Interventions to improve the mental health of children and young people with long-term physical conditions: linked evidence syntheses

Quality appraisal

Quality appraisal was conducted simultaneously with data extraction using criteria adapted from the Cochrane risk-of-bias tool. 86 In addition to criteria on randomisation, allocation concealment, blinding of assessors and selective outcome reporting from the existing tool, we included items on intention-to-treat analysis, between-group similarities at baseline, dropouts, response rates, intervention details and manuals, adherence, follow-up measures and psychometric properties of outcome measures (see Richardson et al. 107). This gave 15 items on which risk of bias and quality of included articles were assessed, providing additional insight into study quality and reporting. Assessment of quality and risk of bias was performed at the article level to evaluate any differences on account of the different outcomes reported in articles from the same study. Quality appraisal decisions were made by two reviewers (from DM, MN and LS) and disagreements resolved through discussion. The appraisals were used to evaluate risk of bias and study quality and were not used to exclude papers.

Categorisation of interventions and outcomes

Data analysis and synthesis

The principal summary measure used to compare clinical effectiveness findings in included studies was differences in mean between intervention and control group post test and, when available, at the longest follow-up time point. For each outcome, mean, standard deviation (SD) and sample size (or figures that could be used to derive these when available) for the relevant intervention and control groups were used to assess differences between groups. When we did not receive necessary data for outcomes from either articles or correspondence with authors, we did not synthesise these outcomes. When such figures were not available for any CYP mental health outcomes in a study, we synthesised findings narratively. Effect sizes for each study were calculated post intervention using Cohen’s d, that is, the difference between the means in the two groups divided by their pooled SD. 108 Ninety-five per cent confidence intervals (CIs) for the effect sizes were also calculated, along with p-values. Effect sizes and CIs were calculated using the ‘metan’ command in Stata^® v13 (StataCorp LP, College Station, TX, USA). The p-values for each effect size were calculated using the ‘ttesti’ command in Stata. Cohen’s guidelines were used to aid interpretation of effect sizes. 108 Thresholds above which effect sizes are considered to be ‘small’, ‘medium’ and ‘large’ are d = 0.20, d = 0.50 and d = 0.80, respectively. Although these classifications are widely used, it is acknowledged that these guidelines do not take account of how the clinical or practical importance of effects might vary between outcome measures. 109

Meta-analysis was considered feasible if multiple studies examined the same intervention type and used the same outcome category and a similar comparator and participant LTC. Random-effects meta-analysis models were fitted to pool effect sizes across the studies, based on the assumption that no two studies were addressing the same research questions in exactly the same way. We calculated 95% CIs for each pooled effect size estimate. The I² statistic (possible range 0–100%) was used to quantify statistical heterogeneity, with higher values indicating greater heterogeneity. 110 When two or more measures assessing the same outcome category were reported in a study, the effects were combined into a single summary effect for that study, calculating the standard error for this effect using the correlation between the measures obtained from the paper itself or other research. 111 When different studies used identical measures, we conducted additional meta-analysis using raw mean differences. All meta-analyses and associated forest plots were produced using the ‘metan’ command in Stata.

We intended to assess publication bias by examining funnel plots for asymmetry using the ‘metafunnel’ command in Stata. However, we were unable to assess funnel plots properly or use more advanced regression-based assessments to assess publication bias owing to the substantial heterogeneity identified across studies and the small number of studies with similar characteristics that could be entered into a given meta-analysis. 112 Therefore, we cannot comment on publication bias in this review.

Intervention characteristics

Report Supplementary Material 1, Table 5, provides details about the interventions reported in the included studies. The intervention aims and structure, as well as details of the delivery site and personnel, the intended recipients and the comparator group, are shown. Eleven intervention categories were used to group together similar interventions. The label and definition of these intervention categories are shown, along with the studies evaluating them, in Appendix 4. Interventions based on CBT were the most frequently studied (n = 7),83,84,124–126,130,145 with 267 participants randomised to this type of intervention. The two studies reported by Szigethy et al. 84,126 explored the effects of the CBT programme PASCET-PI (Primary and Secondary Control Enhancement Therapy-Physical Illness) on various outcomes in children with IBD84 or Crohn’s disease. 143 Masia-Warner et al. 124 and Reigada et al. 125 used versions of TAPS (treatment of anxiety and physical symptoms), the latter authors modifying the intervention to include management of specific symptoms of IBD. Adapted intervention content was seen in eight studies. 84,119,120,124–126,130,145 All but two of these were trials of CBT, with Bignall et al. ’s119 relaxation intervention and Boardway et al. ’s120 stress management training also including adapted content. For each study, Appendix 5 describes the content of each intervention, and whether it was adapted or delivered flexibly. Only Szigethy et al. 126 specifically allowed flexibility in delivery, allowing for individual changes to content based on the developmental stage of the recipient.

The parenting programmes Triple P118 and its modified subsidiary for families in which a child has a disability and behavioural problems, Stepping Stones Triple P (SSTP),128,131 were assessed in three studies, with 93 parents randomised to these intervention conditions. The study by Whittingham et al. 131 was the only one to include more than one intervention condition, as both SSTP with and without additional ACT featured. Group play therapy featured in three studies,133,135,138 with 44 participants randomised to these treatment arms. Palliative care (n = 49),123,137 emotional intelligence training (n = 40)132,134 and stress management programmes (n = 16)120,122 featured in two studies each.

Progressive muscle relaxation (PMR) techniques featured in three studies. 119,133,139 However, they formed part of only two intervention arms, with 47 participants receiving such interventions. Specifically, Bignall et al. 119 trialled an intervention containing breathing retraining with PMR, asthma education and guided imagery, whereas Yang et al. 139 used 30 minutes of audio-recorded guided muscle relaxation nightly before going to sleep. PMR was also an active comparator in the intervention trialled by Diego et al. ,121 in which the intervention was massage therapy. Sometimes the intervention label used in the study indicated that more than one intervention category might be relevant, for instance Hains et al. ’s122 cognitive restructuring and problem solving and Nekah et al. ’s133 structured cognitive–behavioural group play therapy both suggest CBT. The descriptions of interventions provided in articles were used to assess which category was most suitable.

The most common comparator involved treatment as usual (TAU) or usual care (e.g. regular schooling, usual sedation, usual hospital care) within a given setting (n = 20),83,84,118,120,122–124,127,128,130–139,145 with five of these studies employing a waiting list design. 122,124,127,131,137 In total, 393 participants were randomised to standard or usual care control groups. Active comparators in the remaining six studies (n = 157 participants) included asthma education,119 PMR,121 aerobic exercise129 and non-directive supportive therapy. 125,126 Standard care/TAU was often described only roughly and varied considerably between studies, to the extent that determining common components was not possible. During the waiting period of studies with waiting list designs, all participants received usual care.

Interventions were received by parents alone in three studies,118,128,131 by parents and children together in five studies84,123,125,126,145 and by CYP alone in the remainder. When both parents and CYP received the intervention, the goal was usually to give parents skills to help them encourage positive behaviours at home and develop skills to maintain positive effects beyond the intervention end. This occurred as a component of CBT interventions, with the exception of Lyon et al. ’s123 palliative care intervention, in which the aim was to include the family in decision-making. The delivery of interventions occurred across a variety of settings and through a range of personnel. Twelve studies delivered at least part of their intervention in a hospital setting. 84,118,120,122,123,126,128–130,133,136,145 Five interventions took place in other medical centres or university outpatient departments,83,124,125,127,135 three were delivered in school settings119,132,134 and one at home. 139 In Shoshani et al. ’s137 study, which assessed the Make-A-Wish (Make-A-Wish Foundation^® UK, Reading, UK) intervention, children’s wishes were ascertained in their own home, but the nature of their delivery was not reported. Seven studies included some component that required delivery, receipt or practice of the intervention at home. 84,119,123,126–128,131

Interventions were delivered by researchers, clinicians, therapists, psychologists, postgraduate students and teachers. Specific training in the delivery of the intervention was reported in 12 studies. 84,123,125–128,131,132,136,137,139,145 Intervention manuals were referred to in only nine studies. 84,118,123,125–128,130,131

Quality appraisal and risk of bias

Table 1 provides a summary of the quality and risk-of-bias appraisal of included articles. Risk of bias was performed at the article, rather than study, level, as data collection and/or outcomes differed across articles and, therefore, risk of bias may vary where there were several publications reporting on a single study. Eight of the 15 criteria allowed a rating of ‘Yes’, ‘No’ or ‘unclear’. An additional ‘not applicable’ option was available for four criteria (whether or not intention-to-treat analysis was performed, the longest follow-up was at ≥ 6 months, dropouts were described and missing data were explained). For comparative purposes, a response rate at the longest follow-up of ≥ 85% was considered high, 70–84% as moderate and < 70% was considered a poor response rate. For the assessment of outcome measures, only when all outcomes had good psychometric properties was the score deemed positive.

The included articles were often free from risk of bias on freedom from selective outcome reporting, with 33 out of 35 papers rated ‘yes’. Lack of differences or adjustment for differences between groups at baseline was present in 28 out of 35 articles. Twenty-seven articles included only outcomes with good psychometric properties, and five of the eight articles not meeting this criterion were marked down on only one outcome measure. 83,120,123,124,136 Seventeen articles reported a response rate at the longest follow-up of ≥ 85%; however, none of these articles collected follow-up data at a period ≥ 6 months following the intervention. For those eight articles that did include longer follow-ups, response rates were 70–84% in three articles118,142,145 and < 70% in the rest. 128,130,131,140,144

The criteria that indicated risk of bias most often across articles were the assessment of adherence, compliance or fidelity (20/31 ‘no’); explaining missing data (13/31 ‘no’); blinding of assessors and inclusion of a follow-up after treatment (both 17/31 ‘no’). In addition, only half of those articles including a follow-up assessment did this at ≥ 6 months. The selection bias domains scored poorly, largely because of a lack of clarity in reporting.

Four articles were rated as being at a particularly high risk of bias (were judged to be of poor quality in relation to ≥ 50% criteria). 83,127,135,136 There were five other articles in which < 50% of criteria scored positively and > 15% of criteria were unclear, indicating potential risk of bias. 120,121,132,133,139 In all five articles, it was not possible to determine whether or not randomisation was adequate or allocation was concealed. However, these articles were rated as being free from bias in relation to the suggestion of selective reporting, and had a ≥ 85% response rate at follow-up (except Nekah et al. 133), although none included a follow-up assessment at ≥ 6 months.

The articles by Yetwin127 and Hains et al. 122 were rated as being at risk of bias on adequacy of randomisation because both allowed two participants to swap groups after allocation.

The articles demonstrating as being at the lowest risk of bias were by Whittingham et al. ,131,144 both of which were rated positively on 13 criteria. The only risk of bias in the Whittingham articles arose because outcomes were self-reported, therefore assessors were not blinded, and the response rate at the longest follow-up was < 70%. This poor response rate was symptomatic of the few studies to collect data at the 6-month follow-up interval. The two articles reporting on the study by Brown et al. 128,140 performed similarly to the Whittingham articles. The article by Reigada et al. 125 was also rated as being at low risk of bias overall, with only two negative scores.

Children and young people mental health outcomes

Depression was assessed in the four studies whose intervention aimed to improve this outcome. 83,84,126,145 Three of the studies measured depression using multiple measures,83,84,126 meaning that this was the most frequently occurring outcome category for CBT interventions. General mental health was assessed in four studies,84,124,126,145 and two studies whose interventions aimed to improve anxiety also measured it as an outcome. 124,125

Report Supplementary Material 1, Table 6, displays means and SD for control and intervention groups for each CYP mental health outcome assessed post intervention, in which sample size, mean and SD were reported or calculable. Effect sizes (Cohen’s d), 95% CIs and p-values are included, with a positive effect size representing improvement on the measure.

Evidence of the beneficial effect of CBT for measures of CYP mental health can be found in six out of the seven studies featuring this type of intervention. Evidence of the beneficial effect of CBT on depression across all outcomes is provided by one study83 in which large positive effect sizes are seen for all depression measures, although wide 95% CIs for two of the measures include negligible effect sizes, reflecting the imprecision of the estimate of effect [Beck Depression Inventory (BDI): d = 0.85, 95% CI 0.10 to 1.60; p = 0.03; Centre for Epidemiological Studies Depression scale (CES-D): d = 0.86, 95% CI 0.11 to 1.61; p = 0.03]. Evidence for the beneficial effect of CBT on depression from other studies is less clear; although small to large positive effect sizes were seen for the other studies reporting depression outcomes, these effects tended to be imprecise, with 95% CIs typically including negligible or even slightly harmful effects. Reigada et al. 125 provide evidence of a large beneficial effect of CBT on IBD-specific anxiety (d = 1.31, 95% CI 0.36 to 2.27; p = 0.007). However, there was little evidence for a beneficial effect of CBT on anxiety in the study by Masia-Warner et al. 124 (d = 0.27, 95% CI –0.38 to 0.92; p = 0.41). There was some evidence for a large beneficial effect of CBT on general mental health according to Masia-Warner et al. 124 (d = 1.11, 95% CI 0.42 to 1.81; p = 0.002) and Szigethy et al. 84 (d = 0.91, 95% CI 0.25 to 1.56; p = 0.007), although the wide CIs in the latter study include small effect sizes reflecting the imprecision of the estimate of effect. However, Wicksell et al. 145 report a lack of evidence for the beneficial effect of CBT on general mental health (d = 0.40, 95% CI –0.40 to 1.13; p = 0.41) in their trial. There was no opportunity to meta-analyse CYP mental health outcomes for these interventions evaluating CBT because of differing LTCs or comparators across studies in which outcome categories were shared.

Other outcomes

Report Supplementary Material 1, Table 7, displays findings for other outcomes assessed post intervention. Conflicting evidence regarding the beneficial effect of CBT on LTC-specific QoL and LTC symptoms can be seen across the studies featuring CBT interventions. For instance, although the study by Martinović et al. 83 provides evidence of a large beneficial effect of CBT for epilepsy-specific QoL (d = 1.55, 95% CI 0.72 to 2.37; p < 0.001), a lack of evidence for such a beneficial effect on diabetes mellitus control was reported by Serlachius et al. 130 (d = 0.32, 95% CI –0.08 to 0.73; p = 0.12). The study by Masia-Warner et al. 124 provides evidence of some medium to large beneficial effect sizes for CBT on measures of LTC symptoms. Large beneficial effect sizes for self-reported (d = 1.33, 95% CI 0.61 to 2.05; p < 0.001) and parent-reported (d = 0.98, 95% CI 0.30 to 1.67; p = 0.005) pain were found, as well as a medium beneficial effect on somatisation (d = 0.75, 95% CI 0.08 to 1.42; p = 0.03), although the wide CI for somatisation includes negligible effect sizes, reflecting the imprecision of this estimate of effect. The study by Wicksell et al. 145 provides evidence of large beneficial effects of CBT for pain (d = 1.25, 95% CI 0.45 to 2.05; p = 0.002) and pain-related emotional discomfort (d = 1.46, 95% CI 0.63 to 2.28; p < 0.001). However, there was little evidence for a beneficial effect of CBT on other pain and LTC symptoms outcomes measured by Wicksell et al. 145 There was little evidence for a beneficial effect of CBT on LTC symptoms in other studies. 125,126,130

Evidence for the effect of CBT on LTC symptom severity could be investigated further by meta-analysing outcomes from Szigethy et al. 126 and Reigada et al. ,125 both of which studied patients with IBD and used NDST as the comparator. Reigada et al. 125 reported pooled data for the Paediatric Ulcerative Colitis Activity Index (PUCAI) and Paediatric Crohn’s Disease Activity Index (PCDAI) outcomes,125 whereas Szigethy et al. 126 reported separately for participants with ulcerative colitis and Crohn’s disease. Figure 3 is a forest plot of the effects of CBT on these disease activity measures reported in these studies. This meta-analysis demonstrated little evidence of an effect of CBT on LTC symptoms (d = 0.14, 95% CI –0.34 to 0.63; p = 0.56).

