Introduction
Problems with anxiety are extremely prevalent across the life course (Kessler et al., Reference Kessler, Angermeyer, Anthony, De Graaf, Demyttenaere, Gasquet, De Girolamo, Gluzman, Gureje and Haro2007) and have a particularly early age of onset. A recent meta-analysis highlighted that 38% of anxiety/fear disorders are first experienced by the age of 14 years, with a median age of onset of just 8 years of age for phobias and separation anxiety disorder (Solmi et al., Reference Solmi, Radua, Olivola, Croce, Soardo, Salazar de Pablo, Il Shin, Kirkbride, Jones, Kim, Kim, Carvalho, Seeman, Correll and Fusar-Poli2022). Anxiety disorders are also the most prevalent mental health problem in childhood, affecting 6.5% of children and adolescents based on a meta-analysis of international studies (Polanczyk et al., Reference Polanczyk, Salum, Sugaya, Caye and Rohde2015). There is now extensive evidence that cognitive behaviour therapy (CBT) is an effective treatment for child anxiety disorders (James et al., Reference James, Reardon, Soler, James and Creswell2020) but access to CBT remains a major challenge. In a study conducted across England in 2017, despite 65% of families having sought help, only 2% of children identified as having an anxiety disorder had received CBT (Reardon et al., 2019); recent reports suggest that this situation has not markedly improved in recent years (Weale, Reference Weale2022). Elsewhere a similar picture is seen; a recent Australian study found higher rates of access to CBT among children with elevated symptoms of anxiety (19.5%), but the majority of children whose families had sought help for anxiety (66.3%) had not received CBT (Gandhi et al., Reference Gandhi, O’Gradey-Lee, Jones and Hudson2022).
An important observation in the most recent Cochrane review of CBT for child anxiety disorders was that there was no evidence of differences in child outcomes on the basis of therapist contact time, with similar outcomes achieved when therapists provided less than 10 hours of clinical contact compared with higher amounts (James et al., Reference James, Reardon, Soler, James and Creswell2020). Notably, where these brief interventions were focused on pre-adolescents they typically took a parent-led approach, where the therapist works with the parent to support them to implement CBT strategies with their child. We have previously argued that this approach has the potential to increase early access to CBT for children with anxiety disorders in a number of ways including reducing the perceived stigma and disruption to usual activities for children by not requiring them to attend regular clinical appointments, enabling families to manage the child’s difficulties ‘as a family’ (Reardon et al., Reference Reardon, Harvey, Young, O’Brien and Creswell2018), and by reducing the overall therapy time required (as parents can make use of a guided self-help type approach to implement strategies in their child’s day-to-day life) (Creswell et al., Reference Creswell, Parkinson, Thirlwall and Willetts2019). A number of randomised controlled trials (RCTs) to evaluate CBT delivered via parents for child anxiety disorders have now been conducted and recent meta-analyses have concluded that parent-only interventions (Jewell et al., Reference Jewell, Wittkowski and Pratt2022) and parent-only CBT specifically (Yin et al., Reference Yin, Teng, Tong, Li, Fan, Zhou and Xie2021) are effective treatments for child anxiety problems. We have provided an up-to-date list of key RCTs of this approach in Table 1. As shown, studies vary somewhat in the treatment approach used, as well as in the duration of treatment, the settings in which they have been conducted, and the exact child age range included (although they are predominantly with pre-adolescents). There is also some variation in parent engagement, hence our prioritisation of intent to treat outcomes in Table 1 (using the conservative assumption that that drop-outs continued to meet criteria for anxiety disorder(s)). However, all parent-led CBT approaches share a focus on supporting and empowering parents to apply CBT strategies in their child’s day-to-day life with the most important work being done by parents with their children between therapy sessions.
Key randomised controlled trials of parent-led CBT for child anxiety disorders 1
1 Data presented are of children with diagnosed anxiety disorders only.
2 We prioritised diagnostic outcomes, followed by parent-report continuous measures of anxiety given the young child age range in most studies (Creswell et al., Reference Creswell, Nauta, Hudson, March, Reardon, Arendt, Bodden, Cobham, Donovan and Halldorsson2021); we prioritised data on ‘free of primary anxiety disorder’ as this is most commonly reported (James et al., Reference James, Reardon, Soler, James and Creswell2020) but report ‘free of all anxiety diagnoses’ where ‘free of primary disorder’ was not reported.
3 Categorical data based on intent to treat population unless unavailable and then stated.
4 Data not available to calculate effect sizes.
5 The paper did not specify the number of participants randomised to each condition, thus, we have assumed that there were 15 participants in each condition.
6 Child diagnostic status not formally assessed pre-treatment but children ‘appeared likely to have an anxiety disorder upon preliminary interview with a clinical psychologist’.
