-5 RRB codes.

Results:  Of these 254 individuals, 234 (92.1%) received a clinical diagnosis of ASD.  All 234 (100%) met criteria on DSM-IV-TR while only 84.2% met DSM-5 criteria for ASD.  Analysis of the 37 individuals who met criteria on DSM-IV-TR  but not DSM-5 (“lost diagnosis”) revealed that the most common symptom profile (54.1%) was one in which individuals met criteria for all three social-communication symptoms (A codes) but only one of RRB symptom (B codes).  We subsequently attempted to characterize the symptom profiles of a subset of individuals (n=137) upon whom specific symptom data for the four B codes were available. 

Among individuals who met DSM-5 criteria (n=107), the modal number of B codes endorsed was 4 (range 2 – 4, mean = 3.13) and the average number of RRB symptoms endorsed was 4.77 (range 2 – 12).  Among individuals who “lost diagnosis” under DSM-5 (n=20), 11 (55%) lost specifically because they had only one B code but all three A codes.  In this group, the vast majority (73%) met RRB criteria for excessively circumscribed or perseverative interests.  In contrast, the same symptom was scored in only 26% of the individuals who met DSM-5 criteria.  Individuals who met under DSM-5 were most likely to score for adverse response to specific sounds or textures, simple motor stereotypies, and insistence on same routine or food. 

Conclusions:  Consistent with prior research, results suggest that new requirements for RRBs may impact diagnosis of ASD under DSM-5 criteria.  Results suggest differences in RRB symptom patterns which may help to identify those at greatest risk for failure to meet DSM-5 criteria and lend support to concerns about the impact on individuals who were more likely to be diagnosed with Asperger Syndrome or PDD:NOS under DSM-IV-TR .