FIGURE 3.

Forest plot showing the results of meta-analysis of the effects of CBT on IBD symptoms post intervention for included studies. SMD, standardised mean difference (Cohen’s d).

Overall, the seven included studies that evaluated the clinical effectiveness of interventions categorised as CBT aiming to improve the mental health of CYP with LTCs show some promising effects. A number of large beneficial effects for CYP mental health and other outcomes were seen, although effects for particular outcomes were rarely consistent across all studies. The heterogeneity in study characteristics prevented meta-analysis of CYP mental health outcomes. The available sample of studies is small, targeting different aspects of mental health for children with different LTCs. There was also diversity in quality and risk of bias across these studies.

Parenting interventions without acceptance and commitment therapy

Of the four parenting interventions assessed in included studies, two were parenting without ACT (Westrupp et al. 118 and Whittingham et al. 131,144). Westrupp et al. 118 assessed the Triple P in families with a child with diabetes mellitus, whereas one of the intervention arms in Whittingham et al. 131,144 was SSTP alone. SSTP is tailored towards families with pre-adolescent children with a disability, which may be physical or intellectual, and behavioural problems. 148 The sample in the study by Whittingham et al. 131,144 had cerebral palsy. Both studies had similar aims, including improving aspects of CYP mental health, with a focus on behavioural and emotional problems in Whittingham et al. 131,144 as well as improving parenting. A total of 50 participants were randomised to intervention or control arms, with all control participants (n = 32) receiving TAU (Whittingham et al. 131,144 used a waiting list control). In both studies, the primary aim was to reduce child behavioural problems through improved parenting skills. Westrupp et al. 118 examined the effects of Triple P in a group of families with children who had elevated behavioural problems at baseline, as well as a group without elevated behavioural symptoms. We were unable to obtain raw data for the subgroup analysis of families with children with behaviour problems and, therefore, could not calculate effect sizes for the study.

Parenting interventions with acceptance and commitment therapy

The study by Whittingham et al. 131,144 included two intervention groups; as well as SSTP, summarised in Parenting interventions without acceptance and commitment therapy, the authors assessed the clinical effectiveness of SSTP plus ACT. Brown et al. 128,140 also evaluated the clinical effectiveness of SSTP plus ACT. Whittingham et al. 131,144 investigated SSTP plus ACT in children with cerebral palsy, whereas the sample of children in the trial by Brown et al. 128,140 had an ABI, of whom a proportion (7%) had cerebral palsy. Both studies compared the parenting intervention with TAU, with Whittingham et al. 131,144 employing a waiting list control group. Both studies aimed to reduce child behaviour and emotional problems and dysfunctional parenting. Because of the consistent comparator and LTC between studies, similar outcomes were considered eligible for meta-analysis. In total, the parents of 92 children were randomised equally between SSTP plus ACT and control groups. Whittingham was a co-author on both Brown et al. 128,140 papers whereas Boyd and McKinlay were co-authors on all four included papers across both studies. 128,131,140,144

Children and young people mental health outcomes

Report Supplementary Material 1, Table 12, shows the findings for each CYP mental health outcome assessed post intervention. Nekah et al. 133 observed a large beneficial effect for anxiety; however, the wide CI includes negligible effect sizes and raises uncertainty about the effect (d = 1.29, 95% CI 0.11 to 2.46; p = 0.04). Similarly, this study reported a large beneficial effect for depression, although the wide CI includes slightly harmful effects (d = 1.04, 95% CI –0.10 to 2.17; p = 0.08). On the other hand, the study by Zareapour et al. 135 reported a particularly large beneficial effect on depression of art-based group play therapy (d = 3.40, 95% CI 2.12 to 4.69; p < 0.001). In the study by Wang et al. ,138 coping improved with a large effect size (d = 0.93, 95% CI 0.27 to 1.59; p = 0.006), but there was little evidence for the beneficial effect of the intervention on behaviour problems (d = 0.31, 95% CI –0.32 to 0.94; p = 0.32).

Given that Nekah et al. 133 and Zareapour et al. 135 both studied patients with cancer and TAU control groups, meta-analysis was possible for their depression outcomes. Figure 8 shows a forest plot for the effects of group play therapy on child depression post intervention. Meta-analysis revealed that the large beneficial effect size was characterised by very wide CIs that include slightly harmful effect sizes (d = 2.20, 95% CI –0.12 to 4.52; p = 0.06) and large heterogeneity between the two studies’ findings (I² = 86.3%), limiting the evidence for positive effects.

FIGURE 8.

Forest plot showing the results of meta-analysis of the effects of group play therapy on child depression post intervention for CYP with cancer. CDS-A, Child Depression Scale Abbreviated; HADS, Hospital Anxiety and Depression Scale; SMD, standardised mean difference (Cohen’s d).

Overall, the included studies cannot tell us with precision whether or not group play therapy interventions are beneficial for depression or anxiety. Despite the individual findings of a large beneficial effect in the study by Zareapour et al. ,135 when meta-analysed, this effect is uncertain. Behaviour outcomes did not improve in the study by Wang et al. ;138 however, there may be a beneficial effect for coping. The low quality and high risk of bias across the three studies further weakens the strength of conclusions drawn from the evidence.

Children and young people mental health outcomes

Both palliative care studies assessed child anxiety and depression. Shoshani et al. 137 also measured emotional problems, hope and optimism. Report Supplementary Material 1, Table 13, displays findings for CYP mental health outcomes assessed post intervention. Taking the results of Lyon et al. ’s123 trial in isolation gives little evidence of the beneficial effect of FACE-TC for anxiety and depression outcomes. The trial by Shoshani et al. 137 found that the Make-a-Wish intervention had a large, beneficial effect on positive emotions (d = 1.03, 95% CI 0.52 to 1.55; p < 0.001), but evidence for reduction in negative emotions was lacking. Shoshani et al. 137 also reported medium-sized beneficial effects on range of measures, including anxiety, depression, general mental health and hope, but all of these effects were characterised by imprecision, with wide CIs including small effect sizes.

As the study characteristics were similar, meta-analyses were conducted when outcome categories were shared. Figure 9 is a forest plot of the effects of palliative care on child anxiety post intervention reported by Lyon et al. 123 and Shoshani et al. 137 Figure 10 is a forest plot of the effects of palliative care on child depression post intervention reported in these studies. 123,137 Meta-analysis revealed that palliative care interventions show a medium-sized beneficial effect on both child anxiety (d = 0.61, 95% CI 0.19 to 1.02; p = 0.004) and depression (d = 0.61, 95% CI 0.19 to 1.03; p = 0.05). However, there is uncertainty in concluding a positive effect of palliative care interventions on these aspects of child mental health, as wide CIs included small effect sizes.

FIGURE 9.

Forest plot showing the results of meta-analysis of the effects of palliative care on child anxiety post intervention for CYP with cancer. BAI, Beck Anxiety Inventory; BSI-18, Brief Symptom Inventory-18 item version; SMD, standardised mean difference (Cohen’s d).

FIGURE 10.

Forest plot showing the results of meta-analysis of the effects of palliative care on child depression post intervention for CYP with cancer. BAI, Beck Anxiety Inventory; BSI-18, Brief Symptom Inventory-18 item version; SMD, standardised mean difference (Cohen’s d).

Other outcomes

Report Supplementary Material 1, Table 14, provides the findings for all other outcomes assessed post intervention. There was little evidence of a beneficial effect of FACE-TC on the majority of outcomes reported. However, Lyon et al. 123 did find a large beneficial effect for spirituality (d = 0.99, 95% CI 0.19 to 1.78; p = 0.02), albeit with imprecision in the estimate of effect, with the CI including small effect sizes. Shoshani et al. 137 provided evidence of a large beneficial effect for the Make-a-Wish intervention for physical QoL (d = 0.87, 95% CI 0.37 to 1.38; p < 0.001). Both studies measured intervention effects using the Paediatric Quality of Life Inventory (PedsQL) 4.0: Physical functioning subscale, and, therefore, physical QoL specifically could be meta-analysed across the two studies (Figure 11). The result of this meta-analysis provides evidence for a medium-sized beneficial effect of these palliative care interventions on physical aspects of QoL in CYP with cancer (d = 0.73, 95% CI 0.31 to 1.15; p = 0.001). As both studies used the PedsQL 4.0: Physical functioning subscale, meta-analysis using mean differences can be seen in Appendix 6.

FIGURE 11.

Forest plot showing the results of meta-analysis of the effects of palliative care on child physical quality of life post intervention for CYP with cancer. SMD, standardised mean difference (Cohen’s d).

Overall, analysis of the clinical effectiveness of palliative care interventions for CYP with cancer provides tentative evidence that there may be some benefits for child depression and anxiety, child physical QoL and child spirituality. Only two studies investigated the clinical effectiveness of this type of intervention, with differences in findings perhaps reflecting the different intervention content. 123,137 The relatively low quality of the two articles further limits the confidence of evidence regarding whether or not palliative care interventions show beneficial effects in children with mental ill health and cancer.

Children and young people mental health outcomes

Report Supplementary Material 1, Table 15, displays findings for CYP mental health outcomes assessed post intervention. Evidence for the positive effect of PMR is provided by the study by Yang et al. ,139 in which a large, beneficial effect on depression (d = 1.01, 95% CI 0.48 to 1.43; p < 0.001) was observed. Yang et al. 139 also reported a medium beneficial effect on anxiety (d = 0.70, 95% CI 0.19 to 1.20; p = 0.007), although the CI for this effect includes some small beneficial effect sizes. In contrast, there was little evidence reported by Bignall et al. 119 for the effect of their relaxation intervention on anxiety, with small harmful effect sizes reported, although CIs were wide and included positive effects (state anxiety: d = –0.39, 95% CI –1.11 to 0.34; p = 0.30; trait anxiety: d = –0.21, 95% CI –0.93 to 0.51; p = 0.58).

Given that the same asthma education appeared in both Bignall et al. ’s119 intervention and control groups, we meta-analysed the anxiety outcomes alongside the data from Yang et al. ’s139 study, whose control group received TAU (Figure 12). The trait anxiety subscale of the Spielberger State–Trait Anxiety Inventory (STAI) was more aligned with the Screen for Child Anxiety Related Emotion Disorder (SCARED) assessment tool than the state subscale data from Bignall et al. ,119 and thus was entered into meta-analysis instead. Meta-analysis of these outcomes provides a lack of evidence for the effect of relaxation interventions on anxiety (d = 0.28, 95% CI –0.60 to 1.16; p = 0.53). Heterogeneity across the two papers was wide (I² = 75.3%), further impacting the confidence in the pooled effect.

FIGURE 12.

Forest plot showing the results of meta-analysis of the effects of relaxation on child anxiety post intervention for CYP with asthma. SMD, standardised mean difference (Cohen’s d); STAI-T, Spielberger State–Trait Anxiety Index-Trait subscale.

Other outcomes

Both studies included assessments of asthma symptoms. Yang et al. 139 recorded the incidence of daytime and night-time symptoms, as well as the percentages of days without asthma symptoms and of participants experiencing one or more asthma attacks. However, these outcomes were not reported in terms of means and SDs, and so were not included in data synthesis. In addition, Bignall et al. 119 included a measure of LTC-specific QoL, the PedsQL: Asthma module.

Report Supplementary Material 1, Table 16 displays findings for other outcomes assessed post intervention. Evidence for the effect of PMR on asthma symptoms is provided by Yang et al. ,139 with large beneficial effects on the number of daytime (d = 2.32, 95% CI 1.68 to 2.96; p < 0.001) and night-time (d = 1.91, 95% CI 1.31 to 2.50; p < 0.001) symptoms reported. However, this was only corroborated by one measure of asthma symptoms in Bignall et al. ’s119 study, in which a large effect size was found for asthma control, although the wide CIs included some negligible effect sizes (d = 0.80, 95% CI 0.05 to 1.54; p = 0.04). There was a lack of evidence from the study by Bignall et al. 119 for other measures of asthma symptoms. For asthma-specific QoL there was a medium effect size, although this estimate of effect is uncertain, with a wide CI including negligible effect sizes.

Combined LTC symptom outcomes were produced for each study and entered into a meta-analysis (Figure 13), which showed a lack of evidence for the effect of relaxation interventions on LTC symptoms. Despite a large pooled effect, it was characterised by very wide CIs, including harmful effects as well as very large heterogeneity (I² = 97.4%) on account of the differences between each study’s findings (d = 1.10, 95% CI –0.88 to 3.08; p = 0.28).

FIGURE 13.

Forest plot showing the results of meta-analysis of the effects of relaxation on asthma symptoms post intervention for included studies. SMD, standardised mean difference (Cohen’s d).

Overall, the included studies provide mixed evidence regarding the benefits of relaxation interventions for depression, anxiety and LTC symptoms. Yang et al. 139 reported medium to large beneficial effects across these domains. Bignall et al. 119 reported little evidence for the beneficial effects of a combined relaxation intervention. A reason for this may be the asthma education active comparator group in the trial by Bignall et al. ,119 which did improve asthma symptoms compared with baseline. Both studies were rated as being of low quality and high risk of bias. We recommend that the positive effects from Yang et al. 139 are treated with caution.