7 Children were required to have a DSM-IV diagnosis or to score 12 or higher on the Pediatric Anxiety Rating Scale.
8 +1 added in order to calculate odds ratio where events = 0.
9 Negative SMD favours parent-focused treatment.
DSM-IV, Diagnostic and Statistical Manual; MASC, Manifest Anxiety Scale for Children; SCARED, Screen for Anxiety related Emotional Disorders; SDQ, Strengths and Difficulties Questionnaire.
Given the variability in characteristics and methods of the studies in Table 1, unsurprisingly there is also variation in child outcomes across studies, but it is notable that 10 of the 11 studies that compared parent-led CBT with a waitlist control found clear evidence of effectiveness on the measures we prioritised in Table 1. Of note, while the one exception did not find differences on parent-reported anxiety symptoms, it did find that the parent-led CBT outperformed the waitlist on a range of other measures (including total number of child anxiety diagnoses and parent reported functional impairment) and the authors concluded that [with sufficient replication] ‘we may find that empowering parents as lay-therapists increases sustainability and transportability of CBT for child anxiety’ (Smith et al., Reference Smith, Flannery-Schroeder, Gorman and Cook2014). It is also important to highlight that in some of these studies the effects of parent-led CBT are extremely strong; for example, among a rural Australian population, 92% of 6- to 12-year-old children were free of their primary anxiety disorder following an intervention in which parents and children were provided with workbooks and parents received telephone support from a therapist, compared with none of the children in a waitlist condition (Lyneham and Rapee, Reference Lyneham and Rapee2006). Similarly, 95% of children were free of all anxiety disorders following therapist-supported bibliotherapy which consisted of a two-hour initial training for parents, conducted in groups, followed by fortnightly brief telephone support from a therapist, again compared with none of the children in the waitlist condition (Cobham, Reference Cobham2012). Clearly parent-led CBT can be a highly effective brief intervention for child anxiety disorders.
Studies that have compared parent-led CBT with CBT delivered directly to the child or to both the child and parent have also provided encouraging results. Many have failed to find significant differences (even where the parent-led CBT approach has involved considerably less therapist time, e.g. Waters et al., Reference Waters, Ford, Wharton and Cobham2009) and some have even found that brief parent-led approaches can be more effective than more intensive parent and child approaches (e.g. Cobham, Reference Cobham2012; Leong et al., Reference Leong, Cobham, De Groot and McDermott2009), although three (of eight) studies found that a parent-only approach achieved less good outcomes than a combined parent and child approach (Chavira et al., Reference Chavira, Drahota, Garland, Roesch, Garcia and Stein2014; Mendlowitz et al., Reference Mendlowitz, Manassis, Bradley, Scapillato, Miezitis and Shaw1999; Monga et al., Reference Monga, Rosenbloom, Tanha, Owens and Young2015). This variation in findings likely reflects the fact that it is of course not just who you work with that matters but also what you do with them (Lawrence et al., Reference Lawrence, Parkinson, Jasper, Creswell and Halligan2021). Parent-led CBT approaches are often briefer than child only or combined child and parent approaches, which also means that even where outcomes are not better with parent-led CBT, it may still be a more cost-effective approach. Indeed, in the one full health economic analysis of parent-led CBT compared with an alternative treatment, we found a 96% probability that parent-led CBT was cost-effective compared with solution focused brief therapy (including the child and parent) (Creswell et al., Reference Creswell, Violato, Fairbanks, White, Parkinson, Abitabile, Leidi and Cooper2017).
A further important consideration for increasing access to CBT is who can deliver the intervention in the context of a limited pool of trained and qualified practitioners (Stallard et al., Reference Stallard, Udwin, Goddard and Hibbert2007). While there has been little systematic evaluation of this, Thirlwall et al. (Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013) found similar outcomes from their 5 hour 20 minute intervention when the treatment was delivered by ‘novices’ (e.g. assistant psychologists, graduate students) as when it was delivered by therapists with relevant previous experience (e.g. clinical psychologists, psychiatrists). Notably, both groups received specific training (involving a detailed treatment manual and one day training) and regular case supervision (2 hours per week of group supervision provided by an experienced clinical psychologist, including feedback on recordings of treatment sessions). These findings have promising implications for the potential to draw on wider workforces to increase access to psychological interventions for child anxiety problems. In the Thirlwall et al. (Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013) study all therapists had some background in psychology or child mental health, but a recent preliminary investigation reported promising results when school pastoral support workers were trained and supported to deliver the approach (Clarke, Reference Clarke2021). Notably, to date, supervision has typically been delivered by clinical psychologists with extensive experience of CBT and parent-led CBT specifically. Going forwards it will be valuable to establish the training and support needs of wider groups who may be well placed to deliver this intervention to both optimise outcomes and access.
