Wales Autism Research Centre, School of Psychology, Cardiff University, Cardiff, UK.
Cardiff University Centre for Human Developmental Science, School of Psychology, Cardiff University, Cardiff, UK.
Mol Autism. 2024 Jun 6;15(1):24. doi: 10.1186/s13229-024-00603-7.
Brief questionnaires that comprehensively capture key restricted and repetitive behaviours (RRBs) across different informants have potential to support autism diagnostic services. We tested the psychometric properties of the 20-item Repetitive Behaviours Questionnaire-3 (RBQ-3), a questionnaire that includes self-report and informant-report versions enabling use across the lifespan.
In Study 1, adults referred to a specialised adult autism diagnostic service (N = 110) completed the RBQ-3 self-report version, and a relative or long-term friend completed the RBQ-3 informant-report version. Clinicians completed the abbreviated version of the Diagnostic Interview for Social and Communication Disorders (DISCO-Abbreviated) with the same adults as part of the diagnostic process. For half of the assessments, clinicians were blind to the RBQ-3 ratings. We tested internal consistency, cross-informant reliability and convergent validity of the RBQ-3. In Study 2, a follow-up online study with autistic (N = 151) and non-autistic (N = 151) adults, we further tested internal consistency of the RBQ-3 self-report version. We also tested group differences and response patterns in this sample.
Study 1 showed good to excellent internal consistency for both self- and informant-report versions of the RBQ-3 (total score, α = 0.90, ω = 0.90, subscales, α = 0.76-0.89, ω = 0.77-0.88). Study 1 also showed cross-informant reliability as the RBQ-3 self-report scores significantly correlated with RBQ-3 informant-report scores for the total score (r = 0.71) and subscales (r= 0.69-0.72). Convergent validity was found for both self and informant versions of the RBQ-3, which significantly correlated with DISCO-Abbreviated RRB domain scores (r = 0.45-0.54). Moreover, the RBQ-3 scores showed significantly weaker association with DISCO -Abbreviated scores for the Social Communication domain, demonstrating divergent validity. Importantly, these patterns of validity were found even when clinicians were blind to RBQ-3 items. In Study 2, for both autistic and non-autistic groups, internal consistency was found for the total score (α = 0.82-0.89, ω = 0.81-0.81) and for subscales (α = 0.68-0.85, ω = 0.69-0.85). A group difference was found between groups.
Due to the characteristics and scope of the specialist autism diagnostic service, further testing is needed to include representative samples of age (including children) and intellectual ability, and those with a non-autistic diagnostic outcome.
The RBQ-3 is a questionnaire of RRBs that can be used across the lifespan. The current study tested its psychometric properties with autistic adults without intellectual disability and supported its utility for both clinical diagnostic and research settings.
全面涵盖不同信息提供者之间关键受限和重复行为 (RRB) 的简短问卷具有支持自闭症诊断服务的潜力。我们测试了 20 项重复行为问卷-3（RBQ-3）的心理测量特性，这是一个包括自我报告和信息提供者报告版本的问卷，可在整个生命周期内使用。
在研究 1 中，专门的成人自闭症诊断服务机构转介的成年人（N=110）完成了 RBQ-3 自我报告版本，而亲属或长期朋友则完成了 RBQ-3 信息提供者报告版本。临床医生在诊断过程中使用相同的成年人完成了简短版诊断访谈社交和沟通障碍 (DISCO-Abbreviated)。对于一半的评估，临床医生对 RBQ-3 评分不知情。我们测试了 RBQ-3 的内部一致性、跨信息提供者的可靠性和收敛有效性。在研究 2 中，对自闭症（N=151）和非自闭症（N=151）成年人进行了后续的在线研究，我们进一步测试了 RBQ-3 自我报告版本的内部一致性。我们还在该样本中测试了组间差异和反应模式。
研究 1 显示 RBQ-3 的自我报告和信息提供者报告版本都具有良好到优秀的内部一致性（总分，α=0.90，ω=0.90，分量表，α=0.76-0.89，ω=0.77-0.88）。研究 1 还显示出跨信息提供者的可靠性，因为 RBQ-3 的自我报告分数与 RBQ-3 信息提供者报告分数显著相关，总分（r=0.71）和分量表（r=0.69-0.72）。RBQ-3 的自我和信息提供者版本都具有收敛有效性，与 DISCO-Abbreviated 的 RRB 域评分显著相关（r=0.45-0.54）。此外，RBQ-3 评分与 DISCO-Abbreviated 的社交沟通域评分的相关性较弱，表明具有发散有效性。重要的是，即使临床医生对 RBQ-3 项目不知情，这些有效性模式仍然存在。在研究 2 中，对于自闭症和非自闭症组，总分（α=0.82-0.89，ω=0.81-0.81）和分量表（α=0.68-0.85，ω=0.69-0.85）的内部一致性均成立。在组间发现了差异。
由于专门的自闭症诊断服务机构的特点和范围，需要进一步测试以包括具有代表性的年龄（包括儿童）和智力能力的样本，以及具有非自闭症诊断结果的样本。
RBQ-3 是一种可用于整个生命周期的 RRB 问卷。本研究对无智力障碍的自闭症成年患者进行了心理测量特性测试，并支持其在临床诊断和研究环境中的应用。