Children and young people mental health outcomes

Report Supplementary Material 1, Table 17 displays findings for each CYP mental health outcome assessed post intervention. There was little evidence across both studies regarding the clinical effectiveness of stress management programmes on CYP mental health outcomes. On the occasions when medium-sized effects were recorded, these were characterised by wide 95% CIs that included harmful effect sizes. This was the case for state anxiety (d = 0.69, 95% CI –0.41 to 1.78; p = 0.23) and behavioural coping (d = 0.65, 95% CI –0.44 to 1.74; p = 0.25) in the study by Hains et al. 122

Meta-analysis was possible for their LTC-specific stress and coping outcomes. Figure 14 shows a forest plot for the effects of SMT on diabetes mellitus-specific stress post intervention. Meta-analysis revealed little evidence of a beneficial clinical effect for LTC-specific stress (d = 0.10, 95% CI –0.62 to 0.82; p = 0.80). As both studies used the Diabetes Stress Questionnaire (DSQ), meta-analysis using mean differences can be seen in Appendix 6. Figure 15 is a forest plot of the effects of SMT on coping post intervention reported by Boardway et al. 120 and Hains et al. 122 Negative coping from the KidCOPE measure was selected from Hains et al. ,122 as this subscale has most similarities with Boardway et al. ’s120 Ways of Coping checklist. Meta-analysis again revealed little evidence of a clinical effect of the intervention on coping (d = –0.09, 95% CI –0.83 to 0.64; p = 0.81).

FIGURE 14.

Forest plot showing the results of meta-analysis of the effects of SMT on T1DM-specific stress post intervention for CYP with diabetes mellitus. SMD, standardised mean difference (Cohen’s d).

FIGURE 15.

Forest plot showing the results of meta-analysis of the effects of SMT on coping post intervention for CYP with diabetes mellitus. SMD, standardised mean difference (Cohen’s d).

Other outcomes

Report Supplementary Material 1, Table 18, displays findings for each other outcome assessed post intervention. There was again a lack of evidence for the effect of stress management interventions on any outcome. LTC symptoms were meta-analysed for the two studies using the most similar glycaemic control outcomes [i.e. level of total glycosylated haemoglobin (GHb) in Boardway et al. ,120 level of glycated haemoglobin (HbA_1C) in Hains et al. 122]. This meta-analysis (Figure 16) provides little evidence of an effect of SMT on LTC symptoms (d = –0.10, 95% CI –1.77 to 1.57; p = 0.91), showing imprecise effects owing to a wide CI and large heterogeneity across the two studies (I² = 79.9%).

FIGURE 16.

Forest plot showing the results of meta-analysis of the effects of SMT on CYP diabetes mellitus symptoms at post intervention. SMD, standardised mean difference (Cohen’s d).

Overall, the two included studies provide very little evidence regarding the effect of SMT on a wide range of outcomes. Effect sizes ranged from medium-sized and beneficial to large harmful effects, but all were characterised by imprecision in the estimate with large effect sizes, at least in part due to the small sample sizes in these studies. 120,122

Children and young people mental health outcomes

Both EIT studies used the 28-item version of the General Health Questionnaire (GHQ) and reported subscales for anxiety, depression, social dysfunction, somatic symptoms and overall mental health. No other outcomes were reported.

Report Supplementary Material 1, Table 19, displays findings for these study outcomes. Both studies reported very large, positive beneficial effects for each outcome domain. Effect sizes ranged from 2.41 (95% CI 1.58 to 3.23; p < 0.001) for depression to 6.79 (95% CI 5.14 to 8.44; p < 0.001) for overall mental health in the study by Pourmohamadreza-Tajrishi et al. ,134 whereas Ashori et al. 132 reported effect sizes in the range 11.0–17.4 for the same outcomes.

Initial assessment of the two papers suggested that they were reporting on different studies, because of differences in the EIT programme outline and age of the sample. On inspection of the raw data presented for each study, the review team decided that the mean scores were too similar to be confident that these were in fact different samples and, therefore, meta-analysis was not performed. It is also noted that extremely large effect sizes are evident, in part because of the very small standard deviations presented. We attempted to contact the authors for clarification but received no reply.

Despite the reports of very large beneficial effects in the individual studies of Ashori et al. 132 and Pourmohamadreza-Tajrishi et al. ,134 both studies scored poorly on quality appraisal, notwithstanding the conspicuous similarities between the studies. Overall, there is a lack of evidence to confidently determine the effect of EIT on the mental health of CYP with hearing loss.

Children and young people mental health outcomes

Diego et al. 121 assessed anxiety using the Spielberger State–Trait Anxiety Inventory-State subscale (STAI-S) and depression using the CES-D. Report Supplementary Material 1, Table 20, displays these findings. There is evidence that massage therapy had a beneficial effect on child anxiety, with a large positive effect size calculated (d = 1.23, 95% CI 0.35 to 2.10; p = 0.007). Despite a medium positive effect size for depression, there is a lack of evidence for this beneficial effect, given the wide CI that includes negligible effect sizes (d = 0.75, 95% CI –0.08 to 1.58; p = 0.08).

Other outcomes

Diego et al. 121 reported the effects of massage therapy on markers of immune function. Medical charts were reviewed to assess HIV infection progression (CD4 cells per mm³, CD8 cells per mm³, CD4-to-CD8 ratio) and protection against viruses associated with HIV (CD56 cells per mm³, CD56-to-CD3 ratio). Overall, there is a lack of evidence for the effects of massage therapy on symptoms of HIV infection.

The evidence provided in the study by Diego et al. 121 suggests that massage therapy may reduce anxiety and, to a lesser extent, depression in adolescents infected with HIV. However, there was a lack of evidence for effects on markers of immune function. This was the only study reporting massage therapy, and it was of low quality; therefore, there is a lack of unequivocal evidence regarding the effects of this intervention on the mental health of CYP with LTCs. It is notable that the comparator in this study was similar to components of interventions tested in the studies by Bignall et al. 119 and Yang et al. ,139 which were earlier categorised as relaxation interventions.

Children and young people mental health outcomes

Child anxiety and depression were self-reported using the Children’s Anxiety Sensitivity Index and Child Depression Inventory-Short Form, respectively. Report Supplementary Material 1, Table 21, shows findings for these outcomes as measured post intervention. There was little evidence of an effect of heart rate variability biofeedback on either anxiety or depression.

Other outcomes

Yetwin127 assessed QoL, quality of sleep and LTC symptoms (pain). Report Supplementary Material 1, Table 22, shows findings for these other outcomes measured post intervention. There were a couple of medium to large effect sizes reported for both LTC symptom and sleep outcomes; however, these were characterised by imprecision in the estimates, with CIs including negligible effect sizes (e.g. Faces Pain Scale: d = 0.86, 95% CI –0.09 to 1.81; p = 0.08; and Child/Adolescent Sleep–Wake Scale: Falling Asleep: d = 0.78, 95% CI –0.16 to 1.72; p = 0.11). This tentative evidence of beneficial effect also conflicted with other measures of these outcomes that showed a lack of evidence of beneficial effects (e.g. Varni/Thompson Paediatric Pain Questionnaire Worst Pain: d = –0.05, 95% CI –0.95 to 0.85; p = 0.92; Child/Adolescent Sleep–Wake Scale total score: d = 0.27, 95%CI –0.63 to 1.18; p = 0.56).

Overall, the one paper assessing heart rate variability biofeedback provides a lack of evidence for its effect on child anxiety, depression, sleep, QoL or pain in adolescents with chronic pain. In addition to being the sole study to use this type of intervention, Yetwin127 scored very poorly on quality appraisal; therefore, there is insufficient evidence to assess the clinical effectiveness of heart rate variability biofeedback for improving either mental health or other outcomes in CYP with LTCs.

Children and young people mental health outcomes

Depression was self-reported using the BDI. Report Supplementary Material 1, Table 23, shows the findings for this measure and other measures (described in the following section) post intervention. There was a lack of evidence for the effect of resistance training on depression (d = –0.23, 95% CI –1.07 to 0.60; p = 0.59) compared with an aerobic exercise programme.

Other outcomes

Gordon et al. 129 assessed QoL using the physical subscale of the Short Form questionnaire-36 items (SF-36). LTC symptoms were measured using the Fatigue Severity Scale, and several exercise tests were chosen to assess exercise tolerance and muscular endurance, including the number of push-ups completed until fatigue, the sit-to-stand test, the time to fatigue on a submaximal walking test and the metabolic equivalent during the walking test (ratio of energy expenditure during the test to energy expenditure at rest). There was a lack of evidence for the effects of resistance training on these outcomes.

Overall, the one paper assessing the effects of a resistance training programme on depression, physical indices and QoL provides little evidence of a benefit compared with aerobic training. It should be noted that Gordon et al. 129 scored moderately well on quality appraisal, describing randomisation, allocation concealment, blinding of outcome assessors and using an intention-to-treat analysis. However, groups were imbalanced at baseline, when the resistance training group had performed statistically significantly better on the sit-to-stand test, and this was not adjusted for; the intervention was not well described and adherence, compliance and fidelity were not assessed. For this reason, there is insufficient evidence to state with confidence whether or not resistance training has any benefit on the mental health of CYP with LTCs.

Children and young people mental health outcomes

Bufalini136 assessed anxiety at four time points using the modified Yale Preoperative Anxiety Scale. Report Supplementary Material 1, Table 24, shows the findings post intervention, just prior to sedation. There is evidence that music therapy had a large beneficial effect on child anxiety (d = 1.47, 95% CI 0.76 to 2.18; p < 0.001). No other outcomes were measured in this study.

The evidence provided by Bufalini136 suggests that there may be a benefit of music therapy in reducing procedural anxiety. However, in addition to being the sole paper reporting on this intervention, it scored the lowest of all included studies on quality appraisal. Therefore, more evidence is required to increase confidence in the clinical effectiveness of music therapy for improving the mental health of CYP with LTCs.

Reading and re-reading the included studies

This initial stage involved the careful reading of all of the included articles in order for reviewers to familiarise themselves with the studies and their findings. This process began in the screening phase and continued through data extraction and analysis.

Determining how the studies are related

Following data extraction, structured summaries were created to juxtapose the same type of information across articles for consideration of similarities and differences. 159 Information tabulated included study aims, samples, methods, interventions and themes. These summaries included information on participant perspective and whether or not the intervention targeted procedural anxiety to allow categorising of articles for synthesis. Themes from all included studies presented in their findings sections were also tabulated and colour coded separately to allow for early comparison of themes across different studies.

Reciprocal translation of studies

Each article was rated by two reviewers (DM and LS) according to whether the analysis was interpretive or descriptive and whether or not there were more than two pages of findings focused on the experience of the intervention. This facilitated a more in-depth consideration of the quantity and quality of interpretive findings available within each article for analysis. This process identified a smaller number of studies with a large volume of interpretive information on which to focus the initial synthesis.

The reciprocal translation stage involved the extraction of first- and second-order constructs from the studies identified as having a large volume of interpretive findings available for synthesis. First-order construct data were identified as efforts to make sense of daily experiences, often illustrative participant quotes within studies. Second-order construct data consisted of author analysis and interpretation in included articles. Third-order constructs were our interpretations as reviewers, which go further than first- and second-order constructs in individual articles. 220 Table 3 displays examples of first-, second- and third-order constructs contributing towards the analysis. Concept maps showing the relationships between analytical ideas and themes across articles were then developed.

Initially, articles that focused on young people, parents or practitioners were synthesised separately. However, as we observed considerable overlap between perspectives, all of the articles with large amounts of interpretive findings were synthesised. We compared the findings and interpretations of each article to the concept map developed to that point, adding new constructs and themes when relevant. NVivo software was used to code findings and interpretations, line by line, in articles against the ideas seen in the concept maps, as well as new ideas. Refutational translation was conducted when any quotations, metaphors, themes and concepts that offered oppositional or counterarguments to current findings were specifically searched for and incorporated into the findings. This process was completed with the articles in the initial synthesis, and repeated with findings and interpretations from studies selected through the purposive sampling process. When new constructs and themes emerged, studies previously examined were checked to see if they could contribute to these new ideas. Studies evaluating procedural mental health interventions were initially considered separately, but, because of the similarity in the concepts across these articles and other articles focusing on broader interventions, they were then included in the reciprocal translation process during this purposive sampling phase. The remaining articles that neither had large amounts of interpretive findings nor were included in the purposive sampling were then read to see which constructs and themes were supported by the data and if there was any material that refuted the content of these themes. NVivo was used to code any relevant information from these remaining articles. Feedback and ideas relevant to the synthesis from the CYPAG and parent meetings were considered before the reciprocal translation process began and incorporated throughout the different stages of the synthesis, as outlined in Table 2.

Synthesising translations/creating a line of argument

By considering relationships and overlap among the constructs and themes within the synthesis, a line of argument was developed. A line of argument synthesis attempts to make sense of the synthesis as a whole through the development of a new theory or model of understanding. The line of argument consisted of a model that captured how the overarching themes, and the categories that contributed towards them, related to one another and captured the experience of participants with a LTC receiving mental health and well-being interventions, their families and the professionals involved in their care.

Theme: a safe space

‘A safe space’ illustrates how the presence of familiar family or staff within a setting that is already known to them can result in CYP feeling reassured and comforted. These feelings are necessary if CYP are to feel receptive to, and thus benefit from, an intervention. The people who deliver an intervention, or ‘intervention deliverers’, have an important role in helping CYP familiarise themselves with the intervention setting and with other people taking part. These factors, together with an understanding of what the intervention entails, may help reduce anxiety about entering an unfamiliar environment and enable CYP to experience the setting as somewhere that is both emotionally and physically ‘safe’. Nineteen articles contributed to this theme (see Report Supplementary Material 2, Table 6).

The authors of three studies found that CYP may experience anxiety at the prospect of meeting unknown people prior to an intervention172,177,182 and that the presence of a parent or support of a known staff member, such as a social worker, can reduce this anxiety and enable CYP to engage with a previously unknown intervention (see Report Supplementary Material 2, Table 7). A young person with T1DM illustrates this when discussing their anxiety about meeting the person delivering their psychoeducation intervention for the first time:

. . . I hadn’t met anybody and didn’t know where I was going and all those kinds of things, ummm, being quite young, being very ill, very worried. And I personally wouldn’t have coped very well if they had just invited me to go to somewhere. I needed my mum there.

Young person, T1DM182

The above quotation suggests that the involvement of someone who is known to the young person is comforting. The reduction in anxiety as a result of the presence of someone familiar may potentially ‘mask’ a mental health need of CYP taking part. However, within the initial stages of an intervention, a familiar person could transform a previously feared space into one that is ‘safer’, allowing CYP to better engage with the intervention being delivered. The following quotation from a young person taking part in a study examining different management techniques for needle-related distress illustrates this:

If I actually knew the nurse it would be OK, but if I didn’t I would be like ‘what’s going on? I don’t even know you!’

Young person, cystic fibrosis198

Making an environment feel safer may be an important intervention itself, especially for CYP with anxiety around medical procedures. This may depend on the frequency and duration of medical treatments and related levels of distress for the young person.

Fears about not knowing anyone else who was taking part were also acknowledged at the start of a creative writing group172 and online peer-support interventions. 177 Authors of three studies suggest that CYP valued the opportunity to get to know peers more. 162,174,179 As CYP became more acquainted with one another, a feeling of acceptance was created, giving the intervention space a sense of being familiar and safe, helping CYP feel able to share their feelings and experiences with one another. This is illustrated by a participant from a camp:

The camp did help with that [to talk about our different experiences] by bringing us all together and letting us get to know each other for a couple of days so that we felt that we were comfortable enough to go up to each other and just talk about it.