Encouragingly, in contrast to some studies which have predominantly involved CBT delivered directly to the child (e.g. Ginsburg et al., Reference Ginsburg, Becker-Haimes, Keeton, Kendall, Iyengar, Sakolsky, Albano, Peris, Compton and Piacentini2018) good outcomes typically appear to be maintained and may continue to improve beyond treatment. For example, Thirlwall et al. (Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013) reported that 61% of the children who had not recovered from their primary anxiety disorder immediately after the 5 hour 20 minute treatment had recovered by the 6-month follow-up assessment (increasing the overall recovery figure from 50 to 76%). While limited by the fact that only a subset of the original participants were followed up and the absence of a comparison group, it was also encouraging that 3 to 5 years later treatment gains were maintained in 60% of participants, further recovery occurred in 19% without any further input from mental health services, and only 12% relapsed (Brown et al., Reference Brown, Creswell, Barker, Butler, Cooper, Hobbs and Thirlwall2017).
In summary, solutions to increase early access to CBT for child anxiety disorders are urgently needed and brief parent-led CBT appears to be a valuable tool to help achieve this for pre-adolescent children. However, clinicians face inevitable challenges in delivering parent-led CBT and share understandable concerns about how to manage them. We have now delivered or supervised this approach with hundreds of families and the following sections share some of what we have learned along the way about how to overcome common challenges and concerns that clinicians raise, whilst adhering to the key principles of (i) empowering parents, and (ii) creating opportunities that allow children to learn useful new things about their fears and worries (Creswell et al., Reference Creswell, Parkinson, Thirlwall and Willetts2019).
Common concerns and challenges
In our experience of supervising and training others, clinicians have raised various concerns about and challenges in delivering parent-led CBT. These commonly include: (i) concerns that they will not have the opportunity to work individually with the child, (ii) concerns that they lack experience in working with parents/carers, (iii) concerns that parental factors, such as high anxiety, will prevent parents from being able to implement the approach or specific techniques such as exposure effectively, (iv) challenges with engaging parents in the approach, and (v) challenges when the main problems faced by the child occur outside of the home so may not be under the parents’ immediate control. Such concerns and challenges can understandably disempower clinicians, making them hesitant to offer the approach or more liable to going ‘off model’ during treatment (e.g. James et al., Reference James, Reardon, Soler, James and Creswell2020). Here we address these concerns, drawing on relevant literature, clinical case studies, and anecdotes from clinicians we have worked with. Where relevant, we have emphasised how challenges can be addressed by focusing on what we consider to be the two key elements of the treatment approach: empowering parents and providing children with opportunities to learn new information.
No opportunity to work individually with children
Clinician: ‘I came into this wanting to work with children, not parents’
Clinicians commonly express disappointment or concern about working through the parent; they may feel demotivated, fear they will lose the child’s voice in treatment or doubt parents’ capacity to implement strategies effectively, due to their own experience or bias. Understanding the approach’s theoretical foundation helps clinicians to hold in mind the potential advantages of a parent-led approach and so to confidently communicate the rationale to parents. Specifically, the aim of a parent-led approach is to provide alternatives to natural parental responses that may inadvertently contribute to the maintenance of child anxiety (James et al., Reference James, Reardon, Soler, James and Creswell2020), whilst maximising treatment efficiency by capitalising on parents’ capacity to ‘cut to the chase’ and to apply CBT strategies in their child’s daily life in the short and longer term. It is of course also important for clinicians to be aware of the evidence base behind the approach. As one clinician commented: ‘Knowing this will last outside of the therapy room has been ace… it feels to me that parents become an extension of the clinician, which means outcomes are better’. Importantly, while the child is not involved in the regular treatment sessions, it is expected that they would typically be present for the initial assessment and would complete pre- and post-treatment outcome measures (e.g. see Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019) so there are still opportunities for their voice to be heard. Furthermore, we work collaboratively to support parents to agree motivating goals, an exposure plan for children to test out their fears in manageable ways, and rewards with their child, and to practise treatment strategies at home with their child. Consequently, one clinician noted that ‘The sessions are still child-focused but working with parents gives an additional perspective that’s really helpful’.
