Results of Psychometric Testing of the RADS-2 With School-Based Adolescents Seeking Assistance for Sexual Orientation and Gender Identity Concerns. Part 2: Research Brief

Journal of Child and Adolescent Psychiatric Nursing, Aug 2009 by Weber, Scott

PROBLEM: First-stage measures of depressive symptoms need to demonstrate high levels of accuracy in capturing all the facets of depressive illness in order to be maximally useful for nurses performing psychosocial assessments and for nurse researchers.

METHODS: Reliability, validity, and factor analysis testing were conducted on the Reynolds Adolescent Depression Scale (RADS-2). Factor analysis testing examined each item of the instrument to identify the clinical characteristic it measures as well as the broad profiles of symptoms the instrument reveals.

FINDINGS: Four factors were identified and all appear to demonstrate adequate internal consistency reliability.

CONCLUSIONS: Psychometric results demonstrate a firm foundation for nursing clinical and research uses for the RADS-2.

Search terms: Adolescent mental health, depressive symptoms, factor analysis, RADS-2, reliability, psychometric instrument testing, validity

The past three decades have witnessed a growing interest by psychiatric-mental health nurses and researchers in the examination of depression and depressive symptoms in children and adolescents. This follows an extended earlier period during which child and adolescent depression was viewed as either nonexistent or "masked." There was conflicting evidence as to whether adolescent depression was expressed in symptoms similar to or different from those of adults. The latter perspective was primarily expressed in the constructs of "depressive equivalents" (Cytryn & McKnew, 1980; Glaser, 1967; Hollon, 1970; Rie, 1966) in which depression was demonstrated by overt behaviors such as acting-out, aggression, hyperactivity, and delinquency. Other researchers (Carlson & Straber, 2002; Crist, Adler, Isacoff, & Gershansky, 2001; Reynolds, 1995; Yanchyshyn & Robbins, 2000) and the American Psychiatric Association (2000), however, assert that clinical depression is manifested by the same symptoms in adolescence as it is in adulthood. More research with different samples and diverse contexts may help to uncover the true expressed nature and varieties of adolescent depression.

Epidemiological investigations of depressive symptoms suggest that depression is a significant mental health problem among adolescents, and especially among youth who are experiencing personal adjustment challenges related to their sexual orientation or gender identity (Garofalo, Wolf, Kessel, Palfrey, & DuRant, 1998; Martin, 2002; Rivers, 2000; Uribe & Harbeck, 2001).

Using cutoff scores on the Reynolds Adolescent Depression Scale (2nd Edition) and the Beck Depression Inventory (BDI, 2nd Edition) (Beck, Ward, Mendelson, Mock, & Erlbaugh, 2000), 12-18% of adolescents expressed a firststage clinical level of depressive symptomatology (Reynolds, 2002). Although this is not evidence of a formal diagnosis of clinical depression, it does suggest that a significant number of adolescents demonstrate problematic levels of psychological distress. In a study using the original BDI with 300 high school students, Kaplan, Hong, and Weingold (2004) reported that 22% of scores were in the mild to severe range of depression, while Teri (1992) found that, among 568 adolescents, 32% obtained scores in the moderate to severe range on the BDI. In a study of depression in adolescents, Albert and Becklin (2005) administered a revised form of the BDI to 63 7th and 8th grade parochial students in Philadelphia. These authors reported that 36.5% of the students obtained scores in the moderate to severe range.

Purpose of the Study and Report of Current Results

The purpose of this study was to psychometrically test the performance of the RADS-2 in a sample of youth who had sought help from school staff with problems relating to gender identity or sexual orientation during a 1-year period. The purpose was not to screen the sample for prevalence of depressive symptoms, but rather to examine reliability, validity, and factor structure of the RADS-2 with a unique sample of youth. Data were obtained during March and April 2006 from a new sample of school-based adolescents in a large public school district in the Midwestern United States. The sample included 157 adolescent girls and 150 adolescent boys from a broad range of socioeconomic backgrounds. Ethnic and racial profiles of the sample were as follows: 113 (non-Hispanic) Whites, 102 (non-Hispanic) Blacks, 80 Hispanic students of any race, and 12 Asian American adolescents. The subjects ranged in age from 13 to 18 years, and all students were in grades 7-12. The sample was selected from high school students who had sought assistance from school-based nurses, school nurse practitioners, schoolbased psychiatric-mental health nurse practitioners, school psychologists, or counselors, regarding concerns about their sexual orientation and gender identity, and who had completed the RADS-2 in the course of their initial baseline assessment. A school district research registry that was searchable by diagnostic area was used to identify students who had indicated clinical concerns with their sexual orientation or gender identity during the previous 1 year, and all of these students' RADS-2 data were included in the study sample. Unfortunately, "sexual orientation and gender identity" concerns comprised a single classification category in the research registry, so data analysis for each group was not possible. Subjects were enrolled in twenty of the school district's 29 junior and senior high schools, and the total sample of 307 students comprised 0.8% of the total enrollment (40,824) in grades 7-12. The study was considered a secondary data analysis and obtained expedited review and approval from the collaborating university's health sciences institutional review board, as well as the research coordinating review board of the public school district