Young person, craniofacial differences179

First- and second-order data from 11 studies suggests that the individuals who delivered the interventions played a key role in supporting the initial recognition of similarity between peers and encouraging interaction and engagement during the early stages of group interventions. 114,166,172–177,179,184,192 Second-order data from five studies suggests that deliverers of interventions should plan to allow sufficient time for CYP to get to know each other. 166,169,174,175,179

Report Supplementary Material 2, Table 7, indicates that prior experience of a setting as somewhere safe was viewed positively by CYP. 175,176 Below, a young person indicates how they perceived coming back to a hospital environment with which they were already familiar for an introduction to a peer-support programme:

. . . for me, the hospital became like a second home. It was my safe haven. I always felt safe in the hospital. I always felt comforted in the hospital. So having the 8-week group in the hospital, I saw it as somewhere safe that I could come back to.

Young person, mixed175

Authors identified that CYP valued knowing what to expect from an intervention, either through explanation regarding the content of the intervention and its aims by the intervention deliverer182,185,195 or through provision of the opportunity to observe others participating before taking part themselves. 205 Wolf Bordonaro197 discusses how the process of asking for information from the intervention deliverer about the time and place of follow-up sessions by CYP may be a method used by CYP to feel in control within the intervention situation. Many CYP with LTCs experience unpredictable symptom fluctuations and are expected to attend appointments to manage their physical health – aspects of their life over which they may have little control. Feeling in control of a situation related to their LTC could be a novel experience for CYP, providing an opportunity to distance themselves from anxiety associated with their LTC and its management. CYP may feel reassured by the presence of a predictable intervention structure, as illustrated by Gillard et al. ,174 who discuss the use of ground rules and set format within their ‘TeenTalk’ intervention. This use of predictability to help CYP feel safe within the intervention setting is explored further within Theme: boundaries.

This theme indicates that CYP may find it helpful to invite their family members and familiar staff to attend distressing procedures and accompany them in unfamiliar intervention situations. It is also important that CYP are given enough time to get to know one another in group settings and are supported to familiarise themselves with the setting and staff beforehand. Knowing what to expect from an intervention through the use of a predictable routine and requesting information may also help reduce CYP’s worries about attendance and help them feel safe and in control of the intervention situation.

Theme: boundaries

This section considers the theme ‘boundaries.’ Authors of the 34 articles included in this theme (see Report Supplementary Material 2, Table 6) describe the efforts made by the deliverers and/or recipients of an intervention to set certain rules for one another. These boundaries may include maintaining the privacy and anonymity of intervention recipients; setting rules for behaviour or acceptable discussion topics; and having an established meeting time, structure and setting. As indicated within Figure 19, the setting of boundaries appears to facilitate the experience of a setting as ‘safe’, which may in turn enable CYP and their families to discuss their emotions and sensitive issues connected to living with their LTC. This is an example of CYP feeling ‘unconstrained’ by their LTC, an idea that is explored in Theme: unconstrained.

Seven studies suggest that among CYP with LTCs there appears to be a fear of their diagnosis, or other issues they regard as private, being disclosed to others. 167,172,174,177,178,187,196 Authors describe health and social care professionals playing an important role in advertising interventions to an appropriate audience,172 monitoring those who are able to access the intervention210 and locating a secure, safe environment in which to discuss sensitive topics. 176

Three study authors suggest that there is a demand for intervention deliverers to respect the privacy of CYP by creating opportunities for CYP to talk without parents or the deliverer present,179,182,202 as recognised by a parent supporting their child to attend a CBT intervention:

I did wonder whether they should have spoken to her by herself . . . I just felt like there could be things going on that are of worry and whether friends like you etc. that may not want to be said in front of parents.

Parent, T1DM182

This appears to contradict the observation made under the theme of ‘a safe space’ that the presence of a familiar person may enable CYP to access an intervention and talk about their experiences. As CYP become familiar with an intervention and its setting, they become more open to meeting with an intervention deliverer, or other young people, without the support of family or known staff members.

Two studies suggested that online interventions provide the opportunity for greater anonymity than face-to-face interventions, which some CYP and their parents felt facilitated participation and engagement among less confident attendees. 177,178 This is evidenced by a quotation from a young person who took part in an online support and education intervention, supported by older peer mentors:

The internet was better because you’re not looking at them straight in the eye. I guess it is a bit easier. I think I shared my feelings better. Face to face would be hard.

Young person, asthma/allergies178

Three studies suggest that the use of certain rules, such as the use of a deliberate turn-taking structure within sessions, can help CYP feel safe enough to express their emotions,174,189 as well as ensure that these emotions are appropriately managed. 187 This is illustrated by a family member discussing taking part in a family system intervention:

I liked how each of us had a turn to speak and share our experiences. I also liked how everybody else listens on what you have to say.

Sibling, ABI189

Although Gan et al. 189 do not indicate how the precedence for this turn-taking was established, the topic manual developed to guide the implementation of their family intervention indicates that time to introduce the intervention and develop an agreed contract between participants and intervention deliverers was included during the early stages of the intervention. One mother taking part in a family art therapy intervention agreed that boundaries around acceptable behaviour should be established early on in the intervention:

The kids, you know, I think it’s extremely important when you are dealing with children, regardless of illness or age, that they know from the get go, what is expected, and what their limitations are . . . then when they are imposed after the fact, it feels restrictive.

Parent, mixed164

Here, the late introduction of boundaries to curb potentially disruptive behaviour appears to have been experienced as critical and restraining. Perhaps the introduction of boundaries at an earlier stage of an intervention may ensure that the CYP’s behaviour remains respectful of others and also enables them to relax, have fun and feel ‘uncontrolled’. This possibility is supported by Gillard and Watts,173 who highlight how camp staff supported CYP to take part in ‘transgressive activity’ while attending camp, such as the challenge to overcome ‘security’ measures to win a box of sweets for their cabin group. This was seen as an opportunity to engage in risk-taking behaviour within set appropriate limits. Four studies highlighted how behavioural rules ensured that CYP respected the personal space and boundaries of others,173,181,185 which contributed towards the creation of a supportive community within peer group interventions:

That it’s got a sense of community . . . that everybody respects everybody . . . and you can talk about everything and anything . . . Some people disagree with you, but they don’t bark at you.

Young person, cerebral palsy/spina bifida170

One topic in which discussion appeared to be restricted was in relation to life-threatening LTCs. Within some interventions, the avoidance of the discussion of death and dying was noticeable. For one study, this was a result of the intervention’s explicit aim to provide the child with respite from worries regarding their LTC,173 whereas within an online support site for CYP with cystic fibrosis and their parents, the avoidance of discussing the life-limiting nature of the LTC was not explicitly discussed. 183 Perhaps this reflects the desires of participants within this group and may highlight the need to discuss ground rules at the beginning of an intervention, to ensure that participants know what to expect.

The conscious or unconscious avoidance of discussions of death may reflect the need for CYP with certain diagnoses, and those involved with their care, to change the focus of mental health and well-being interventions to consider ‘how to make the best’ of the time left and live as normal a life as possible. In contrast, two studies aimed at alleviating distress for CYP with cancer undergoing stem cell transplantation suggest that the interventions provided the opportunity for CYP to confront their own mortality and the possibility of dying. 201,202 These two apparently contradictory concepts illustrate the important role mental health interventions may play in facilitating both the acceptance of the possibility of death and ensuring that life is as fulfilling as possible for CYP with potentially terminal diagnoses. The balance between these two alternative outcomes may be affected by LTC type and prognosis, as well as by the choice of coping strategy preferred by the CYP and their family. The evocative nature of the work, particularly in supporting CYP at the end stage of their illness, necessitates the presence of an intervention deliverer who is adequately trained and the resources to ensure that they are provided with adequate supervision. The need for an intervention to be adequately resourced and consider the type of LTC and personal preferences of CYP is discussed within the ‘availability’ and ‘engagement’ themes under Construct: getting in and staying in.

Setting aside time to spend together can be difficult because of the challenges of meeting the needs of different family members and balancing family life alongside the demands of managing a LTC. 164,188 First- and second-order construct data from two studies evaluating family-orientated interventions indicated that CYP and their parents appreciated having a scheduled opportunity to spend time together as a family. 164,189 This appeared to facilitate communication between family members, as described by two parents taking part in a family system intervention, facilitated by a social worker/rehabilitation counsellor:

Opened my eyes as to what [the adolescent with the ABI] was thinking.

Parent 1, ABI189

Helped us to discuss concerns and feelings that otherwise might not surface.

Parent 2, ABI189

This opportunity to communicate more openly facilitated insight into what the young person with the LTC was thinking and meant that the family could use the space to better understand each other’s feelings. Authors of two studies also indicated that CYP valued the opportunity to meet with peers. 163,174 This may prove beneficial for CYP in building respectful, understanding and empathetic relationships with family and peers with a LTC, as discussed in Construct: social support.

In this section we discuss the value of establishing boundaries, and their role in ensuring that CYP experience an intervention setting as safe and feel able to express their emotions. Intervention deliverers and participants can be involved in the creation of boundaries, which through the sharing of experiences can facilitate the development of understanding and caring relationships.

Theme: unconstrained

This theme explores how the opportunity to meet similar peers seemed to be important to CYP. This experience appeared to provide CYP with the freedom to fully express their emotions without being constrained by the expectations or views of others. The theme also discusses how mental health and well-being interventions can provide an experience of freedom or respite from the constraints and daily routine of having a LTC and the anxiety and distress that can result. Twenty-eight articles contributed to this theme (see Report Supplementary Material 2, Table 6).

Although the idea of an intervention setting being both ‘safe’ and ‘unconstrained’ appears to be contradictory, there is a large degree of overlap in the 12 studies that contribute towards both of these categories,162,164,166,173–177,179,181,182,205 across a diverse range of interventions and LTCs. The ‘boundaries’ category suggests that a therapeutic space must have limits if it is to be experienced as safe by CYP. This may account for these two different sets of ideas from the same studies. The use of ‘boundaries’ within an intervention may need to be flexible to consider the changing needs of participants and facilitate their ability to access ‘a safe space’ or feel ‘unconstrained’ as required. A possible example of this is that CYP may wish to think about the risk of dying on some occasions, but not others.

Interventions that provided the opportunity for CYP to be themselves, speak openly and experience freedom from stigma or judgement were particularly valued163,166,167,169,173,174,176,187 and facilitate a safe space in which CYP felt able to talk and share their experiences:

I feel open that I could talk to whoever, however I want like I can talk open about what I feel and what I need.

Young person, HIV176

One way in which this appears to have been achieved is by CYP interacting face to face with peers with similar LTCs (see Report Supplementary Material 2, Table 7). Six of these interventions were residential camps;163,166,167,169,173,174 however, this opportunity was also provided during two weekly peer support group interventions. 176,187

The opportunity to feel free from the perceived judgement of others appeared to be particularly relevant for CYP whose LTCs result in a visible difference in their appearance or which are associated with high levels of perceived stigma. 163,166,167,172–174,176,187 Although the majority of the interventions included in this review were delivered to CYP with the same LTC, there was some evidence to suggest that CYP, including those with stigmatised diagnoses,167 also enjoyed and benefited from interacting with peers with different LTCs. 164,177,178,217 This evidence suggests that the ability of CYP to empathise with others who are experiencing similar difficulties to them may transcend diagnosis.

In addition to the positive benefits of spending time with peers in similar situations, the authors of five studies explicitly discussed the value of CYP having the opportunity to access an ‘adult-free space’ to meet by themselves169,179 or separately with the intervention deliverers. 164,182,202 Four of these studies interviewed participants between 9 and 24 years of age; thus, the developmental stage of participants may have influenced their expressed need and comfort with a space separate from their parents. 169,179,182,202 A flexible approach, depending on timing and needs of CYP, might be appropriate, as suggested by CYP with craniofacial conditions receiving a camp-based intervention:

It is good to have a leader [adult deliverer] but maybe not [one] who does everything . . . like especially when we didn’t know each other at first it was really tough to get us to talk . . . So it might be good to have a leader at first but once we got to know each other it was a little easier.

Young person, craniofacial conditions179

These peer relationships also appeared to provide CYP with a release from perceived pressure from health professionals and parents, as illustrated by a young person with T1DM talking about their relationship with their mentor, a slightly older peer with lived experienced of their LTC:

The mentor manages to come into both worlds, the world of the adults and the world of my worries, he knows them, you can choose what to share with him and what not to share, he’s tolerant. He doesn’t pressure me.

Young person, T1DM162

In contrast with the need for structure highlighted within ‘boundaries’, Barnetz and Feigin162 highlight how young adolescents can learn to accept their condition through relationships based on enjoyment and interest, unconstrained by an adult-orientated approach to relationships that emphasises structure, contribution and development. This apparent freedom from adult rules offers a contrast to life as usual for CYP, when opportunities for experiencing autonomy may be complicated or limited by the LTC. The ability to learn and express feelings through unstructured, informal contact with peers was also identified by studies evaluating camp-163,166,169,174,179,185 and technology-based interventions. 177,183 The role of CYP being able to express themselves emotionally to receive emotional support and learn new skills is discussed in more detail in Theme: emotional and informational support and Theme: managing myself.

Another way CYP were able to feel ‘unconstrained’ by their LTC was through interventions that provided respite from the ‘daily suffering and uncertainty’ associated with treatment for a LTC,202 as well as a temporary reprieve from worries about daily life and the limitations enforced on them by the LTC. This is illustrated by a quotation from a parent of a child receiving music therapy:

He looked forward to the [therapeutic music session]. He knew it was going to be a day where he didn’t have to worry about if they were coming in to give him a pill or if they’re coming in to start the chemotherapy.

Parent, cancer201

Interventions associated with feelings of respite seem to be those that are not part of routine treatment, such as camps, or interventions that aim to provide an alternative environment or distraction from the hospital setting. 177,201,205 This release from the daily burden of living with a LTC could potentially improve the mental health of CYP. CYP taking part in camp interventions spoke of the bodily freedom they experienced in being able to take part in activities that they had never tried before because of the physical limitations placed on them by their LTC and risk management strategies used by adults they came into contact with. 165,166,169,173,174 Authors of five camp studies discussed how, owing to adaptations such as the provision of trained staff and specialist equipment to support physical health needs,165,166,173,174,185 CYP were able to take part in an activity commonly available to American CYP without LTCs. This brief opportunity to experience being like any other child may be a novel experience for CYP with LTCs and is discussed further under Theme: a new normal.