Lack of experience in working with parents/carers
Clinician: ‘I was nervous, and slightly intimidated, to work with parents before … I thought that because I didn’t have children myself, they would question how I could understand …’
It can feel daunting to work with parents, especially for clinicians with limited experience, but as noted above, similar outcomes have been found when parent-led CBT was delivered by novices or by more experienced practitioners (Thirlwall et al., Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013). Qualitative interviews from the same study indicated that although some parents may have been initially surprised at being so involved in the treatment, they found the approach acceptable, reported empowerment and skill development, and identified wide-ranging benefits both for the child and the wider family (Allard et al., Reference Allard, Thirlwall, Cooper, Brown, O’Brien and Creswell2021). Clinicians have also described parent-led CBT as an ‘empowering’ and ‘practical’ approach (Evans et al., Reference Evans, Hill, O’Brien and Creswell2019) with high satisfaction reported by clinicians from a range of professional backgrounds when delivered to families individually (Creswell et al., Reference Creswell, Hentges, Parkinson, Sheffield, Willetts and Cooper2010) and in groups (Evans et al., Reference Evans, Hill, O’Brien and Creswell2019).
Clinician: ‘We’re not teaching them how to parent or telling them what to do, we’re just giving them ideas that we know have worked with others … we don’t have to have all the answers … the approach is socratic and collaborative and encourages parents to seek help from other sources themselves. And when you see parents feeling validated that they are doing the right thing already or successful trying out new ideas, that gives you a lot of confidence!’
For clinicians to develop skills to overcome concerns in a model-adherent way, regular and effective clinical supervision is crucial for building confidence and preventing drift (Thirlwall et al., Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013). In our current research trials, low intensity practitioners receive one hour of individual clinical case management and one hour of group clinical skills supervision per week with a trained high intensity CBT therapist/clinical psychologist in line with published supervision guidance (IAPT, 2011; UCL, 2015a; UCL, 2015b) and we strongly advocate for this as the minimum supervision requirement for safe, effective and enjoyable practice.
Concerns about delivering parent-led CBT in the context of high parental anxiety
Clinicians often express concern that parents who experience anxiety difficulties themselves may struggle to implement a parent-led CBT approach to address child anxiety problems, and this is something that parents can worry about too. This may be an understandable concern given that (i) anxiety disorders run in families to some extent (Thirlwall et al., Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013), (ii) previous research has shown significantly poorer outcomes for children following (child-focused) CBT for anxiety disorders when parents are highly anxious (Cobham et al., Reference Cobham, Dadds and Spence1998) or have anxiety disorders themselves (Hudson et al., Reference Hudson, Newall, Rapee, Lyneham, Schniering, Wuthrich, Schneider, Seeley-Wait, Edwards and Gar2014), and (iii) both are hypothesised to be at least partly accounted for by parental behaviours (e.g. modelling or transferring anxious information, parental over-protection, and parental over-involvement) which may limit children’s opportunities to experience new or challenging situations (Murray et al., Reference Murray, Creswell and Cooper2009). However, it is also possible that, for the same reasons, a parent-led approach may be particularly beneficial in these circumstances as parents are supported to develop their confidence in helping their children to test out their fears and are provided with alternatives to instinctive (and understandable) responses which may inadvertently maintain children’s problems with anxiety. This has not been directly assessed but encouraging findings came from Hiller et al. (Reference Hiller, Apetroaia, Clarke, Hughes, Orchard, Parkinson and Creswell2016) who compared standard parent-led CBT to a version with additional techniques to help highly anxious parents tolerate their children’s negative emotions. In contrast to hypotheses, both treatment arms were associated with good outcomes with similar proportions of children free of their primary anxiety disorder in both arms as found in trials which did not specifically target parents with high anxiety suggesting that ‘standard’ parent-led CBT may well be effective even when parents are highly anxious (Hiller et al., Reference Hiller, Apetroaia, Clarke, Hughes, Orchard, Parkinson and Creswell2016).
The case illustration in Fig. 1 describes a family where both the parent and child experienced difficulties with anxiety. Here we outline the ways we empowered this parent to support their child to test out her fears. Exposure is a central component of this treatment approach which provides children with opportunities to learn new information about their fears and their ability to cope in feared situations. The importance of exposure for successful CBT for child anxiety disorders has been emphasised in various recent studies (see Creswell et al., Reference Creswell, Waite and Hudson2020). Despite suggestions from the adult literature that exposure in a graded format may not be the optimal approach for rapid and sustained learning (Craske et al., Reference Craske, Treanor, Conway, Zbozinek and Vervliet2014), we have retained a step-by-step approach so that exposure is manageable for both the child and parent (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). Thus, empowering parents to engage their child in their step-by-step exposure plan to test predictions is crucial. Parents can understandably be reluctant to help their child test out fears through exposure due to concerns about how their child will react and how they will cope with their child’s emotions or behaviours. In our supervisory experience, clinicians can be similarly concerned about encouraging parents to approach exposure, particularly where parents are anxious or when anxiety problems are longstanding, and it is important to be alert to this in supervision to ensure that parents are empowered and children get the opportunity to test out fears in a supported way.
Case illustration.
