Efficient Sufficiency: A qualitative evaluation of a 1 year pilot study of young people and parents accessing a mental health drop‐in centre in a paediatric hospital

Abstract Background Children and young people with long‐term physical health conditions (LTC) are known to have higher levels of co‐morbid mental health problems than medically healthy children. Evidence‐based treatments for mental health problems are effective in children who also have an LTC. This study aimed to explore the factors associated with participants' perceived acceptability and impact of a transdiagnostic mental health centre offering brief psychological assessment and treatment for children and young people and/or their families with mental health needs in the context of long term physical conditions. Methods One‐hundred twenty‐eight patients attending the drop‐in centre were invited to participate. Overall, 35 participated (31 parents/carers; 4 children and young people) in semi‐structured interviews (either in person or by phone) exploring their experience of the centre. Interviews were audio‐recorded, transcribed and checked. Framework analysis was then conducted on all transcripts. Results Overall, participants found the drop‐in centre highly acceptable and reported a positive experience. Reasons for this varied but broadly focused around four themes: (1) efficient sufficiency; (2) autonomy; (3) fusion of process and content factors and (4) (dis)parities of esteems and ‘seeing both sides of the coin’. Conclusions Participants found the intervention acceptable. A mental health drop‐in centre in a paediatric hospital appears to be a positive and valued adjunct to supplement existing mental health services.


| INTRODUCTION
Children and young people with long-term physical health conditions (LTC) are known to have higher levels of psychiatric co-morbidity than medically healthy children (Zhou et al., 2019). Psychiatric comorbidities are known to negatively impact the young person's quality of life (Baca et al., 2011) and to be expensive for health services (Zima et al., 2016). Whilst evidence-based treatments for mental health problems in children exist and are effective in those who also have a LTC (Moore et al., 2019), many young people with LTCs are not able to access mental health services (Children's Commissioner, 2016), and those that are may not receive evidencebased intervention (Welch et al., 2018).
To increase access, brief CBT interventions have been developed based on traditional CBT 'high-intensity' protocols consisting of 12-16 sessions and evaluated in randomized trials. These brief interventions typically comprise either just a single-session, for example, of psychoeducation, or 6-8 sessions of an intervention, involve technology or other self-help materials and are delivered by health professionals (Lorentzen et al., 2020;Schleider & Weisz, 2017).
Brief psychological interventions have been shown to be effective for common mental health problems in general  and in young people with LTCs (Catanzano, Bennett, Sanderson, et al., 2020).
A drop-in centre in a paediatric hospital delivering brief cognitive behavioural interventions may be one way for more children to access evidence-based treatments. This was the goal of the 'Lucy project': to evaluate the acceptability, feasibility and impact of a 'Mental Health and Psychological Wellbeing Drop-in Centre' in a tertiary paediatric hospital setting  and the quantitative outcomes are reported elsewhere Clarke et al., 2022). The project was an uncontrolled trial of young people, their siblings and carers attending a national paediatric hospital. A 'drop-in' booth served both as a focus for recruitment and for raising awareness of the project. Once families had consented and completed baseline measures, an initial triage assessment was carried out by newly qualified clinical psychologists, trained psychological well-being practitioners (i.e., individuals trained specifically in low-intensity therapies through a specific programme as part of UK Improving Access to Psychological Therapies initiative) or a junior doctor. All participants were then discussed in a weekly triage meeting with a consultant child and adolescent psychiatrist and allocated to an intervention.
Participants were allocated to (i) provision of/direction to selfhelp materials and/or online resources, (ii) further assessment in the form of either a neurodevelopmental assessment and/or computerized mental health diagnostic assessment (the Development and Wellbeing Assessment), (iii) signposting/referral to appropriate internal or external services (including mental health services for adults if the parent had significant symptoms of anxiety and depression) and (iv) a brief modular psychological intervention defined as up to 6 sessions (6 hours total) of either telephone or face-to-face (videoconferencing software was not used in this study) guided self-help. These categories were not mutually exclusive and participants could be allocated to more than one intervention.
As part of the Lucy Project's pilot phase , a qualitative study was conducted aiming to explore the factors associated with participants' (i.e., young people with long-term conditions and their siblings/parents/carers) perceived acceptability and impact of a transdiagnostic mental health centre offering brief psychological assessment and treatment for children and young people and/or their families with mental health needs in the context of long term physical conditions. 2 | METHODOLOGY 2.1 | Design cases, this was over the phone, but in some cases, the 6-month follow-up was conducted face to face. If they expressed interest, informed consent was then taken, and participants were interviewed either in person or by phone according to their preference. Interviews were semi-structured and adapted from a schedule devised by the study team from a previous qualitative study .
This included both key questions and prompts, which interviewers could use to help participants share their experience (Table S1). At the end of each interview, a series of questions were asked about the experience of the interview itself. The interview schedule was piloted with one participant, and no changes were made as a result. This pilot interview is aggregated with the full dataset. In cases where more than one member from the same family was interviewed, we interviewed the parent and child separately (n = 4). The interviews were conducted either by phone (n = 31) or in some cases face to face (n = 4) in clinic rooms at the hospital depending on patient preference. The only people present were the researchers and the participants. Interviews were conducted by MC (male doctor and PhD student, working on the research project as a researcher involved in delivering the intervention) and/or KF (female undergraduate placement student in psychology, working on the research project as a researcher not involved in delivering the intervention). In cases where MC had been the primary therapist for a particular participant, KF would carry out the interview so as to limit bias.