The emphasis on how much CYP enjoy fun and engaging activities was acknowledged by 18 studies114,162–168,172–174,177,178,185,187,189,205,206 and could be considered as another important factor that allows CYP to relax and ‘just be a kid’. 173 This is an important component to consider when designing interventions that CYP find engaging, discussed later in Theme: engagement. However, it may not always be appropriate to incorporate an element of ‘fun’ into a mental health intervention. For example, in the case of CYP receiving procedural interventions, the focus of a mental health intervention may be on reducing the distress of CYP enough to support them to engage with the procedure. In addition, three study authors acknowledged that there was the risk of CYP learning information about their illness or those of others that may upset them while interacting with peers with similar LTCs. 163,177,185 For example, one CYP with cancer talks about the uncertainty created through meeting peers at camp who had experienced a relapse in their illness:

It worries me sometimes, because they [campers who relapsed] thought that some day they’d be sure, and they could go on with their lives. And it didn’t turn out that way. So once in a while, I think, ‘Well, maybe I will relapse, and I won’t be around very long.’

Young person, cancer163

Nicholas et al. 177 highlight the importance of having a member of staff to address any anxiety or distress that may be created through peer interaction.

The studies contributing to this theme indicated that CYP appreciated the chance to speak openly, particularly when they did not often have this opportunity within their everyday lives. 167,174 CYP also valued opportunities to escape the routine and pain associated with their LTC through new environments, having fun and trying new activities.

Theme: therapeutic relationships

This theme, receiving contributions from 42 articles (see Report Supplementary Material 2, Table 6), characterises therapeutic relationships as the availability of trustworthy individuals who make CYP feel cared for, respected and valued. These relationships offer CYP the opportunity to share their stories and be listened to in a respectful environment. Studies included in this review indicate that relationships of this quality are not just limited to interactions between CYP and a trained intervention deliverer. This theme also considers the importance of relationships with peers and families in similar situations in helping them to feel secure within the intervention setting.

The authors of seven studies162,166,174,175,183,186,197 note that CYP receiving interventions valued the knowledge that helpful people were consistently available (see Report Supplementary Material 2). This is illustrated by the following quotation from a young person, describing their perception of counsellors at their camp:

That is what camp means to me, I love the counsellors. They are awesome and if you need help just ask them a question and they will help you.

Young person, congenital heart disease (CHD)166

Within the camps included in this review, camp counsellors referred to volunteers, often previous camp attendees,167,173 who provided support and supervised CYP while they completed activities. This suggests that therapeutic relationships can exist between CYP and individuals without specialised mental health training.

Along with interest and understanding,186,190,196,210,219 Report Supplementary Material 2, Table 7, indicates that mental health and well-being interventions provided the opportunity to build empathetic connections with both peers and intervention delivers. 114,183,205 This opportunity is evidenced by a young person discussing an internet-based coping skills programme with a discussion board component:

The scene presented was true . . . but, most of all I liked knowing that other teenagers know how I feel. That made me feel good.

Young person, T1DM114

These empathetic interactions appeared to provide the opportunity for CYP to receive recognition and validation of their experiences. 163,164,172,175,177–179,186 First- and second-order construct data from 10 studies suggest that these connections may elicit feelings of being cared for and valued. 164,167,169,172–174,176,178,185,197 This is evidenced by a young person describing the emotional effect of their relationship with fellow campers:

I would probably take that feeling of like being wanted because I know, like, for some kids . . . they don’t feel as, like, wanted or like people want to be around them. . . . it’s not, like, very often that you’re included in a lot of stuff.

Young person, mixed167

This suggests that it is important that CYP feel that they matter to other people, who can see and value them for the person they are, beyond their LTC. This may help CYP build trusting relationships with their peers and intervention deliverers and to express themselves and thus create the opportunity for them to receive social support from peers, as discussed in Construct: social support.

This opportunity for self-expression and being heard by an empathetic audience may in turn have reinforced the experience of the intervention space as safe and trustworthy. For example, a parent observed the following of the relationship between their child and the deliverer of a movie-making programme designed to reduce CYP’s anxiety while receiving radiotherapy:

She placed a lot of trust in them and she really loved them. . . . And I think because of that she felt safe. She felt she was not going to be let down and that she could go through the process with the people around her.

Parent, cancer206

These therapeutic relationships, including the positive regard of both peers and those delivering the intervention,197 influence both how the child feels about themselves and their coping outside the intervention setting. This is illustrated by a young person discussing how their camp experience could affect them during their everyday life:

The love that I received here made me a better person because I was able to have more independence yet still have a lot of fun whenever I want and always find the bright side of things instead of things that are negative and pessimistic . . . [I] have more confidence in myself.

Young person, mixed167

The evidence within this theme suggests that relationships with empathetic peers, intervention deliverers and family members can play an important role in helping CYP feel validated, cared for and valued. This appears to facilitate the expression of emotions and increase the likelihood that CYP’s psychological needs will be met. As a result, high-quality therapeutic relationships could be considered as a critical component of mental health interventions in their own right.

Summary: therapeutic foundation

The importance of a sense of familiarity, feeling unconstrained, establishing boundaries and developing therapeutic relationships has been discussed in the context of creating a ‘therapeutic foundation’. The creation of such a foundation is something that is needed to form the basis of an intervention. It provides a healing space in which CYP can be themselves, reinforces the engagement of CYP with an intervention and enables them to access social support from their peers and intervention deliverers. In turn, this social support, as discussed within Construct: social support, may reinforce that it is safe for CYP to talk about their feelings and experiences. Having a safe space and supportive peer network can also enable CYP to learn new skills and fresh perspectives to help them manage their LTC that can contribute towards the building of resilience and hope, concepts that are explored within the overarching themes below.

Theme: I am not alone

This theme discusses how CYP value the opportunity to connect with others who are similar to them, and the feelings of safety and belonging that this can provide. Meeting and interacting with peers helped CYP to realise that they were not the only ones living with such difficulties and normalised their experiences. This experience can help reassure CYP and reduce feelings of loneliness and isolation. Forty-six articles contributed towards this theme (see Supplementary Material 2, Table 6).

Six study authors discuss how, for some CYP, meeting people with similar experiences was a novel experience. 162,166,169,172,176,185 Conditions such as cystic fibrosis actively prevent CYP from meeting face to face in groups because of the potential risk to their physical health. 183 Other LTCs such as HIV infection and sickle cell disease can be associated with high levels of stigma, poor understanding and grief within the wider community, which can prevent CYP with these LTCs from connecting with peers with the same conditions. 167,176

Ten studies indicated that CYP valued the sense of being understood and accepted through interactions with peers with similar conditions. 162,163,166,171,174–176,179,216,217 This is illustrated by a quotation from a young person accessing a week-long camp intervention:

At home I’m like Clark Kent . . . but like here, I’m Superman. I’m just like, I’m more open, more talkative, more inspired . . . ’Cause when I’m at my home it’s like I’m hiding from everybody else . . . I just don’t communicate. Camp is more open, it’s just friends all over.

Young person, HIV174

The quotation suggests that feelings of being understood and accepted by similar peers may be related to CYP’s perception that they were free from stigma or judgement from others, facilitating feelings of safety and self-expression as discussed in Theme: unconstrained.

Nineteen studies suggested that the opportunity to connect with others who share their experiences appears to facilitate feelings of belonging in CYP. 114,162,164–166,169,171,172,174,177,179,183,186,187,189,200,202,216,217 This is evidenced by experiences of a variety of interventions, which ranged from structured family art therapy-style or support groups for CYP164,172,176,189 to more unstructured interventions, such as camps:166,169,174,179

Here it’s like a happy home so it’s like, . . . we are all like a family and our grandma lived here, and grandma invited us all over for dinner every 3 months, that’s how it is, like whenever we come it’s like ‘Hey, good to see you’.

Young person, HIV176

The authors of five studies indicate that the communication of experiences with peers or understanding that intervention deliverers can help reduce feelings of isolation within CYP,162,166,171,186,202 as supported by a quotation from a young person experiencing an online peer-mentoring intervention:

When I was doing the chats, I was involved with the conversation, and I wasn’t thinking about [loneliness] . . . [The chats] made me realise that I’m not alone; like with my disability . . . there’s lots of people out there that go through the exact same thing as I do.

Young person, CP/SB170

These reduced feelings of isolation also occurred through use of online interventions,114,177,178,183,216,217 although some parents suggested that the inclusion of an opportunity to meet either before or after commencement of online sessions would enhance bonding between CYP. 178,216

The reduced isolation through recognition that there were other people who were similar to them and the validation achieved by connecting through shared experiences could also be thought of as a form of ‘affirmational support’ as conceptualised by Stewart et al. 178 Stewart et al. 178 describe affirmational support as CYP sharing challenges and solutions and thus receiving validation of their concerns and reassurance that they are not alone and are able to cope well with their LTC.

Theme: emotional and informational support

Forty articles contributed to this theme (see Report Supplementary Material 2, Table 6), which explores how both emotional and informational support are important components of social support experienced by CYP. We used Stewart et al. ’s178 model of social support and their conceptualisation of ‘emotional’ and ‘informational’ support to inform the set of ideas within this theme. In this section, ‘emotional support’ aims to encompass the relief and reassurance experienced by CYP as a result of expressing their feelings to peers who can empathise with them. Access to people with similar stories to tell and experienced intervention deliverers also enabled CYP to receive information and advice on how to manage and cope with their LTC. This informational and emotional support may reinforce CYP’s experience of the intervention setting as a safe space where they can express their emotions freely and enable them to acquire skills and knowledge, as discussed later in Construct: resilience.

Report Supplementary Material 2, Table 7, indicates that the opportunity to be in the same environment and share experiences with similar others enabled CYP to access emotional support and reassurance,162,172–174,176–178,183,207,216,217 as illustrated by this quotation from a participant attending a camping intervention for CYP infected with HIV:

You’re just always getting reassured that it’s OK that you have this [HIV], and it’s just a good feeling.

Young person, HIV174

Children and young people appeared to experience the sharing of their emotions with peers as a form of stress relief174,217 and valued the opportunity to talk about anxieties they worried that parents would not understand. This is illustrated by a young person with T1DM discussing their older peer mentor:

You can share deep experiences and fears that other people won’t understand or don’t know how to calm.

Young person, T1DM168

This suggests that the presence of peers may facilitate emotional self-expression for CYP, owing to their perception that their peers can understand what they have been going through, as discussed under Theme: unconstrained. CYP may perceive that it is safe to discuss difficult feelings if they know others will be able to understand and empathise with them.

The authors of three studies suggest that those delivering interventions have an important role in maintaining the ‘emotional safety’ of CYP during discussion of sensitive topics and reassuring CYP that other people will understand and listen to them. 178,187,189 Peers and intervention deliverers also provided CYP with the opportunity to express their emotions and think about how to manage them:

When I used to get angry, I just handle matters myself. I’d get in trouble. Here I don’t get in trouble. They just talk to me for a while, about like, how it’s not worth it.

Young person, HIV174

In addition to emotional support, CYP also expressed a desire for information on what to expect in the future, including information on medications; activity limitations; variations in disease severity; and possible impact on family, friend and romantic relationships. 168,171,189 This is indicated by a young person discussing what they would like as part of a psychoeducation programme:

I’d like to know what could happen, so you’ve got no surprises later on.

Young person, juvenile idiopathic arthritis (JIA)171

This suggests that informational support gained from individuals delivering an intervention may help to reduce anxiety associated with not knowing what could happen next or how their LTC may impact on their lives in the future. The quotation below from a young person infected with HIV illustrates how they intended to use the knowledge from a peer support group intervention to prepare them to answer questions from others in the future:

. . . I always had to ask how do I get it again, how do I transmit it . . . and yeah cause I always had to ask the second questions because I always have to make sure that I’m at a point where if someone asked me, I’m ready to answer their questions.

Young person, HIV176

Second-order construct data suggest that CYP valued information from same-age peers, older peer mentors and intervention deliverers on how to manage relationship challenges, such as teasing because of appearance, or LTC management issues166,178,216,217 and how to manage issues of disclosure and sexual relationships. 168,172 Listening to how peers managed these difficult situations may help to inspire CYP through the recognition that they may be able to do this too, an idea explored in Theme: hope and inspiration.

The data within this theme suggest that mental health interventions should aim to incorporate emotional and informational support components, as there is evidence that this may benefit CYP’s mental health and well-being. These components may be delivered through contact with peers with similar conditions and/or intervention deliverers.

Theme: improved relationships with family and healthy friends

This theme discusses how an intervention targeting CYP mental health may also improve connections with family members and healthy peers. This may occur through enabling CYP to help their family members and peers better understand their feelings and the impact of the LTC on their lives. Alternatively, interventions can directly target communication styles between CYP and family members or peers. Both of these approaches can improve relationships between CYP and key people in their lives. This may in turn result in increased social support for CYP outside the intervention setting. Thirty-four articles contributed towards this theme (see Report Supplementary Material 2, Table 6).

The authors of four studies suggest that one method of changing dynamics within the family is by altering parenting styles through both family-directed interventions164,184,189 and those focused on CYP. 185 Changes reported by parents as a result of these interventions included improved recognition of protective parenting styles185,189 and increased attempts to focus on aspects of family life other than their child’s LTC. 164

Seven studies suggest that an intervention can provide a distraction from the LTC. 164,177,181,190,201,205,206 One young person discussed how they enjoyed the alternative focus provided by a family art therapy intervention:

It’s a lot of fun. And I get to do things with my mum. And there are things, finally, that we can have fun at doing. When I was going through so much of my sickness, I wasn’t able to do fun things. It’s just a healthy experience.

Young person, mixed164

The idea that CYP appreciate interventions that consider other areas of their life in addition to their LTC is discussed further under Theme: more than just my illness.

Interventions that aim to improve CYP mental health and well-being may also indirectly improve relationships between the CYP and other members of their family. The authors of 12 studies indicated that improved communication between family members and increased empathy for the child with a LTC resulted from interventions aimed at individual CYP184,190,196,200–202,205,206 or peer groups. 172,175,177,181 CYP receiving peer support interventions valued advice from peers with similar conditions, which helped them to resolve communication and relationship issues with family members. 178,216,217 Communication skills training may be an important component of a mental health intervention and be particularly relevant to young people transitioning into adulthood, with the resulting change in family dynamics.

The impact of interventions that directly and indirectly target family communication is illustrated by two quotations below. The first is from a parent describing the impact that attending a creative writing group for CYP infected with HIV had on her child; the second is from a young person discussing the impact that a family intervention to support them with their ABI had on their family:

She opens up more to her mother and her family. She used to be pretty shy, but now she’s like so talkative!

Parent, HIV172

We as a family were able to come together to better understand my brain injury and how I live my life . . .

Young person, ABI189

The above quotations suggest that increased empathy and understanding of the impact of LTCs on CYP and their communication can improve family relationships.

Seven studies reported that CYP and parents noted that relationships with healthy peers could also improve following an intervention:163,172,178,201,202,206,217

Lisa was facing some bullying so she showed the DVD [digital versatile disc] to her class to help them comprehend what was going . . . So as far as coping and being accepted at school I think it helped a lot.