First, the clinician and Maria discussed current theory around the maintenance factors relevant to childhood anxiety, including the role that ‘other people’ (e.g. parents, teachers, siblings, etc.) can potentially play in the maintenance of these difficulties (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). Specifically, the clinician sensitively explored whether the understandable responses of any key people in April’s life may inadvertently be maintaining April’s difficulties with anxiety (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019) by preventing her from having an opportunity to learn new information about her fears and her ability to cope in feared situations. Second, the clinician used open, curious questions and normalised common parental concerns about exposure to explore Maria’s fears about it. They collaboratively identified that Maria had a specific anxious expectation: ‘If April tries a step she will scream and cry and I don’t know how to cope with that’. This was crucial to identify early on so that Maria and the clinician had a shared understanding of what might get in the way of doing exposure and could plan what Maria would say and do if April did get distressed. The clinician spent time revisiting the rationale for exposure with Maria, explaining that she was being given permission to risk her child experiencing a temporary increase in anxiety in a safe way in order to learn new information about her fears. The extra time taken for these conversations helped Maria to feel more confident in trying a step and prevented Maria and the clinician inadvertently avoiding exposure together.
The clinician further empowered Maria to put her own fears to the test by encouraging her to have a go at supporting April to engage in exposure so that Maria could also learn new information about her own fears and her ability to cope if April experienced distress. To facilitate this, the clinician and Maria identified a first small step that felt manageable for both April and Maria (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). Peterman et al. (Reference Peterman, Read, Wei and Kendall2015) suggest that exposures that are fun (as well as anxiety provoking) can increase children’s engagement – thus, it was agreed that the family would play a game where each family member would say the words ‘sick’ and ‘vomit’ progressively louder and louder, for 5 minutes. The clinician explicitly reviewed what both April and Maria learned from this task to consolidate their learning (Craske et al., Reference Craske, Treanor, Conway, Zbozinek and Vervliet2014) and to empower Maria to continue engaging April in her step-by-step plan; for example, the clinician asked Maria questions about how April reacted and how she coped with this, and how this would influence how she might respond differently to April when they try the next step. Lastly, the clinician highlighted the various ways in which Maria had supported April to complete a step of their plan and praised this, to help empower Maria to continue supporting April to face anxiety-provoking situations (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019).
As a result of these actions, Maria was able to support April to complete her step-by-step plan, which consisted of watching videos of other people being sick, making their own ‘sick’ with cold soups and food from their kitchen, and being able to sit next to someone who felt unwell – testing out predictions as they went. Accordingly, April’s routine outcome measure scores moved from the clinical range at pre-treatment to the non-clinical range at post-treatment. At the one-month follow-up appointment, Maria shared that she had learned that drawing attention to April’s phobia in treatment had not made April’s anxiety worse and that April’s anxiety did pass and she could tolerate April getting upset more than she thought. In fact, Maria found that encouraging April to face her fears in a step-by-step way had helped April to overcome her phobia and meant that Maria felt confident to support April with any fears that arose in the future.
Challenges with engaging parents
Parents commonly feel disempowered at the start of treatment and may have some reluctance to engage in a parent-led approach. This is understandable given the difficulties parents often face accessing evidence-based treatment for their child (Crouch et al., Reference Crouch, Reardon, Farrington, Glover and Creswell2019; Reardon et al., Reference Reardon, Harvey and Creswell2020) and their common concerns about being negatively judged by professionals (Owens et al., Reference Owens, Richerson, Murphy, Jageleweski and Rossi2007; Pullmann et al., Reference Pullmann, VanHooser, Hoffman and Heflinger2010). Furthermore, parents often arrive with the expectation that a clinician will work directly with their child to help them to overcome their difficulties (Allard et al., Reference Allard, Thirlwall, Cooper, Brown, O’Brien and Creswell2021). Using the case illustration in Fig. 2, we will highlight the key ways through which we were able to empower this parent to engage whilst adhering to the treatment model.
Case illustration.
In the first treatment session, the clinician provided Lehka with the opportunity to share her concerns about the treatment approach. The clinician acknowledged her concerns and collaboratively explored the potential benefits of the approach, referring explicitly to the available research evidence. Second, the clinician sensitively outlined current understandings of the development and maintenance of childhood anxiety and explored its relevance to Saira’s experience, using this as an opportunity to normalise Lehka’s frustration as a common parental response to child anxiety (Fisher et al., Reference Fisher, Masia-Warner and Klein2004). This ensured Lehka did not feel blamed for her understandable responses to Saira’s difficulties and helped her to recognise the powerful position she was in to help Saira to overcome her difficulties with anxiety (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). Third, equipping Lehka with skills for how to talk to Saira about anxiety helped to further empower her. The second treatment session provided Lehka with an opportunity to role-play key skills with the clinician, including asking open curious questions, being empathetic, and checking understanding, to help Lehka and the clinician to identify what Saira needed to learn to overcome her difficulties with anxiety (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). Learning these skills (and having success with implementing them at home) enabled Lehka to gain better insight into Saira’s difficulties with anxiety and to become more empathetic (Allard et al., Reference Allard, Thirlwall, Cooper, Brown, O’Brien and Creswell2021).