| Analysis
All interviews were audio-recorded, transcribed (anonymizing patient data) and checked. Framework analysis (Gale et al., 2013;Ritchie & Spencer, 1994) was applied to all transcripts using NVivo 12 (QRS International Pty Ltd). This involved the following steps: familiarization, identifying a thematic framework, indexing, charting, mapping and interpretation (Ritchie & Spencer, 1994). We primarily used an inductive ('bottom-up') approach, whereby code development was driven by the data. A priori questions around barriers/facilitators and study aims were also used to develop the initial thematic framework.
The two main analysts, MC and KF, met regularly during the process of developing the themes. Any discrepancies or disagreements were discussed with SB. MC's involvement in delivering the intervention conferred some advantages in terms of insight and deeper knowledge of the topic. On the other hand, there are disadvantages to being a part of the intervention being assessed, as it may lead to positive bias; that is, positive comments get focused on, and critical feedback gets minimized. We sought to address this by maintaining curiosity about the data, working closely with members of the research team less involved with delivering the intervention and welcoming a variety of views. MC and LX then re-applied the final themes to the data to check fit. Inter-rater reliability was calculated and was found to be strong (kappa = 0.87). The number of participants who endorsed each theme was then calculated. Outcomes are reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (Tong et al., 2007). Finally, descriptive statistics were carried out using SPSS statistical analysis software (version 25, IBM). Participant demographics and symptom profiles were compared to those of the wider sample who were not interviewed. We conducted respondent validation to ensure the researchers' understanding of the interviews was accurate and whether they agreed and/or had anything to add to the summary. All participants were invited to take part in respondent validation. The two participants who responded said they agreed and the themes did not change as a result of respondent validation.

| Sample
Of the 128 who consented to the study, 35 participants from 31 families took part (in the case of 4 families, the child was interviewed in addition to the parent). Four out of 31 parents/carers were fathers, and one out of four young people was a boy. A breakdown of reasons for non-participation can be seen in Figure 1.
Participants who took part in the qualitative interviews were more likely (49%) to have been allocated to a brief psychological intervention, compared to those who did not (21%) take part (p < 0.05). There were no other significant differences between those who participated and those who did not participate in the qualitative interviews (p > 0.05). Of those where child mental health data were available at both time points, 15 showed 'improvement', 7 'deterioration' and 2 'no change'. Scores on the child mental health measure for the subset who took part in the qualitative study are presented in Table S2. Interviews were on average 24 minutes in length (range: 7-47 minutes). Demographic characteristics can be seen in Table 1.

| Overview of findings
Overall, participants reported that they experienced the intervention as 'really helpful', 'an amazing resource' and 'like it changed their life'.
Responses were organized into the following themes: (1) efficient sufficiency; (2) autonomy; (3) fusion of process and content factors and (4) (dis)parities of esteems and 'seeing both sides of the coin'. The breakdown of themes endorsed by each participant are detailed in Table 2 and by intervention in Table 3.

Efficient sufficiency
The first theme relates to the tension between increasing access and giving participants a sufficient 'dose' of an intervention. Overall, participants reported finding the intervention sufficient and highly efficient. Factors such as being referred, but not being seen by the time the interview was conducted, explained the minority of cases who would have liked to receive 'more'.

Autonomy
The second theme focused on the idea of the intervention being flexible in that participants had the ability to make their own choices and adjustments so that the interventions could be tailored around their lives; for example, they liked being able to receive the intervention by phone or face to face. Participants explained that they learnt how to become their own therapists and that this increased their freedom and ability to make choices in their lives; for example, young people described being able to go to the cinema and see friends following treatment for anxiety.

