A paediatric therapeutic alliance occurs with distance intervention

Introduction

We have developed a novel system for delivering mental health services, called Family Help, that has been designed to overcome access barriers and improve retention for children and families receiving mental health treatment. Family Help provides manualized treatment (i.e. structured, protocol guided) at a distance by means of handbooks, videos and weekly telephone sessions with a non-professional coach. The Family Help programme consists of 12 weekly telephone sessions scheduled at the family's convenience, reducing the need to travel, take time off work or remove the child from school.

The coach is trained to problem-solve and provide programme skills education with the family as outlined in the Family Help protocol. Coaches follow strict risk management protocols involving immediate reporting to the Family Help health professional (a nurse or psychologist) of any suspected abuse, neglect or safety issues that may be disclosed by the parent or child during telephone sessions.

The Anxiety and Recurrent Pain Family Help programmes both include a cognitive-behavioural approach with emphasis on successful implementation of coping strategies (i.e. positive self-talk and relaxation techniques). The focus of the Anxiety programme is on gradual exposure using a hierarchical process. The Pain programme integrates learning of cognitive-behavioural stress management strategies with information about medication, diet and exercise. The child completes weekly homework assignments that include successful learning and implementation of a new skill (e.g. belly breathing, deep muscle relaxation or mini-relaxation, positive thinking/self-talk, role-playing and/or gradual exposure). The role of the parent is to be an ‘at home coach’ encouraging the child to learn and practice the new skill.

During weekly telephone sessions, the Family Help coach problem-solves with the parent and child, customizing treatment to address the individual child's problems. For example, if the child had a specific phobia of dogs, the coach would guide the design of a hierarchical worry list, ensure that the child learned the coping strategies to equip him/her for gradual exposure to dogs and evaluate success each week. However, it is not known whether a therapeutic alliance can occur between a child and coach in the absence of face-to-face contact. We have therefore conducted a study to determine if a therapeutic alliance occurs between the child and Family Help coach, and between the parent/primary caregiver and coach.

Methods

The study participants were adult caregiver–child pairs who completed the Family Help treatment programme for paediatric anxiety (6–12 years old) or recurrent headache/abdominal pain (9–16 years old). Children were referred to Family Help by family physicians and were eligible if the Family Help psychologist found evidence of a mild to moderate DSM-IV diagnosable disorder. A sample size of 55 parent–child pairs was needed to detect a difference of 5 points in therapeutic alliance scores with 90% power and alpha 0.05, including a 10% failure rate. All participants provided verbal consent to complete a telephone administered questionnaire. The study was approved by the appropriate ethics committee. Information was collected from January 2004 to February 2007.

Fifty-six parent–child dyads participated in the study. The reliability of one child's response data was questionable (i.e. it was apparent during the interview that he did not take it seriously and the responses were not consistent). Therefore, the data were excluded from analysis leaving 55 parent–child pairs (Table 1).

The majority of the participants were Caucasian, living in rural Nova Scotia. Most of the parent participants were female primary caregivers above 35 years of age (mean 39 years; SD 5). Half of the children were aged 9–11 years. The Family Help treatment had an average duration of five months and participants were interviewed about the therapeutic alliance at the end of treatment. All participants had a female telephone coach.

Results

Paediatric and parent therapeutic alliance scores are shown in Table 2. Four parents and seven children had missing data. Missing data were managed by replacement observation calculated from the mean item score within the specific subscale. The scores for parents and children were compared with paired t-tests (Table 3). Overall, the parent scores were significantly higher than the child scores. There was no significant correlation between the child and parent scores.

In a separate questionnaire, 93% of the primary caregivers strongly agreed they were satisfied with the service received. The correlation between parent total WAI scores and treatment services satisfaction was r = 0.14 (P =0.28). When asked if they would encourage other families to use Family Help to treat their child, 94% strongly agreed. The results of content analysis performed on participant responses when asked what they found most helpful about Family Help are summarized in Table 4.

Discussion

The results of the present study were compared to an earlier Family Help distance treatment study with adults and an adult face-to-face treatment study (Figure 1). Unfortunately, we were unable to find a study in which the 36-item WAI scale had been used to measure paediatric therapeutic alliance. However, a paediatric face-to-face therapy study reported a parental mean total WAI score of 230 (SD = 18), ² similar to the results of the present study.

OPEN IN VIEWER

Figure 1

WAI composite and subscale scores

A positive therapeutic alliance existed between the paediatric client–therapist and the parent–therapist dyads, in the absence of face-to-face contact. The paediatric WAI scores were similar to the WAI scores reported in adult face-to-face intervention ^1,3 and in other distance treatment studies. ^4,5 The results imply that Bordin's theory (mutual bond, goal and task agreement) can be generalized to paediatric distance therapeutic alliance. However, it is difficult to know whether this theory adequately defines the paediatric therapeutic alliance constructs, especially in distance therapy. Perhaps the child therapeutic alliance is more complex ^6,7 given the dual nature of alliance in parent-assisted intervention (e.g. both the parent and the coach work with the child). There is insufficient knowledge of child and parental views about paediatric therapeutic processes ⁸ and inadequate development of theories specific to the paediatric population. ^7–9

The present study was one of the first to examine differences between child–therapist and parent–therapist therapeutic alliance. The power of this study was sufficient to detect a small difference between the parent–child pair WAI scores. However, the significant mean difference of 8 points (out of a maximum score of 252) may not be clinically meaningful.

The lack of correlation between parent–child WAI scores and parent total WAI scores and treatment programme satisfaction was probably due to ceiling effects and to data truncation, and should be interpreted with caution. Similar effects have been found in paediatric face-to-face therapeutic alliance research, ^10–12 limiting predictability. However, studies have shown that parents and children both report that a positive child-therapist alliance is the most important part of treatment. ^13,14 Perhaps the effectiveness of manualized therapy ^10,11 and the specialized therapist attributes required for child therapy may heighten the treatment experience.

To achieve positive health outcomes, the child must fully participate in frequent treatment sessions ^6,10,15 to learn how to implement coping strategies (such as relaxation techniques used to facilitate the gradual exposure process for anxiety). The present study provides evidence that a young child with an internalizing disorder will engage in manualized treatment from a distance and form a strong bond with a non-professional coach whom they have never met in person. The content analysis findings provide validation of the parent and child therapeutic alliance scores with 73% of the sample describing positive coach attributes, including comments about the relationship. Moreover, the attrition rate in the Family Help study was low (about 1%) indicating that therapy delivered from a distance is possible and highly accepted by children and adults. ^4,16 The majority of the participants were very happy with the service that they received and many commented about positive health outcomes.

Programme design attributes reported in the satisfaction survey are similar to previous findings. ¹⁶ The convenience of home delivery eliminates the need to travel. Families seem to prefer the flexibility of treatment sessions offered after usual business hours. ^16,17 In the present study, the peak times for coach telephone sessions were between 18:00–20:00 hours, indicating user preference for after-hour access to treatment services. Increased service accessibility and convenience may enhance the therapeutic alliance and treatment compliance, decreasing drop-out rates.