The development and implementation of a step-by-step plan for Saira to learn new information about her fears and her ability to cope in feared situations (in a way that was manageable way for both Saira and Lehka) served as an opportunity to empower Lehka further. The step-by-step plan is devised collaboratively, and the clinician’s role is to encourage the parent to generate their own ideas for possible steps to test anxious predictions and to praise their suggestions (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019). This ensured Lehka felt able to implement the step-by-step plan and would be confident to devise a new plan in the future, if needed. At subsequent treatment sessions, the clinician praised Lehka for successes with the plan, explicitly recognising the role she had played in bringing about positive change.
Where Lehka experienced difficulties implementing the plan, this was collaboratively problem-solved to empower her to consider ways she could resolve these challenges and continue to support Saira to overcome problems with anxiety. By focusing on empowering Lehka throughout this treatment, she was able to support Saira to ask questions in class and to order her own food at a café. These improvements were reflected in Saira’s routine outcome measures, which shifted from the clinical range at pre-treatment to the non-clinical range at post-treatment. At the one-month follow-up appointment, Lehka shared that she now had a better understanding and empathy for Saira’s difficulties, which meant that Saira now felt more able to discuss her anxieties with Lehka, and that they could both work together to help Saira to face her fears in a step-by-step way. Lehka also noted wider improvements in her relationship with Saira, identifying that they now spent time with each other and enjoyed each other’s company.
Challenges when problems occur outside the home
Child anxiety problems are frequently located, or have the biggest impact, outside the home environment, such as in school. This raises a potential challenge for a parent-led approach as the parent is not present when the child is struggling and so is limited in how much they can address stressors in the environment and directly help their child to implement treatment strategies. Addressing problems within schools can also feel particularly daunting for parents who feel that communication and collaboration with school staff and/or other professionals have broken down. Similar challenges can arise when families have other systems supporting them such as social care or other healthcare professionals, or where the need for additional help becomes evident during treatment.
The key to good outcomes for the child remains the same as we have emphasised elsewhere: for the system, including the parent, to feel empowered to communicate and build (or rebuild) relationships to support the child and family with treatment strategies (or additional help-seeking) (Carr, Reference Carr2016; Creswell et al., Reference Creswell, Parkinson, Thirlwall and Willetts2019). Indeed, preparing the family and wider system to manage difficulties autonomously is a component of effective systemic intervention (Carr, Reference Carr2016). Similarly, in CBT for child anxiety disorders, actively focusing on transferring control to the parent has been found to be associated with significantly better long-term outcomes, with 82% of children free of their anxiety diagnosis one year after treatment compared with 53–63% where there was limited parental involvement and without explicit focus on control transfer (Manassis et al., Reference Manassis, Lee, Bennett, Zhao, Mendlowitz, Duda, Saini, Wilansky, Baer and Barrett2014).
Empowering Davis to engage Jonah and his teacher in a step-by-step exposure plan that could be implemented to test out fears within school was crucial to helping Jonah reintegrate into class (see Figure 3 case example). Firstly, the clinician modelled a curious stance (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019), exploring Davis’ worry about approaching school by asking open questions about what might happen if he asked to speak to school. Davis shared his previous experience of telling a senior member of staff that he experienced anxiety himself and felt blamed; he worried ‘they will not listen and will think it’s my fault’. The clinician validated Davis’ experience and understandable anxious expectation, then stayed curious by wondering aloud whether there were other possible outcomes and people who could help (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019); Davis identified Jonah’s teacher as a potential ally.
It could be tempting at this point to take control of communication with school or professionals, especially when parents are feeling disempowered. However, parents are generally in a better position (than children and, often, clinicians) to liaise with school or other agencies to encourage the implementation of useful strategies. Parents typically want control over decision-making and treatment planning, and fostering parental capacity and strengths is associated with stronger self-efficacy and greater perceived helpfulness of support (Dunst et al., Reference Dunst, Trivette and Hamby2007; Dunst and Trivette, Reference Dunst and Trivette2009; Mak et al., Reference Mak, Hiebert-Murphy, Walker and Altman2014). Therefore, the question for the clinician is, ‘How do I empower this parent in these circumstances to communicate with X?’, and for the supervisor, ‘How do I support this clinician to empower this parent to communicate with X?’.