Parent, cancer220

These improved relationships could stem from the creation of an age-appropriate means to tell their story and increase peer empathy (e.g. Baruch200) or through the acquisition of social skills learnt through communication with peers with LTCs (e.g. Stewart et al. 178,217).

The authors of nine studies discuss how the interventions evaluated helped CYP ask for the support they needed from family, peers and practitioners, or improved relationships with health professionals. 164,177,178,191,197,200–202,217 This is illustrated by a quotation from a parent of a young person taking part in an online intervention, discussing how the intervention influenced their child’s relationship with his health-care provider outside the online setting:

[through the online network the ill child] has gotten to know [a health-care provider] a lot better so he feels much freer . . . to go ask her for help.

Parent, mixed177

Thus, interventions that can support CYP to tell their story and engage in supportive interactions with peers and intervention deliverers may enable them to seek support from parents, healthy peers and health professionals when required.

There was some evidence (see Report Supplementary Material 2, Table 6) to suggest that the impact of these interventions on family and peer relationships endured beyond the end of the intervention period,164,174 as illustrated by a parent discussing the impact of a family art intervention:

I’ve changed because it does not have to be the first thing I tell my husband about when he calls me at night – about what’s going on with [ill child] . . . I’ve really tried to make a conscience [sic] effort not to do that . . .

Parent, T1DM164

The integration of changes to family lifestyle and relationships into daily life may be an important component for maintaining the effects of the intervention on CYP’s mental health over time. This will be explored further in Theme: keeping it going. The potential improvements to family and peer relationships in terms of increased empathy and understanding, which are achieved through enhanced communication, may have a significant impact on CYP’s mental health and well-being in the long term.

Summary: social support

The affirmational, informational and emotional support described above are all components of social support as described by Stewart et al. 178 The social support received by peers may facilitate their experience of an intervention setting as a safe space, which in turn may increase the likelihood of receiving emotional and informational support. Support to share their experiences can result in feelings of reduced isolation and improved relationships with healthy peers and family. This may have important implications for extending the impact of an intervention on CYP mental health and well-being beyond the intervention setting. Social support seems to be an important mechanism that may facilitate the development of new skills and knowledge and help CYP create a more hopeful view for the future, as discussed below.

Theme: hope and inspiration

This theme discusses how CYP value having a sense of hope while living with a LTC. Hope and inspiration can be provided through access to role models who are living successfully with a LTC. Alternatively, mental health and well-being interventions may provide opportunities to experience success and mastery by encouraging CYP to persevere with challenging situations. A sense of inspiration may also be elicited by interventions that encourage CYP to focus on positive activities. Thirty articles contributed towards this theme (see Report Supplementary Material 2, Table 6).

Twenty studies focused on CYP’s interactions with peers who have an LTC as a source of encouragement, motivation and inspiration. 162–165,167,168,171–175,181,183,185–187,201,214,216,219 CYP appeared to find it particularly powerful seeing an example of someone older who was leading a fulfilling life despite their illness,162,178,216 as illustrated by the recipient of a peer-mentoring intervention below:

It was fun like it’s never been before, I saw someone who is in control of the situation and isn’t afraid of diabetes. If he can do it, I’ve got to succeed.

Young person, T1DM162

The advice received through contact with peers of the same age in both formal and informal settings also showed that it was possible for others in the same position to overcome difficulties associated with the LTC or challenges within the intervention itself:

. . . makes you want to do something and get up and go for it. Because you see all these kids, the kids in wheelchairs and they’re doing all this stuff, and they’re not letting anything get in their way.

Young person, cancer173

I saw that the other girls were fine with exercises I was like, ‘OK you are worrying over nothing’.

Young person, fibromyalgia186

Role models did not have to be someone CYP had met personally. Barlow et al. 171 suggest that CYP could watch vignettes of peers online to observe people who are managing to live successfully with the disease. Role models may also help encourage CYP to think about goals for other areas of their life, outside managing their LTC, as suggested by a young woman discussing important components for a group intervention for young women infected with HIV:

Ask them who their role models are. They could be somebody famous or something like that. Be like, you see how they work hard to get that, you want to do that for yourself. You want that life for yourself.

Young person, HIV168

This quotation suggests that feelings of hope and inspiration may encourage CYP that they too can affect positive changes within their own lives outside the interventions setting. This interpretation is supported by three studies whose authors suggest that the inspiration provided by peers can also provide CYP with the motivation to persevere with intervention activities or physical health procedures they perceive to be difficult. 184–186 This is illustrated by a parent discussing their views on the impact that attending camp will have on their child:

I think that [friendships among children with CHD] will inspire him . . . when he sees other kids do things that he might be afraid of, I think it will help him to take that leap of faith and inspire him. And he might think ‘oh well let me try that’.

Parent, CHD185

The inspiration provided by peers may encourage CYP to achieve success within challenging situations. The potential feelings of empowerment and improved self-esteem that may result could be considered as important protective factors against mental ill health, as discussed further in Construct: resilience. However, there was some evidence that CYP found talking about their emotions with peers uncomfortable1,92 and advice from peers was perceived as irrelevant if CYP felt that they were in a worse position than their peers. 171

The authors of four studies suggest that activity-based interventions that provide CYP with an opportunity to receive positive feedback, such as receiving beads following a painful procedure,200 or provide an alternative focus through taking part in a creative activity can provide CYP with a sense of comfort and positivity. 197,201,204 This positive feedback may inspire CYP to continue with challenging procedural treatments and/or have a beneficial effect on their mental well-being.

The evidence above illustrates how hope and inspiration can be created through contact with peers with a similar illness whether face to face, online or by video. Interventions that provide CYP with a source of hope may also inspire increased feelings of self-confidence that they too can live a fulfilling life with a LTC.

Theme: more than just my illness

This theme draws on data from 33 articles (see Report Supplementary Material 2, Table 6) and discusses how CYP value interventions that consider more than just their LTC and explores to what extent certain interventions included in the review tried to meet this need.

Eleven studies indicated that CYP valued interventions that focus on other aspects of their lives in addition to their LTC. 114,167,168,174,176,178,184,187,195,205,206 In addition to addressing psychosocial or educational needs162,171,177,183,189 and acknowledging how CYP’s physical and mental health could affect one another,114,186,195 interventions also created opportunities for CYP to exercise control and self-expression. 189,197 However, there was also some evidence that some interventions failed to adequately address CYP’s psychosocial needs. 171,178,182,183,216 Below, a young person describes what she would like to see as part of a group intervention for young women infected with HIV:

My perfect programme wouldn’t just not only [sic] be focused on the infection. It would just really be building self-worth, building self-esteem like all the way around.

Young person, HIV168

Second-order construct data indicate that CYP appeared to particularly value interventions that lead to them feeling recognised as a ‘normal’ person,206 and let them explore other aspects of their identities. 167,169,174,205,217 One intervention deliverer discussed how CYP used a peer support group to explore how they wanted to incorporate HIV into their identity:

So really a means for them to figure things out on their own and figure out how they wanted to represent themselves and HIV or if they wanted that to be a part of who they were.

Group leader, HIV172

This suggests that CYP value the opportunity to explore who they are in addition to their LTC, and services that reflect their additional needs.

The authors of five studies indicated that interventions that affect or involve other family members may also have a positive impact on the mental health of the CYP by addressing family interaction styles:171,177,185,189,206

My mum and dad get stressed out too and I learned some things to try to calm down when I am frustrated.

Young person, ABI189

Consideration of the impact of family on CYP’s mental health ensures that CYP are not burdened with the expectation to manage their LTC and psychosocial needs by themselves. It may also encourage family members to access appropriate support for their own needs and enable them to support their child. This could be a relief for CYP and may lead to increased empathy and understanding between family members, creating the opportunity for improved family relationships, as discussed in Construct: social support.

Theme: a new normal

This theme explores the importance of an intervention enabling CYP to continue taking part in day-to-day activities, while accepting any limitations resulting from their LTC. The impact of CYP being able to observe how other CYP cope and adjust to life with a LTC is also discussed. Thirty-nine articles contributed towards this theme (see Report Supplementary Material 2, Table 6).

Thirteen studies found that CYP value interventions that enable them to take part in everyday activities, whether that is household tasks or doing things that friends or family members without a LTC also do. 165,167,169,171,173,177,181,182,185,186,196,197,201 Below, a parent talks about the importance of a karate class that her child with cerebral palsy could attend with his healthy brothers:

There are things that go on that normal kids with no disabilities do that he can enjoy also. This helps him out the most. His brother and other kids can take karate, and so can he.

Parent, cerebral palsy181

Online interventions also help CYP to have a link with their normal life, as illustrated by one parent who talked about how an online environment allowed their hospitalised child to engage in fun activities as he would at home:

Because he can’t always go out and play, I think to be able to do it vicariously through the computer has been really good for him.

Parent, mixed177

The data above suggest that interventions that reduce CYP’s sense of being different may help them feel more ‘normal’ and reduce their sense that they are excluded by their LTC. However, O’Callaghan et al. 204 suggest that some CYP may feel stigmatised by being asked to take part in a mental health and well-being intervention, which may act as a barrier to participation.

Eleven interventions supported CYP to accept the limitations that their LTC imposes on their lifestyle and to acknowledge that, although their LTC is a part of their lives, it does not necessarily need to take over. 162,165,169,172,176,180–182,186,187,202 Below a parent discusses how camp could help their child to realise that having a LTC does not need to stop him from doing the things that he wants to do:

[interacting with children with CHD] is good for Adam to see that just because you have [CHD] doesn’t mean that you can’t live a full life . . . the activities that they will do together will just help them to see that you can just be as normal as others kids . . .

Parent, CHD185

In contrast to the above, parents from two studies evaluating interventions to relieve procedural distress in CYP with cancer reported that the interventions could provide CYP with the opportunity to come to terms with their own mortality. 201,202 For CYP with a terminal illness, an intervention should focus less on creating hope and more on achieving a sense of peace and acceptance.

Having the opportunity to meet or observe peers with similar difficulties to their own meant that CYP were able to compare the impact the LTC had on the lives of their peers and the coping strategies used by others to their own. Sometimes this encouraged CYP to develop a new, more positive, perspective on their personal circumstances. 162,164,167,169,171,173–175,177,178,185,187,200,201,216,217 This is illustrated by a young person with JIA comparing themselves with peers they saw in a psychoeducational video:

Well, I haven’t got it in many places so I thought that I was lucky because other people have got it worse.

Young person, JIA171

Despite the apparent desire to feel normal, second-order construct data from six studies suggest that, at times, CYP may enjoy the opportunity to feel ‘special’ as a result of an intervention. This may be through the recognition that an intervention is specifically for them181,191 or through recognition by others that they are ‘brave’ or ‘special’. 166,173,200,206 This is a possible extension on the typical developmental need to feel valued by others.

Theme: mutuality – I have something to offer to others

This theme focuses on the realisation that some CYP experienced while receiving an intervention that they too had something to bring to their relationships with peers or those delivering the intervention. This included input on setting up new interventions or providing information, advice or empathy. This helped CYP view themselves differently, as an equal part of the relationship, instead of passive recipients of care. This has implications for CYP’s feelings of empowerment and self-esteem. Thirty-three articles contributed to this theme (see Report Supplementary Material 2, Table 6).

As later acknowledged under Theme: engagement, CYP and their families have a wealth of experience that can be called on when designing an intervention. 114,168,170,171,175,184,191–193,211,218,219 CYP draw on their knowledge and experience to provide social support to their peers,178,216,217 as illustrated by an adolescent discussing how they were asked to provide a peer with some informational support at camp:

His parents wanted me to . . . give him some advice and help him relate to how it is going to be. That really makes me feel important . . . I get to share my life experience and relate to him with his life experience.

Young person, CHD166

Children and young people were able to recognise that they were able to provide support to both peers with a LTC and the intervention deliverers who had been asked to support them,162–164,166,186,217 as highlighted by a young person talking about their relationship with their peer mentor:

I know him no less than he knows me, I help him, he doesn’t only help me.

Young person, T1DM162

This represents an alternative, more positive, way of viewing their LTC and the impact it has on their identity, and indicates that CYP can have an impact on the lives of others. This is a possible alternative identity that may offer a source of hope for the future for CYP and act as a protective factor against mental ill health. The recognition that they too can offer beneficial advice to others can have a positive impact on how CYP see themselves within the context of their relationships. The LTC is no longer the reason for needing help from others; rather, it is also a rich source of experience to draw on so that the CPY can be ‘the helpers’ instead. This may improve CYP’s self-esteem and sense of empowerment and have a beneficial impact on their mental well-being.

Data shown in Report Supplementary Material 2, Table 7, indicate that by recognising that they too had something to bring to their personal relationships, CYP were empowered in their interactions with people outside the intervention setting, exemplified by teaching others new skills119 or challenging the stigma narratives around their illness:

It has influenced me to become more active in AIDS [acquired immune deficiency syndrome] and HIV [sic] in teaching others about AIDS and HIV, becoming more open about my disease because I do live in a community where I have certain people around me who will not judge me for my disease.

Young person, HIV172

The realisation that they too have something to offer others may facilitate behaviour change and give CYP a voice within the wider community. The impact of having their voice heard within an intervention is discussed in Theme: empowerment.

Summary: a hopeful alternative

This construct encapsulates how valuable CYP find the hope that it is possible for them to live fulfilling lives with a LTC. In addition to the importance of social support and the positive influence on peer mentors in providing a source of inspiration, it is also important for interventions to focus on more than just CYP’s physical illness. Interventions should consider how CYP can be encouraged to take part in activities that make them feel ‘normal’ despite their physical limitations. CYP also appear to value the opportunity to recognise that they too have something to offer other people and are not just the passive recipients of help. These factors can help CYP to feel empowered and to develop positive self-esteem, factors that may contribute towards improved mental well-being, discussed in Construct: resilience, and may help CYP develop and maintain links with peers and family, as discussed in Construct: social support.

Theme: managing myself

This theme explores how being taught skills and knowledge to manage both their physical and mental well-being can enhance CYP’s ability to cope with difficult life circumstances and lead to a sense of empowerment and improved self-esteem. Forty-nine articles contributed to this theme (see Report Supplementary Material 2, Table 6).

Twenty-three studies indicate that CYP seemed to find it helpful to have the opportunity to learn skills that would allow them to take responsibility for managing their LTC as part of an intervention aiming to improve their mental health. 114,119,162,163,166,168,171,172,174,176–178,182–184,186,190,191,193,195–197,214 Such learning appeared to take place through interventions in one-to-one, group and online settings and increased CYP’s confidence to manage their LTC in the future:192,199,211

Learning the techniques helped my asthma and me as a person, because if I know how to control my asthma now, I can help in the future and stop things from occurring.

Young person, asthma119

One CYP receiving CBT described how the skills they learned provided them with a sense of control over their fibromyalgia:

It gave me something to work with, which is something I haven’t had since I was in school.