Flexibility of therapy
Participants (21/35; 60%) really appreciated the flexibility of the intervention. This was true in terms of scheduling appointments around existing commitments, but also the mode of delivery (e.g., phone or face-to-face): Signposting to resources only 8.5 Not allocated to intervention 8.5 Note: Core participant demographics are shown for all participants who were interviewed along with the median and interquartile range (IQR) and percent (%) of cases (where relevant) for all data. NB: Indices of multiple deprivation (IMD) have been developed in England and Wales to encompass material deprivation and aspects such as health, education and crime. An IMD decile is a dimension that places the deprivation scores of individual areas into one of 10 groups of equal frequency, ranging from the 10% most deprived areas (score of 1) to the 10% least deprived areas (score of 10). These can be used to show the relative deprivation level of the area the participant lives in.
T A B L E 2 Themes and subthemes endorsed by each participant and intervention received

Fusion of process and content factors
The third theme highlighted the synergy between process and content factors of the intervention. Families valued both process and content factors. This synergy was best embodied by the therapists who were able to simultaneously be compassionate and professional, but also a source of practical support, with concrete strategies, goals and outcomes.

Content factors
The large majority of participants (32/35; 91%) appreciated the content of the practical resources (e.g., handouts detailing the strategies) that were given to them by the therapist and these were felt to be highly acceptable: The handouts we were getting, the video support, Young people and parents found tracking their progress using goal-based outcomes helpful and reported a sense of achievement when they could see improvements: The questionnaires I think that they should keep doing them as well because they really sort of help you look at them and look back and sort of say well I'm a 1 then but now I've made some progress to a 6. However, some participants did mention that in cases where the child was an inpatient, parents would not feel able to remove themselves and go to a booth, even if it was in the hospital: When a parent is in that situation yes they feel all these feelings but me personally I would not take There were a broad range of opinions on the importance of therapists as 'LTC experts' ranging from a preference for them having no existing knowledge to it being crucial for effective treatment: As long as they are sort of aware of what that will mean to the child, I do not think they need to know the ins and outs of every single illness or diagnosis.  . To the authors' knowledge, this is the first qualitative study to explore participants' experiences of a mental health drop-in centre in a paediatric hospital for young people with long term conditions and their families.

| Acceptability
Overall, participants' experience of rapidly accessing a brief intervention was positive. This was especially the case in those who received a brief psychological intervention (e.g., guided self-help). Conceptually,   (Bennett et al., 2015;Jones et al., 2022), diabetes, bronchiectasis unrelated to cystic fibrosis, and epidermolysis bullosa (Jones et al., 2022). The high levels of acceptability described in the study overall may in part be explained by a high degree of overlap between the intervention model (e.g., co-location, integrated care, self-referral and facilitating onward referral to local services) and the ways families would like tertiary paediatric outpatient clinics to facilitate access to mental healthcare (Jones et al., 2022).

| Impact
Participants reported having choices they did not have before, by for example being able to go for a promotion at work, once the strategies for managing challenging behaviour had been learned. The perceived increase in autonomy (i.e., families feeling more 'free', 'in control' and with more 'choices') may be one way participants were impacted by the intervention. Previous research highlighted in the review by Moore et al. (2019) suggests that empowerment, self-esteem and self-management, may feed into each other to affect a young person's resilience and be influenced through psychological intervention, especially when these include learning skills. This resonates with findings in this study, whereby participants felt that once the strategies had been learnt, they felt able to continue managing difficulties on their own. This may be a particular advantage of guided self-help and CBT interventions, which emphasize learning strategies. Most participants did not report changes to their physical health because of the intervention, although there were some notable exceptions (e.g., a young person with psoriasis). This is in line with quantitative data .

| Limitations
Study limitations mean the results should be interpreted cautiously.
All participants were invited to interview, but in practice, the majority who accepted were participants who had completed a brief intervention (29/35). It is possible that this led to a degree of selection bias, where for those who consented to be interviewed, improvements from the intervention motivated them to take part and 'give back to the project' by taking part in the qualitative study at follow-up. This was counterbalanced by attempting to recruit participants who had not completed the intervention (see Figure 1), and although numbers were smaller (n = 6), they still consisted 17% of the sample. Only a minority of participants interviewed had not started the intervention (n = 3), making comparisons with the rest of sample limited. Overall, there appeared to be some similarities in terms of participants' recognition that integrated care and accessibility were important (even when they did not receive an intervention themselves). The majority of participants interviewed were mothers (27/35), limiting the extent to which our findings can be applied to fathers and young people. As discussed in more detail elsewhere Clarke et al., 2022), the lack of control group is a limitation of the wider study, as improvements may have been due to time or other confounders. We did not use validated measures of service use, which would have captured use of all interventions/support outside the project, as it was felt that questionnaire burden was already high.

| Implications for future research
These findings, in conjunction with analysis of quantitative outcomes Clarke et al., 2022) and the acceptability and feasibility of remotely delivered training (Batchelor et al., 2020), may be used to help guide the design and implementation of future randomized controlled trials evaluating the effectiveness of brief psychological treatments for young people with LTC. This may be done by, for instance, retaining a degree of flexibility in terms of scheduling appointments, using goal-based outcomes and routine outcome monitoring as part of the intervention and offering brief CBT/guided self-help.