The clinician encouraged Davis to advocate for Jonah, helping him to recognise his capacity based on when he had done this previously (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019), and reminding him that it is OK to continue to ask for help for his child and that school will likely want to provide support if they feel they can. A step-by-step exposure plan that could be implemented at school and reinforced at home was identified as a tool that could be used to structure a conversation with school staff (Creswell et al., Reference Creswell, Parkinson, Thirlwall and Willetts2019); it provided a clear concrete plan with a specific goal that benefited everyone and could help to reframe school’s possible perception of Davis to one of him taking agency. The clinician adopted a problem-solving stance throughout (Halldorsson et al., Reference Halldorsson, Elliott, Chessell, Willetts and Creswell2019), supporting Davis to plan how he wanted to approach staff and how the clinician could support him. It was collaboratively agreed that Davis would ask for a meeting with Jonah’s teacher and the SENDCo at pick-up time, and Davis would share the suggested plan. During the meeting, Davis asked his pre-prepared questions to open a collaborative conversation: what steps did the teacher think would work in their classroom, and how could they quickly reward Jonah and tell Davis he had completed a step? The teacher helped Davis edit some steps to make them fit easily into usual class activities to reduce staff burden, and the teacher spoke to Jonah about how they could tell his father, using Jonah’s idea of a sticker sheet.
Jonah achieved his step-by-step plan and by the follow-up session he was staying in class all day every day. His routine outcome measures moved from the clinical range at pre-treatment to the non-clinical range. Crucially, Davis and school said they now had a good working relationship and the tools to support Jonah going forward, and Davis felt confident asking for help from school and others in future.
Discussion
There is a growing evidence base for parent-led CBT as a brief treatment for child anxiety problems and it presents a promising option for increasing access to psychological therapies for children with these common problems. Despite this evidence, both clinicians and parents often voice concerns about how appropriate or effective it will be. We hope that we have illustrated how some of these concerns, as well as common challenges, can be addressed while keeping within the treatment model and by particularly reflecting on two core principles of empowering parents and providing opportunities for new learning for children. While we of course do not expect all children to recover following this brief treatment, outcomes are generally good and there is a lack of consistent evidence that other approaches are better. Furthermore, consistent predictors of outcomes from brief parent-led CBT for anxiety problems have so far failed to be identified (e.g. in terms of age, gender, baseline anxiety severity, anxiety disorder subtype; Thirlwall et al., Reference Thirlwall, Cooper and Creswell2017) which means we have no empirical guide to tell us not to offer it in particular circumstances. One exception comes from an analysis of pooled data from international CBT trials, which found that children with a primary diagnosis of specific phobia benefited more from individual child-focused CBT, than group or parent-led CBT, but outcomes were good across all treatment conditions leading the authors to conclude that ‘future research … is needed to determine whether the additional clinical benefits justify the additional resources required’ (McKinnon et al., Reference McKinnon, Keers, Coleman, Lester, Roberts, Arendt, Bögels, Cooper, Creswell and Hartman2018), and highlighting the potential of parent-led CBT within a stepped care model given its fit with a low intensity format. Further research is also clearly required to help us better understand who does and does not benefit from this approach, and who might benefit with appropriate adaptations (for example, autistic children with anxiety problems, children with chronic physical health conditions and anxiety problems, children with obsessive compulsive disorder (OCD), children whose parents face particular challenges). Various studies are underway to explore these possibilities and we look forward to seeing what they find out.
As shown in Table 1, parent-led CBT for child anxiety problems can be delivered efficiently and this has been particularly the case when delivered remotely (over the telephone) or in a hybrid way with accompanying workbooks (e.g. Lyneham and Rapee, Reference Lyneham and Rapee2006; Thirlwall et al., Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013). As shown in Table 1, parent-led CBT has commonly been delivered in groups. While this may bring advantages from peer support for some parents, this format may also introduce barriers for others (Allard et al., Reference Allard, Thirlwall, Cooper, Brown, O’Brien and Creswell2021). From our interactions with clinical teams we are aware that groups are often used as a way to deliver the intervention efficiently, but notably cost-effectiveness evaluations comparing groups with the individual approach have not yet been conducted. Recent developments in online psychological interventions also highlight the potential to increase accessibility and efficiency (e.g. Schröder et al., Reference Schröder, Berger, Westermann, Klein and Moritz2022) yet only one RCT so far has evaluated online parent-led CBT (Donovan and March, Reference Donovan and March2014). In this study with parents of 3- to 6-year-olds, although the online treatment was superior to waitlist the outcomes were relatively modest (see Table 1), but the authors concluded that ‘given that the program … was not tailored specifically for parents of children in this age group and was somewhat makeshift in nature, the positive results obtained are particularly promising’ (Donovan and March, Reference Donovan and March2014). Further promising evidence that an online approach may be a feasible, efficient and acceptable approach that is associated with good outcomes comes from a recent clinical case series in which we examined routinely captured outcome measures from a new online parent-led CBT intervention that was designed in collaboration with parents, children and clinicians (Hill et al., Reference Hill, Chessell, Percy and Creswell2021).