Young person, fibromyalgia182

Three studies highlighted how learning how to manage their LTC and what to expect in the future could help CYP feel prepared and increase feelings of self-efficacy, mastery and being in control. 168,171,177 This may have important implications for helping reduce CYP’s anxiety around not knowing if they will be able to cope with physical, emotional and social challenges arising from their LTC.

Children and young people and their parents also appreciated skills learnt during mental health and well-being interventions that directly enabled CYP to manage stress114,119,162,171,177,183,184,189–191,196,200,205 and in some cases showed that they were able to use these new skills when required during their day-to-day life,178,182,186,190,196,216 as illustrated by two participants of a neuromuscular and CBT training programme:

Mini-relaxation I would do all the time when I felt a little tense . . . muscle relaxation I would usually do in the morning or at night before I went to bed.

Young person 1, CFS186

Pleasant imagery I usually did before bed or if I was having a lot of pain . . . calming statements . . . I did mostly when I felt . . . anxious or when I couldn’t sleep.

Young person 2, CFS186

Interventions that acknowledged the interaction between physical symptoms and mental well-being seemed to be appreciated,114 particularly for LTCs such as fibromyalgia and CFS. 186,195

Acquiring skills to manage challenging social situations seemed to be particularly important for CYP living with LTCs associated with a high degree of stigma, such as HIV infection for which there were issues around managing intimate relationships and disclosure. 168,176,187 Hosek et al. 168 highlighted a demand for group interventions that support young women who have a HIV infection to manage the emotional distress resulting from negative consequences following disclosure:

It doesn’t mean that because I told my friend that I’m HIV positive she’s gonna accept me. It doesn’t mean that because I told her I have to accept that she has to accept me . . . She’s out of my picture. So how do we deal with it?

Young person, HIV168

Eleven studies included in this review indicated that distraction can be a valuable skill that CYP can use to cope with distress and pain and escape the boredom or uncertainty associated with long hospital stays. 164,169,173,174,177,191,201,202,205–207 The evidence presented within this theme suggests that gaining skills to managing their LTC, social situations and distressing procedures through a mental health intervention can help CYP to take responsibility for managing their physical and mental well-being.

Theme: empowerment

Forty-seven articles contributed to this theme (see Report Supplementary Material 2, Table 6), which discusses how mental health and well-being interventions can empower CYP to take responsibility for their own health-care needs on day-to-day and health-care settings, enable them to have a voice within their local community and increase their confidence in managing tasks of everyday living. CYP can feel empowered as a result of an intervention that includes the incorporation of activities that allow them to experience mastery and achievement, and by having their voices heard within an environment that encourages reflection on progress. The experience of feeling empowered within the intervention setting can have an impact on how able CYP feel to effect change in their day-to-day lives and is thus an important component of resilience. Feeling empowered can also affect how CYP view themselves in relation to others and thus relates to the ‘mutuality’ theme above. Other third-order constructs that contributed towards this theme can be seen in Report Supplementary Material 2, Table 8.

Evidence from 11 studies suggest that interventions can offer opportunities for CYP to make choices and exercise autonomy by deciding how they would like others to interact with and support them. 162,170,173,177,178,183,185,195,197,198,207 An example of this is provided by a parent describing how their child instructed staff on how best to support her while she was undergoing needle insertions:

[The child] wrote out a needle plan, and she talked it through at home and worked it out with the psychotherapist here at the hospital . . . every time there was a new nurse or doctor she’d hand the card to them and they had to read it . . .

Parent, cystic fibrosis198

It appeared to be important for interventions to provide opportunities for CYP to experience having their voices heard. This could be achieved through allowing CYP to vote on important decisions as part of the wider group and giving them the chance to tell their story. 171,173,176,181,197,198,200,202 These experiences enabled children to make their voices heard within the wider community. 164,172,176 One parent observed how a camp for CYP with CHD allowed their child to develop confidence in her ability to speak up for what she needed later while attending school:

Since going back to school . . . she knows that from camp as well, that she has to be an advocate for herself. That she’s the one that has to kind of put the foot down and say ‘this is why I need to stop and I’m going to stop’.

Parent, CHD185

Data from Report Supplementary Material 2, Table 7, also indicate that activities incorporated into interventions that promote recognition of achievement and mastery of challenges may help CYP gain a sense of confidence in their abilities. 119,165–167,169,171,173–175,177,179,185,186,198,200,201,205,206 This is supported by a quotation from an intervention deliverer from a camp intervention:

For some of these people, they’re very sheltered and their parents don’t let them do anything, and they’re climbing rock walls and going down zip lines, and the fact is that they’re overcoming their fears and obstacles.

Intervention deliverer, cancer173

Having this sense of confidence in their own abilities may enable CYP to feel empowered outside the intervention setting. Some of the outcomes that were less directly related to mental health included the willingness of CYP to become more involved in their care,198 to take responsibility for completing homework exercises119 and to complete other tasks in order to manage their health needs. 119,162,171,176,178,184,185,187,195,197,216 For example, in a study that examined the clinical effectiveness of breathing retraining for asthma, one adolescent reported that it was they who took responsibility for remembering to do the exercises:

Nothing has gotten in the way, I just remember, no one reminds me.

Young person, asthma119

Overall, this theme illustrates how mental health interventions that offer opportunities for CYP to exert control, gain experiences of mastery and feel that their voices are heard can enable them to feel empowered within their everyday lives beyond the intervention setting. This may contribute towards CYP’s development of ‘resilience’ and a positive view of themselves.

Theme: empowerment

This theme details how improved self-esteem may contribute towards the development of resilience. A young person’s self-esteem may be improved by an intervention that encourages them to recognise their capabilities and areas of achievement, as well as providing opportunities to interact with others who can reflect the positive parts of their identity. Twenty-seven studies contributed to this theme (Report Supplementary Material 2, Table 6).

Data from four studies indicate that taking part in those interventions reinforced CYP’s belief in their own ability to manage their condition, emotions and negative responses from other people. 178,185,197,206

The authors of six studies suggested that CYP value the opportunity to recognise positive things about themselves. 164,168,179,186,189,190 One study encouraged CYP to practise making positive affirmations about themselves during the day, and there was some evidence to suggest that CYP could then use this to ameliorate emotional distress in other settings:

Thought about what I was proud of or what made me happy, which helped me to calm down and be happy again.

Young person, T1DM190

The above quotation demonstrates the impact of CYP having the opportunity to learn skills to manage their emotions on their sense of well-being and suggests that opportunities to build CYP’s self-esteem can provide a valuable resource for CYP to draw on in times of emotional need.

The authors of 17 studies suggested that positive recognition and affirmation from peers, family, intervention deliverers and themselves were found to be important for improving CYP’s self-esteem and/or mood. 114,162,164,166,173,178,179,181–183,185–187,189,200,201,206 One parent felt that their child found it valuable to recognise that he was capable of offering support to peers through a computer-based intervention and that other people also recognised this:

I think it encouraged [the ill adolescent] . . . to know that he had the ability to ‘do’. I think it really boosted his confidence, knowing that he had something to offer other kids, and that someone saw that while he was on the computer.

Parent, mixed177

The authors of three studies indicate that opportunities for CYP to tell their story to others help CYP communicate their experiences and receive positive recognition from others. 200,201,206 Kashikar-Zuck et al. 186 indicate that CYP appreciated the tailoring of intervention tasks to their individual abilities, alongside the opportunity for progression through stages of increasing difficulty. One quotation suggests that this structure provided CYP with the chance to reflect on their achievements with pride:

I am glad that I went through it . . . now I can look back . . . at how far I have come . . . today that stuff, wasn’t hard. It just gives you this sense of pride, like if I can overcome that, what else can I overcome?

Young person, fibromyalgia186

The importance of the individual delivering an intervention being able to relinquish control in an intervention setting to help CYP recognise that they too have something to offer is highlighted in Report Supplementary Material 2, Table 7, under Theme: empowerment. The below quotation highlights the emotional impact on CYP if they perceive an intervention deliverer as being unable to acknowledge their views as equal to their own:

I spoke to [therapist’s name] about it . . . and I said to him that it wasn’t working and I didn’t know why, he basically told me it was my fault and that if it wasn’t working it must be because I wasn’t doing it properly.

Young person, CFS195

This quotation suggests that this young person was left feeling blamed and criticised when they felt that the intervention deliverer had not respected their views. This highlights the need for empathetic, supportive, therapeutic relationships with intervention deliverers, as discussed in Construct: a therapeutic foundation, and the potential impact of these relationships on CYP’s self-esteem.

Second-order construct data from four studies indicate that having the opportunity to reflect on what they have been through may be meaningful for CYP. 167,173,174,200 Such reflection appeared to be facilitated through the receipt or creation of ‘mementos’ documenting their LTC and treatment story and the chance to observe change over the course of an intervention. Combined with the positive responses these mementos elicit from others, they may reinforce CYP’s self-concept as someone who is brave, special and resilient. This may bolster their sense of self-esteem, which could be a protective factor against the development of mental ill health.

Having the opportunity to achieve a sense of mastery by successfully completing activities not only empowered CYP, as described above, but also improved their mood181,186 and/or increased their sense of self-esteem:167,177,179

I have to say my favorite [sic] memory from camp would probably be climbing the tower. It was an incredible experience. When I got to the top I felt like I accomplished, you know, what I wanted to do. The zip line was probably the best part also.

Young person, mixed167

Quotations from five studies suggest that interventions that encouraged focus on areas of ability within the individual child, rather than what they were not able to do, were also valued. 166,168,181,185,201 This is illustrated by a parent describing the importance of a karate class to provide an example of what they could do for their CYP with cerebral palsy:

I don’t want his sister or brother to be in a private karate class. This has really helped him a lot. Even during the week when they do things that he can’t do, we mention the karate thing. I tell him that he does karate and they don’t.

Parent, cerebral palsy181

Opportunities that allow CYP to feel capable and included may boost CYP self-esteem by supporting them to develop a more hopeful view of their capabilities and of the future, as described in Construct: a hopeful alternative.

This section discusses the role of mental health and well-being interventions that encourage positive recognition and experiences of mastery in boosting the self-esteem of CYP. This may result in CYP having a more positive image of themselves, which may act as a preventative factor against experiencing symptoms of mental ill health.

Summary: resilience

This theme explored three inter-related components that may promote resilience within CYP with LTCs. Theme: managing myself detailed how it is necessary that CYP are supported to gain the relevant skills and knowledge they need to be able to take age-appropriate responsibility and have a voice within the intervention process and the wider community. This knowledge transfer can occur both through formally ‘taught’ interventions and by more informal social learning through contact with peers in group interventions. Having the appropriate skills may facilitate the development of ‘empowerment’. In Theme: empowerment the importance of allowing CYP opportunities to make choices about the care they receive and take responsibility for managing the different areas of their lives affected by their illness was explored. The recognition of both their success in managing their LTC and achievement in other areas of their life facilitates a sense of positive self-esteem, the third theme conceptualised here as being necessary to promote resilience.

Theme: availability

This theme draws on 30 articles (see Report Supplementary Material 2, Table 6) and centres on the need to recognise the mental health needs of children with a LTC and ensure the availability of relevant resources to address them.

The authors of three articles acknowledge the difficulty that health services face in identifying CYP and/or their families in need of support to maintain their well-being. 162,171,187 One parent expressed frustration at the difficulty of getting the mental health needs of their child recognised by health professionals focusing on physical symptom management:

They really don’t understand the problems you have at home . . . The doctors see them for half an hour . . . and examine the joints . . . but, it doesn’t just affect their joints. It affects them mentally and that’s what we have to put up with.

Parent, JIA177

In seven studies it was reported that participants commented that, when mental health interventions are provided, the availability or range of options offered is sometimes limited:168,171,175,176,186,189,197

[The intervention] wasn’t easy to find. I was looking for some support for [my son] because he was struggling last year . . . They said they only had one social worker for the outpatient department and it was all very difficult for them to see him.

Parent, mixed175

The authors of two studies highlight the need for staff expertise to design and set up an intervention114,171 and to ensure that a suitable space for the intervention to take place in is available. 114,176 The authors of a further nine studies highlighted the need to ensure that staff delivering the intervention have the appropriate skills, experience and training so that the intervention is delivered as planned and that the emotional or behavioural needs of the CYP are met. 162,166,171–174,178,189,197

Second-order construct data from three studies suggest that, when mental health and well-being interventions are available, some CYP may not feel able to seek support for their well-being. 162,182,207 This may be for a variety of reasons: a desire not to discuss their anxieties or their LTC in front of parents,162 fear of physically hurting someone by holding their hand while undergoing painful procedures207 or the view that the setting in which they are being seen is not appropriate to their perceived needs. 182 Overall, this theme highlights the need for CYP’s mental health needs to be recognised by staff that they may come into contact with and relevant interventions to be provided to CYP in a timely manner.

Theme: accessibility

This theme focuses on factors that may affect whether or not CYP and their families are able to access an available intervention. Factors that may facilitate access include pre-existing relationships with health and social care staff, ensuring that interventions consider the individual needs of the CYP and addressing practical issues such as ensuring that the intervention setting, duration and timing are convenient. Fifty-two articles contributed to this theme (see Report Supplementary Material 2, Table 6).

Seven studies suggest that both health and social care professionals play an important role in identifying CYP who may benefit from existing interventions, ensuring that families are aware of the support available and inviting them to participate in the intervention. 166,169,172,175,177,187,197

Approaching CYP directly via text message or letter was an effective means of supporting them to access an intervention,175,187 and CYP may observe others taking part and wish to participate. 172,177,182,205 There was the suggestion that more formal processes could ensure that interventions are accessible by those who need them, including introduction of referral systems197,211 and ensuring that CYP from certain ethnic minorities or with more severe illness are able to attend. 169 Parents felt that the means of introducing mental health interventions relevant to their child should be incorporated into inpatient discharge processes:

You’ve got the teachers, you’ve got the wards, you’ve got the social workers. It should be just part and parcel of coming around on discharge, you know that you are aware that there is support for young people, maybe not now, but when it may be appropriate.

Parent, mixed175

The idea that the unpredictable nature of a LTC can prevent CYP from accessing, or participating fully in, treatment for their mental health was recognised by 14 studies. 164,165,173–175,177,182,183,185,186,191,201,202,205 Therefore, the recognition of physical health needs by those planning mental health and well-being interventions and consideration of how the intervention can be altered accordingly is required to ensure that the intervention is accessible to those who need it. 173

By ensuring that CYP’s physical health needs are met, intervention deliverers and services can reduce some of the anxiety around attending an intervention, which may help it to be perceived as a safe space for CYP and their families.

Facilitators in online interventions may also play an important role in stimulating discussion and ensuring the relevance of the material to those taking part. 114,178,216 This may help CYP who are less confident access intervention material and maintain their engagement. Six studies suggested that the communication skills and understanding of those delivering the intervention are also key in making the material accessible. 162,166,185,186,195,200 One young person highlights how they felt that the person delivering the intervention did not understand the material and this meant that the information delivered was not fully understood:

The physiology stuff at the beginning I didn’t find helpful because I don’t think it was fully explained, ’cause to be honest I don’t think they fully understand it themselves.

