Construct validity and parent–child agreement of the six new or modified disorders included in the Spanish version of the Kiddie Schedule for Affective Disorders and Schizophrenia present and Lifetime Version DSM-5 (K-SADS-PL-5)

Abstract

Changes to the Diagnostic and Statistical Manual of Mental Disorders fifth edition (DSM-5) incorporate the inclusion or modification of six disorders: Autism Spectrum Disorder, Social Anxiety Disorder, Intermittent Explosive Disorder, Disruptive Mood Dysregulation Disorder, Avoidant/Restrictive Food Intake Disorder and Binge Eating Disorder. The objectives of this study were to assess the construct validity and parent-child agreement of these six disorders in the Spanish language Schedule for Affective Disorders and Schizophrenia for School Age Children Present and Lifetime Version (K-SADS-PL-5) in a clinical population of children and adolescents from Latin America. The Spanish version of the K-SADS-PL was modified to integrate changes made to the DSM-5. Clinicians received training in the K-SADS-PL-5 and 90% agreement between raters was obtained. A total of 80 patients were recruited in four different countries in Latin America. All items from each of the six disorders were included in a factor analysis. Parent-child agreement was calculated for every item of the six disorders, including the effect of sex and age. The factor analysis revealed 6 factors separately grouping the items defining each of the new or modified disorders, with Eigenvalues greater than 2. Very good parent-child agreements (r>0.8) were found for the large majority of the items (93%), even when considering the sex or age of the patient. This independent grouping of disorders suggests that the manner in which the disorders were included into the K-SADS-PL-5 reflects robustly the DSM-5 constructs and displayed a significant inter-informant reliability. These findings support the use of K-SADS-PL-5 as a clinical and research tool to evaluate these new or modified diagnoses.

Introduction

The semi-structured diagnostic interview Schedule for Affective Disorders and Schizophrenia for School-Age Children, Present and Lifetime Version (K-SADS-PL), which is based in the Diagnostic and Statistical Manual of Mental Disorders (DSM, fourth edition) (APA, 1994; Kaufman et al., 1997), has been validated in its Spanish version (Ulloa et al., 2006). Recently, the Spanish version of the K-SADS-PL was modified in order to integrate the changes made to the fifth edition (DSM-5) (APA, 2013), especially regarding six new or modified disorders: Autism Spectrum Disorder (ASD), Social Anxiety Disorder (SAD), Intermittent Explosive Disorder (IED), Disruptive Mood Dysregulation Disorder (DMDD), Avoidant/Restrictive Food Intake Disorder (ARFID) and Binge Eating Disorder (BED).

There were several changes in the DSM-5, the manner in which disorders are grouped (i.e. ASD), in their diagnostic criteria (i.e. SAD and IED), as well as the addition of new disorders (i.e. DMDD, ARFID and BED) (APA, 2013). Changes for ASD included the removal of a sub-diagnosis, arrangement of diagnostic criteria into two domains (social communication/interaction and restrictive and repetitive behaviors) instead of three and an independent evaluation of symptom severity for each diagnostic domain. These changes have been reported to result in higher specificity, sensitivity and stability for ASD (Frazier et al., 2012; Gibbs et al., 2012; Huerta et al., 2012; Mandy et al., 2012; McPartland et al., 2012; Tahery and Perry, 2012). Changes in the diagnostic criteria for SAD focused on a fear of performance evaluation, making it a generalized disorder rather than a phobia (APA, 2013). The inclusion of these new criteria points towards SAD as a valid diagnosis in children and adolescents (Bögels et al., 2010). Changes in IED diagnostic criteria addressed the lack of definition and frequency of aggressive behavior (APA, 2013); validity studies lead to an improved detection of subjects with recurrent and problematic impulsive aggressive behavior (Coccaro, 2012; Coccaro et al., 2014, 2017). DMDD was incorporated into the DSM-5 in order to provide a diagnostic location for chronic irritability instead of considering it as bipolar non-specific manifestation (Rao, 2014); nevertheless DMDD and bipolar disorder present phenotypes that commonly overlap in clinical settings (Mitchell et al., 2016). Furthermore, DMDD could not be differentiated from Oppositional Defiant Disorder (ODD) in a general population child sample study (Mayes et al., 2016) nor in a clinical research setting (Axelson et al., 2012), which in turn suggests a low diagnostic delimitation of the DMDD construct with other disorders. A previous diagnosis, Feeding Disorder of Infancy and Early Childhood, was rearticulated as Feeding and Eating Disorders (FED) to include subjects of all ages. This FED dimension includes BED and ARFID. An extensive review examining the evidence for BED in reference with DSM-5 diagnostic criteria supports the duration/frequency criterion and emphasizes the importance of loss of control and marked distress as a primary marker around binges (Wilfley et al., 2016). Clinical (Fisher et al., 2014; Nicely et al., 2014) and non-clinical (Kurz et al., 2015) studies have reported that patients with ARFID are distinct from those with other FED. Restrictive eating disturbances in a non-clinical middle childhood sample were evaluated using the Eating Disturbances in Youth-Questionnaire, identifying three variants of early-onset restrictive eating disturbances and weight problems, inadequate overall food consumption, limited accepted amount of food and food avoidance based on a specific underlying fear (Kurz et al., 2016).

Given that these changes have been only recently incorporated and have resulted in large quantitative and qualitative modifications, new validity studies in diagnostic clinical interviews are needed.

Diagnostic clinical interviews often involve summarizing information from both, parents and their children. Investigations using multi informants have shown a moderate to high agreement between parents and adolescents with ASD about behavioral, emotional and functioning problems (Jepsen et al., 2012). There are no specific reports of parent-child agreement for SAD, but across studies that use different structured or semi-structured interviews appeared that children are better informants describing their mental states, somatic symptoms and relational problems than their parents (Reutersköld et al., 2008; Weems et al., 2011). A similar situation occurs with IED. While there are no studies describing parent-child agreement about this disorder, some studies in youths with Disruptive Behavior Disorders (DBD) have shown a low correlation between parents and offspring reports (Nguyen et al., 1994), yet other reports show that parents provide better information than children describing externalized disorders (Edelbrock et al., 1986). Parent-child agreement in the questionnaire on eating and weight patterns showed no concordance in terms of the number or type of binge eating, overeating episodes or compensatory weight control behaviors (Steinberg et al., 2004). No information was found describing parent–child agreement for neither DMDD nor ARFID. In order to understand the variability between parent-child responses, for instance, in terms of age and sex, inter-informant reliability studies for these six new or modified disorders are needed.

The main objectives of the present research were to establish construct validity and evaluate parent-child agreement for the six new or modified disorders included in the Spanish version of the K-SADS-PL-5.