An apparent benefit of parent-led CBT for increasing access to psychological interventions for child anxiety problems is that good outcomes can be achieved without relying on highly specialised clinicians (e.g. Thirlwall et al., Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell2013). While highlighting the potential of the approach in IAPT-type services for children, it also raises the interesting question of how broad a range of individuals might be able to deliver it effectively. For example, there is evidence that parenting interventions for disruptive behaviour problems in children can be successfully delivered by parent peer facilitators and that this approach may address particular barriers experienced by some underserved populations (Day et al., Reference Day, Michelson, Thomson, Penney and Draper2012).
A major limitation of the available literature on parent-led CBT is that the majority of trials conducted to date have been in Western countries (Australia, UK, USA) which leaves questions about the cross-cultural acceptability and efficacy of the parent-led CBT approach for child anxiety. Reasons to be encouraged about the potential transportability of the approach include a recent study that found excellent outcomes from therapist-guided online psychotherapy (iCT for social anxiety disorder) in culturally contrasting countries (UK and Hong Kong; Thew et al., Reference Thew, Kwok, Lissillour Chan, Powell, Wild, Leung and Clark2022) and from studies that have shown good outcomes from CBT for child anxiety disorders in non-Western countries (Ishikawa et al., Reference Ishikawa, Kikuta, Sakai, Mitamura, Motomura and Hudson2019) – both with minimal adaptations. Furthermore, Chavira et al. (Reference Chavira, Bustos, Garcia, Reinosa Segovia, Baig, Ng and Camacho2018) specifically examined the feasibility of telephone-assisted parent-led CBT for rural Latino youth with anxiety problems and concluded that the approach was feasible, acceptable, tolerable and safe in an underserved rural community, although some useful observations were made with implications for increasing accessibility and engagement. The overall conclusion that the approach may be a promising way to increase access to psychological therapies across cultures, albeit potentially with some minor adaptations, is consistent with anecdotal reports from international colleagues, as shown in Fig. 4.
Case illustration.
International experiences of parent-led CBT.
Increasing access to effective psychological interventions for common mental health problems in childhood remains a major challenge. Creative solutions are needed that overcome the many barriers that families face in accessing treatment while maintaining good treatment outcomes when treatments are delivered at scale. From the extant evidence and from our own experience of delivering and supporting others to deliver parent-led CBT for child anxiety problems we conclude that it is a valuable component of the toolkit that is required to ensure children and families get the support they need when they first need it. While there will be inevitable concerns and challenges in applying the approach, many of these can be successfully addressed with reference to the empirical literature and by embracing the key principle of empowering parents and working with them to create opportunities for children to learn what they need to learn to overcome their problems with anxiety.
Data availability statement
Not applicable.
Acknowledgements
We thank all the families that we have worked with and learned so much from. We also thank our clinical team within the TOPIC group who have contributed valuable feedback that has informed this paper and international colleagues who have shared their experiences. We thank Katie Fletcher for administrative support.
Author contributions
Cathy Creswell: Conceptualization (equal), Investigation (equal), Methodology (equal), Writing – original draft (equal), Writing – review & editing (equal); Chloe Chessell: Conceptualization (equal), Investigation (equal), Methodology (equal), Writing – original draft (equal), Writing – review & editing (equal); Gemma Halliday: Conceptualization (equal), Investigation (equal), Methodology (equal), Project administration (equal), Writing – original draft (equal), Writing – review & editing (equal).
Financial support
Cathy Creswell receives funding from the National Institute of Health and Care Research (NIHR) Applied Research Collaboration Oxford and Thames Valley at Oxford Health NHS Foundation Trust. Gemma Halliday is funded by the National Institute of Health and Care Research (NIHR) ICATS: i2i study; Chloe Chessell is funded by Economic and Social Research Council [ES/P00072X/1] and receives funding from the National Institute for Health and Care Research (NIHR) Applied Research Collaboration Oxford and Thames Valley at Oxford Health NHS Foundation trust. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.
Conflicts of interest
Cathy Creswell is the author of the books Helping Your Child with Fears and Worries and Parent-led CBT for Child Anxiety: Helping Parents Help Their Kids, and receives royalties from sales.
Ethical standards
The authors have abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the BABCP and BPS. Ethical approval was not needed.
Open access
Comments
No Comments have been published for this article.