The study was approved by the University of Reading Research Ethics Committee (UREC 18/28). We used the techniques suggested by Mays and Pope [18] to ensure the quality and rigour of the study, and followed the COREQ checklist (see Online Resource 1) for explicit and comprehensive reporting of qualitative studies [19].
Recruitment and participants
Participants were recruited through two large mixed state secondary schools in Berkshire, UK, as part of a wider study, including whole school screening for anxiety and depressive disorders (Radez et al. under review). The process of recruitment for the current study is outlined in Fig. 1, and described in more detail in Online Resource 2.
Of 26 adolescents (aged 11–18) who took part in the diagnostic assessment, 24 met diagnostic criteria for an anxiety and/or depressive disorder. These adolescents and their parents/carers were invited to take part in qualitative interviews. Although the primary focus of the study was adolescents’ perceived barriers/facilitators, their parents were also invited to take part in a separate qualitative interview for the purpose of data triangulation. Each participant (adolescent and parent) provided written consent to take part in the interview and to allow the researcher to audio record the interview. If the young person was under 16 years, they provided written assent and their parent provided written consent. In total, 22 adolescents and 20 of their parents took part in the qualitative interviews. The lead researcher (JR) conducted all interviews with adolescents and parents separately, and all interviews were conducted within one session. During qualitative interviews adolescents and parents also reported other diagnoses (e.g., autism spectrum disorder, gender dysphoria, and physical conditions), which have not been assessed during the diagnostic assessment. Adolescents were interviewed one-to-one in a quiet, private room in their school, and parents were interviewed over the phone at a time that was convenient for them. In 18 cases, parent interviews were conducted with adolescents’ mothers and in two cases with adolescents’ fathers. Two parents/carers did not take part without giving any reason. Each family that took part in the qualitative interview was given £10 voucher to reimburse them for their time. Adolescents’ demographic and clinical characteristics are outlined in Table 1.
Table 1 Adolescents’ demographic and clinical characteristics Of the 22 adolescents, 16 (72.7%) identified as White-British and 6 (27.3%) as other varied ethnic groups. Seven (31.8%) adolescents and/or their parents also reported that a young person had additional physical or mental health difficulties (e.g., chronic physical illness, autism spectrum disorder, attention deficit hyperactivity disorder, dyslexia, Tourette syndrome, and gender dysphoria) that had been diagnosed by other professionals.
Measures
Questionnaire measures
Revised Child Anxiety and Depression Scale, Child Version—RCADS-C [20]. The RCADS-C is a 47-item self-report questionnaire measure of symptoms of anxiety and low mood in young people, aged from 8 to 18. The questionnaire consists of six subscales that correspond to DSM-IV anxiety/depressive disorders—separation anxiety disorder (SAD), social phobia (SP), obsessive–compulsive disorder (OCD), panic disorder (PD), generalised anxiety disorder (GAD), and major depressive disorder (MDD). Respondents rate how often each item applies to them, using a four-level scale from 0 (‘never’) to 3 (‘always’). The RCADS demonstrates favourable psychometric characteristics when applied in various settings (e.g., clinic and community) and in different countries [21]. In the current study, subscale scores and anxiety/depression/total scores and standardised T scores were calculated using syntax provided on the author’s website. A T score of > 70 indicated a clinically significant level of anxiety/depression symptoms. Adolescents’ scores on anxiety total scale and on six subscales were used to identify participants for the diagnostic assessment.
Moods and Feelings Questionnaire, Child Version MFQ-C [22]. The MFQ-C is a 33-item self-report screening tool for depression in children and young people, aged between 6 and 17. Respondents are asked to report how they have been feeling or acting in the past 2 weeks. For each item, they can respond with ‘not true’ (0), ‘sometimes’ (1), or ‘true’ (2). Research studies suggest that the MFQ provides reliable and valid measure of depression in children and young people in both clinical and community samples [23]. In the current study, the MFQ-C total score was calculated by summarising participants’ responses to all 33 items. Based on previous research [24], we used the cut-off score > 26 to identify adolescents with a clinically significant level of depressive symptoms.
Help-seeking questions Each adolescent was asked three questions about seeking/accessing professional help in the last 12 months. Adolescents reported whether they (1) had spoken to a professional (e.g., teacher or GP) about feeling anxious/depressed in the last 12 months, (2) had received any support from a professional to help them with difficulties with anxiety/depression in the last 12 months, and (3) felt that they would benefit from professional support for anxiety/depression. Adolescents’ responses to these three questions were used to purposively sample participants for diagnostic assessments.
Diagnostic interviews
The following diagnostic assessments were administered to identify participants who met diagnostic criteria for an anxiety and/or depressive disorder and were therefore eligible for the qualitative study. All interviews were administered by the first author (JR), trained to reliably deliver the diagnostic assessments. Following assessment, each case was discussed in diagnostic supervision with co-author (FO), who has extensive experience of delivering, training, and supervising these diagnostic tools. Agreement between JR and FO was excellent [presence/absence of diagnoses, κ = 0.820, Clinical Severity Rating (CSR) rating, ICC = 0.956].