Young person, CFS195

This has important implications for CYP acquiring the skills and knowledge required to manage their physical and mental well-being as described under Theme: managing myself. First- and second-order data in Report Supplementary Material 2, Table 7, suggest that interventions within this synthesis used different ways to ensure that their material was accessible through considering the age, educational and developmental level of the CYP. 114,119,164,168,171,174,182,184,187,191,197,200,201

Second-order construct data from 13 studies also suggest that the timing of the intervention and choice of venue need to be considered to ensure that the intervention does not interfere with CYP’s education or parents’ employment and that the duration and financial burden of travel does not prevent attendance. 119,164,172,175,176,182,183,186,187,191,195,201,216 The usefulness of co-ordinating delivery of physical and mental health interventions was highlighted,176,191 and is illustrated by a quotation from a young person who had attended a hospital-based support group:

It’s convenient and just you can do everything all in one day and see everybody in one day, and just get everything done.

Young person, HIV176

Considering CYP’s physical, psychosocial and family needs together may not only reduce the burden of attending multiple appointments, but also reinforce that CYP and their families have other needs in addition to their LTC and are part of a wider family system, as discussed in Theme: more than just my illness. This may help CYP feel validated as a whole person and make them feel supported and understood by services, necessary components to build a therapeutic foundation.

Interventions delivered online can overcome some of the practical issues around access. One of the perceived benefits of such interventions was that they enabled children to access the intervention independently and overcome the physical limitations of their illness, although sometimes technological issues hindered this:177,178,183,216

[a window with a picture of] a dog comes up and says that they’re unable to locate their site or for safety purposes it’s been blocked. And a lot of the time they go to put on a game and they can’t get it to work. So that gets frustrating.

Health-care provider, mixed177

However, the authors of six studies suggest that CYP and their families would like flexibility in terms of how an intervention is delivered. 114,168,178,179,189,190 Parents acknowledged the convenience of interventions supported by technology, but stated that they would also like CYP to have the opportunity to meet peers and/or intervention deliverers face to face. 178,189,190,216

Care is required that the intervention itself is not too burdensome for CYP or their families. The authors of four studies suggest that the pace and duration of interventions should be considered to ensure that interventions are not too long, yet allow enough time for the material to be covered without overwhelming participants. 164,178,186,195 The quotation below describes how a particularly long intervention session became too burdensome:

It was . . . really taxing the first day because it started at 10 and didn’t finish until about 7 in the evening, and there were also about seven other people there which is like . . . considering I had been like housebound for years I hadn’t really . . . saw anyone . . . I was absolutely shattered.

Young person, CFS195

This quotation suggests that a balance is needed between the convenience of holding multiple appointments at the same time and ensuring that CYP and their families do not feel overwhelmed.

The perceived burden of sessions may be affected by the type of LTC experienced by CYP as well as the stage of their illness. Studies in this review support this idea by discussing the perceived importance of ensuring that a mental health and well-being intervention acknowledges that the LTC may affect the mental health of CYP and address this by considering LTC-specific challenges or worries166,167,171,173–175,182,184,186,187,189,197,205 and stage of illness. 171,178,180,184,187,189,195,197,201,205,217 An example of the role of the intervention deliverer in ensuring that exercises and activities are within the physical capabilities of CYP is given below, within a cognitive–behavioural neuromuscular training programme for CYP with fibromyalgia:

[The exercise physiologist] would work with you personally to make sure you were comfortable with where you were . . . make sure the exercise wasn’t too easy or too hard.

Young person, fibromyalgia186

The role of the intervention deliverer in establishing boundaries to ensure that participants feel safe enough to discuss emotive subjects is discussed in Construct: a therapeutic foundation. Having opportunities to chat before the interventions and including activities that CYP and their families find enjoyable also appears to make the more formal therapeutic aspects of intervention accessible. 164,204 As an example, the art activities provided as part of a family intervention seemed to provide an alternative focus and aid disclosure:

Was Living Well what you expected it to be?

Actually, it wasn’t. And that’s good. I thought about everybody talking about their kids would be kind of depressing, but it turned out, the artwork and everything kind of opened everybody up.

Parent, mixed164

The authors of five studies acknowledged that the interventions being evaluated were not able to address CYP’s every need or replace other interventions that may be useful. 114,178,182,183,189 This highlights how CYP with LTCs have other needs in addition to their physical health, as discussed under Theme: more than just my illness.

This theme summarises different factors that may influence how accessible CYP and their families find an intervention and the role of the intervention deliverer in adapting the intervention to ensure that it meets CYP’s psychosocial needs.

Theme: engagement

Although the availability of an accessible intervention is necessary for its provision, CYP may not necessarily want to engage. Different factors that may influence engagement with an intervention are explored next. Fifty-four articles contributed to this theme (see Report Supplementary Material 2, Table 6).

The role of the intervention deliverer in adapting the intervention to ensure its relevance to CYP is discussed in Theme: accessibility. Report Supplementary Material 2, Table 7, indicates that there are multiple ways in which mental health and well-being interventions can be flexibly delivered according to the needs and individual characteristics of CYP. The authors of six studies suggest that interventions that consider the individual behavioural, social and emotional needs are valued by CYP and their families. 168,173,179,181,182,202 Interventions included in the review also considered stage of illness,171,187,189,195 gender162,168,169,184,197,205,217 and the individual interests of CYP taking part in the intervention162,172,177,182,191,201,206 as factors that could ensure a mental health intervention’s relevance to CYP and help sustain their engagement with the intervention.

In contrast, the authors of seven studies suggest that CYP can engage with an intervention at a level that meets their needs, sometimes with no formal adaptation of the intervention process. 162,163,167,174,177,183,206 For example, Gillard and Allsop167 discuss how a camp intervention appears to meet differing needs among CYP, as influenced by their type of LTC. This fluid interaction of CYP with the interventions may be because the design of the interventions allows for it. CYP can choose which aspects of the intervention are relevant to their needs and access the intervention at this ‘level’. Not only can this be empowering for CYP, as discussed above, but it may also mean that some interventions could meet the mental health needs of CYP as they change over time. This idea is explored further in Theme: keeping it going.

Explaining the rationale or theory behind an intervention can also facilitate the engagement of CYP with an intervention. 167,195,205 One young person discusses how helpful they found understanding the rationale behind an intervention to treat their CFS:

The explaining of the process, the background information definitely helped because it helps you realise how the process works, and there is no point doing something unless you understand how it works . . . otherwise you won’t believe in it enough.

Young person, CFS195

Both CYP and their families appeared to engage with interventions that were evidence based or developed through consultation with other families, as this appeared to give the interventions validity and made them worth ‘buying into’:

This would be a very good tool. [This has] techniques developed through research with other families and [would be] very beneficial.

Parent, cancer191

The above suggests that CYP and their families feel able to trust and perhaps put their faith in interventions that are tailored to meet their needs and ‘make sense’. This could be considered an important part of knowing what to expect from an intervention, as discussed earlier in Theme: a safe space.

Second-order construct data from Report Supplementary Material 2, Table 7, indicate that CYP and their families can have a variety of different expectations prior to the beginning of an intervention182,195,196 and the extent to which these expectations match their experience of the interventions may affect how effective the intervention is perceived to be. 195 CYP expressed anxieties about encountering new health professionals, not knowing anyone else or what to expect and a fear of not knowing what to say to someone who was going through difficult experiences or dying. 163,164,182

Most CYP’s concerns appeared to dissipate over the course of the various interventions. 175,179,181,186 Some CYP and authors of studies believed that an initial tolerance of anxiety, discomfort or frustration was necessary to both begin and maintain engagement with an intervention:171,172,177,186,197

Don’t be shy because if you’re shy you miss out on a lot.

Young person, HIV172

The below quote from a young person attending a camp intervention aimed at CYP with CHD emphasises that it is important that CYP are given enough time to develop relationships with peers:

It’s not long enough, you know we don’t get enough time to bond, but it’s like, ‘how much can you do in 1 week right’?

Young person, CHD169

First- and second-order construct data from Report Supplementary Material 2, Table 7, suggest that fun or pleasant activities appeared to be important in motivating CYP, particularly those who are younger, to engage with an intervention. 114,162–168,172–174,177,178,181,185,187,189,197,201,205,206,216 Other components that enhanced engagement with, or adherence to, an intervention included the provision of a meal during the session, the receipt of rewards or ‘gifts’, and the opportunity to keep a product of the intervention as a memento or display it for others to see. 164,173,174,187,190,200,201,211

Some evaluations of online support interventions acknowledged the importance of clear website layout and attractive graphics to maximise user engagement. 177,190,218,219 A quotation from a young person evaluating an online self-management programme for CYP with JIA emphasises how large amounts of text can be off-putting for young people:

Like this, it looks like a lot of stuff to read, and I probably, as a teenager . . . wouldn’t take the time to read it.

Young person, JIA218

Information from six studies indicates that many families appreciated telephone calls or text messages from intervention deliverers or researchers to remind them to attend appointments, carry out the intervention at home or use the intervention when required. 114,177,186,190,191,201 Three additional studies suggested that CYP and their families may benefit from information that they can refer back to at the end of an appointment or intervention. 119,171,189 In addition to facilitating engagement with an intervention, these memory aids can also help CYP and their families to incorporate advice and skills practice into their daily lives. This may extend any potential benefits of a mental health and well-being intervention beyond the scheduled end of the intervention, a concept which is discussed below in Theme: keeping it going.

Theme: keeping it going

The content of this theme draws on 42 articles (see Report Supplementary Material 2, Table 6) and discusses the importance of ensuring that interventions are adequately resourced and have components that can be adopted into everyday life by CYP and their families. Some interventions may encourage the development of skills such as emotional management by CYP, which may increase the likelihood that any benefits on CYP’s mental health and well-being can be sustained beyond the end point of an intervention.

The importance of ongoing training or supervision with peers or colleagues with specialised mental health training, particularly for intervention deliverers with non-mental health backgrounds, was recognised by the authors of three studies. 162,178,197 It was observed that individuals delivering more specialised mental health interventions, such as art therapy, receive a higher degree of training. 172,189,197 Without such ongoing support, it would be difficult to sustain delivery of an intervention long term and maintain engagement with CYP by ensuring that the intervention meets their mental health and well-being needs.

The need for training prior to the onset of an intervention to ensure that intervention deliverers are able to meet the psychosocial needs of CYP taking part was implied in a further three studies. 166,172,200 This is illustrated by a quotation describing how intervention deliverers needed to be able to respond quickly to manage the disclosures of CYP during a group creative writing intervention:

And then there’s the risk that a kid starts to write something that maybe puts them in a precarious situation that we have to address in the class. We didn’t know what was going to come out, what kids were going to write, what they were going to say.

Intervention deliverer, HIV172

This interpretation is closely linked to the need to provide appropriately trained staff to deliver an intervention, as discussed in Theme: availability. The skills of individual intervention deliverers and the support provided to them may determine how well an intervention of longer duration can meet the needs of CYP as these needs change over time. Manualised protocols may also help facilitate the training of health/social care staff and consistent delivery of mental health interventions. 186,189,200

Financial support to buy appropriate equipment that CYP can use if required173,177,205 and to fund staff time189,191,198,206 was also recognised as essential to ensuring that an intervention is sustainable. One young person from a family art therapy intervention indicated their concern regarding issues relating to the funding of an intervention:

I worry that they’re not going to get funded because there might not be a huge change in someone, but when you’re chronically ill, you don’t need a huge change to make a difference . . . I think whoever grants you all the money needs to know that . . .

Young person, mixed164

The above quotation hints at the need for services to prove their efficacy in order to receive funding, and the potential impact on the mental health of CYP who are aware that funding for an intervention they find helpful may not be secure. 189 Resource and funding issues may have implications for the ability of intervention deliverers and services to provide an intervention that can be experienced as a ‘therapeutic foundation’ by those who access it.

Data from 16 studies suggest that, over time, the psychosocial impact of a mental health intervention may be seen outside the original intervention setting, as discussed in Theme: improved relationships with family and healthy friends and Theme: managing myself. 162,166,167,172–175,177,178,185,186,196,197,205,206,217 One factor that appears to facilitate this is the opportunity for CYP to practise the skills they have learnt, both in an intervention setting186,195 and in their everyday lives. 119,164,186,191,195 One young person describes how having materials to support practice of relaxation skills was helpful for encouraging practice outside a formal intervention setting:119,189,190

I found the CD [compact disc] to be most helpful. I practiced twice a week. Will still use it even after the study.

Young person, asthma119

The authors of 10 studies suggested that technology, including social networking sites, can provide the opportunity for CYP to continue supportive relationships developed during peer-based interventions. 163,165,166,169,174,175,179,183,185,217 A quotation from a young person indicates how these relationships can endure across time, despite relatively little face-to-face contact:

She was the one I called when I got cancer. I don’t know. Even though like we never see each other, I feel like she is one of my closest friends. Like, closer than the ones I have at school.

Young person, CHD165

The authors of six studies suggest that intervention deliverers can play a role in extending the impact of the intervention beyond its time-limited delivery period, by creating opportunities to meet again after the intervention has ended166,174 or by providing ‘top up’ intervention sessions based on the changing needs of the child. 171,189,195,205

Positive feelings regarding interventions can permeate CYP’s lives. Six studies suggest that merely the memory, facilitated by mementos or anticipation, of the intervention could have a positive impact on their feelings of well-being for both the CYP and their family:166,167,173,174,200,203

The meaning of camp in my life is everything. It’s what I want to do every single summer. I look forward to it every single summer . . . So it’s really, really fun.

Young person, mixed167

I’ll tell you that it’s something we won’t forget. It will be a memory that we will share. Together, the four of us. And you know, we talk about things that we did at Living Well.

Parent, mixed164

Four studies suggest that having a memory from the intervention to think about, or an object, such as a toy, to use, can be a source of comfort for CYP, particularly those undergoing distressing procedures. 174,191,200,205 These physical objects or memories can act as a reminder to CYP of their resilience by reminding them of how they have successfully overcome challenges in their physical health treatment and in an intervention setting. 174,200 These ‘transitional objects’ may act as a source of emotional comfort191,205 and be utilised in times of stress to improve emotional well-being.

Summary: getting in and staying in

‘Getting in and staying in’ acknowledges that getting CYP involved with an intervention that may benefit their mental health is a process, requiring the availability of an accessible intervention that is both tailored to the needs and characteristics of the CYP it intends to support and capable of adapting over time if their needs change. To ensure that an intervention is sustainable, adequate resources must be available to both implement the intervention and continue delivery over time. Interventions demonstrate sustainability when CYP are using aspects of the intervention or developing relationships outside sessions. It appears that both parents and service professionals have an important role in supporting CYP to access and engage with an intervention and encourage long-term behaviour/emotional change.