Anxiety Disorder Interview Schedule—Child Version—ADIS-IV-C [25]. The ADIS-IV-C is a standardised diagnostic interview, based on the DSM-IV-TR designed to assess anxiety and other disorders in children and adolescents. In the present study, the anxiety sections of the ADIS-IV-C were used to determine whether the adolescent met diagnostic criteria for any anxiety disorder. Minor adaptations to the interview schedule were made, so the diagnoses were assigned based on the DSM-5. If the adolescent met symptom criteria for a diagnosis, then the assessor would assign a Clinician Severity Rating (CSR), ranging from 0 to 8; a CSR of 4 or more would indicate that the young person met criteria for diagnosis. The diagnosis with the highest CSR was considered as the primary diagnosis. Studies using the ADIS-IV-C provide strong evidence for its good psychometric characteristics, which has been especially the case for the anxiety section [26]. Furthermore, ADIS-C provides reliable and valid information even when administered with child only, and reliability of child report is especially high for older children/adolescents [27].
Kiddie Schedule for Affective Disorders and Schizophrenia—Present and Lifetime Version—K-SADS-PL [28]. The K-SADS-PL is a semi-structured interview for affective disorders and schizophrenia, based on DSM-5. In the present study, the depression and mania sections of the K-SADS-PL child interview were used to determine the presence of depression in adolescents. The diagnosis of the major depressive disorder (MDD) was assigned if a young person met at least five criteria for MDD. In addition, CSR scores were assigned in a similar way as the ADIS-C to provide a comparable estimate of the symptom severity. K-SADS-PL is a diagnostic interview with favourable psychometric characteristics, and is recommended over ADIS in terms of identifying mood disorders in young people [29], with adolescent self-report being particularly informative and reliable [30].
Qualitative interviews
The interview topic guides (see Online Resource 3) were developed by the first author (JR), with input from co-authors (TR and PW), drawing on findings of a recent systematic review on barriers/facilitators to seeking professional help for mental health problems in young people [8] and interview guides used in previous similar studies [31]. Areas of inquiry and sample questions for adolescent interview are outlined in Table 2. Although areas of inquiry for parent and adolescent interview were similar, adolescent responses partially guided their parent’s interview. Prior to the data collection, interview questions were piloted with two families (two adolescent girls and their mothers) to help pace the interview and to test the appropriateness of the questions.
Table 2 Areas of inquiry and sample questions from adolescent interview topic guide The semi-structured interviews were conducted by the first author (JR), a female PhD student in psychology, trained in qualitative research methods and with a background of working in mental health research settings. As JR also conducted the diagnostic assessments with the adolescents and their parents, she had already established a relationship with them; this may have helped them feel more at ease and able to open up, but may also have affected what information they gave in the interview. As English is not the interviewer’s first language, during each interview, she frequently summarised information provided by the participant to ensure that her understanding was accurate [32]. Field notes and initial ideas were written after each interview, and used to partially guide the remaining interviews. All 42 interviews were audio-recorded and transcribed verbatim by JR. Adolescent interviews ranged from 13 to 48 min (M = 28:17, SD = 8:06), and parent interviews from 14 to 77 min (M = 35:36, SD = 14:05).
Data analysis
Data analysis started, while data collection was ongoing. Data were analysed by the lead researcher (JR), following six phases of the reflexive thematic analysis [33, 34]. We approached the data from an essentialist/realist epistemological orientation, which draws on the experiences, meanings, and the reality of participants. We analysed the dataset inductively (directed by the content of the data) and semantically (reflecting the explicit content of the data). JR familiarised herself with the data by listening to the audio recordings and transcribing the interviews. During transcribing, all identifiable information was removed and participants were given pseudonyms. Adolescents’ interview transcripts were coded following guidance by Saldana [35]. Data were managed and stored using software NVivo, Version 12, QSR International Pty Ltd [36]. Coding was iterative and cyclical, and systematic over all adolescent interview transcripts (i.e., giving full and equal attention to each aspect of the dataset and coding for implicit and explicit contents). Coding was led by JR with regular discussion/input from other team members (TR, PW) with qualitative expertise, to reflect on the coding process. Although all 22 adolescent interviews were coded, additional data did not contribute to new codes after the first 15 transcripts, and therefore, we judged data saturation to have been reached. After all adolescent interview transcripts were coded, JR coded parent interviews using the final set of codes identified in adolescent interviews (i.e., a ‘top–down’ approach). Notably, as parents were interviewed only for the purpose of the data triangulation for the current study, only sections relevant to the research question were coded and analysed. Adolescent and parent interviews were treated as separate datasets, and JR especially looked for elements in the parent dataset that appeared to contradict or was not contained in the adolescent dataset. JR then organised the final set of codes into preliminary themes and subthemes that explained the vast majority of the adolescent and parent perspective. Themes and subthemes were reviewed and revised by regular discussion with other research team members (TR, PW, and CC) to develop the final set of themes/subthemes. During these discussions, the research team also reflected on the lead researcher’s and the whole research group’s prior assumptions and knowledge in the field of help-seeking. Finally, JR produced a report of the analysis by elaborating identified themes and subthemes and using data extracts (quotes) related to the research question.