Psychological Interventions for Parents of Children with Cancer: A Systematic Review and Meta-analysis

Context and Objectives

Childhood cancer represents one of the leading causes of death in children aged 1-14 years in developed countries, affecting approximately 140.6 per million person-years. Although survival rates have improved significantly (up to 90% for acute myeloid leukemia in some cases), the diagnosis and treatment of childhood cancer generates profound psychological impact on the entire family. Parents face multiple difficulties including loss, uncertainty, fear of relapse, anxiety, depressive symptoms, and the need to assimilate complex information about their child's health status.

Previous literature has documented that parents experience four main categories of stress reactions: uncertainty, anxiety, depressive symptoms, and posttraumatic stress. These responses affect family quality of life and may compromise the child's well-being. Psychological interventions have been widely recommended to promote short and long-term family adaptation.

The primary objective of this study was to conduct a systematic review and meta-analysis of randomized controlled trials evaluating the efficacy of psychological interventions specifically directed at parents of children with cancer, providing a rigorous and updated assessment of the effectiveness of these interventions.

Method

Participants

The meta-analysis included 14 studies meeting selection criteria, conducted between 1998 and 2017. The total sample size comprised 1,561 individuals: 809 subjects formed the 14 treatment groups (M = 57.79 participants, SD = 55.48, range = 10-217 participants), and 752 participants constituted 14 independent control groups (M = 53.71 participants, SD = 53.96, range = 10-213 participants). Attrition rates ranged between 0% and 32.5%.

Design

A systematic search was conducted in electronic databases (PsycINFO, Medline, PsycARTICLES, and PsycTESTS) during January 2019. Selection criteria were: (a) face-to-face intervention, (b) specifically designed for parents of children diagnosed with cancer, (c) with control group (waiting list or no-intervention control), and (d) random assignment to experimental conditions. Studies analyzing interventions in specific medical procedures, treatments, or palliative and end-of-life care were excluded.

The initial search yielded 925 references. After applying inclusion criteria, 21 articles underwent full-text review. Of these, 7 were excluded during full-text review: in 1 study, both conditions received the intervention but there was no control condition for comparison; 2 had questionable descriptive adequacy (studied only child adjustment); and 3 studies had no active control group. The remaining 14 studies were included in the systematic review and meta-analysis.

Intervention/Conditions

The examined studies included different participant categories (children, siblings, parents) and diverse intervention methods: face-to-face interventions, written disclosures, internet-based interventions, behavioral guidance (randomized controlled trials and quasi-experimental), and covered a wide range of topics (grief, acupuncture, radiotherapy, etc.). Most of the 14 studies (12 studies) focused on emotional symptoms and two on quality of life.

Instruments

For depression, the Beck Depression Inventory-II (BDI-II) and the "Depression/Depression Detection" subscale of the Profile of Mood States were used. For anxiety, the State-Trait Anxiety Inventory (STAI), Beck Anxiety Inventory (BAI), Hamilton Anxiety Rating Scale (HAM-A), and "Tension/Anxiety" subscale of POMS were employed. Posttraumatic stress was measured with the Impact of Event Scale-Revised (IES-R). The Perceived Stress Scale (PSS) and Maslach Burnout Inventory (MBI) were also used. Quality of life was measured with the 36-Item Short-Form Health Survey (SF-36) and the Care of My Child with Cancer Questionnaire.

Analysis

Study characteristics were coded: design type, statistical techniques, control group type, demographic characteristics, and outcome measures. Individual effect sizes were calculated for each of six potential outcome measures: depression, anxiety, posttraumatic stress, stress/burnout, general indicator of emotional symptoms, and quality of life. A random-effects model was used to calculate weighted effect sizes and 95% confidence intervals.

Heterogeneity was assessed with the Q statistic. The I² index was used as an indicator of the degree of heterogeneity. Values of 0%-25%, 50%, 75%, and greater than 75% in the I² statistic were considered indicative of no, low, moderate, and high heterogeneity, respectively.

Results

Study Selection: Of 925 identified references, 21 articles met preliminary inclusion criteria. After full-text review, 14 studies were included in the meta-analysis.

Sample Characteristics: The 14 included studies comprised a total of 1,561 individuals. The treatment group included 809 subjects (M = 57.79, SD = 55.48, range = 10-217), while the control group included 752 participants (M = 53.71, SD = 53.96, range = 10-213).

Primary Results by Outcome Measure:

Depression: The only statistically significant effect size was for depression at posttreatment (d = 1.19), indicating a large effect size. However, moderation analysis indicated this substantial effect size was due to two studies with low methodological quality. The relationship between methodological quality and efficacy in reducing depression was negative (ES = -0.50, p = .049).
Anxiety: Weighted effect sizes for anxiety were not statistically significant at posttreatment (d = 0.21) or follow-up (d = 0.31).
Posttraumatic Stress: Effects were not statistically significant (d = 0.05).
Stress/Burnout: The effect size was small and not statistically significant (d = 0.90).
General Indicator of Emotional Symptoms: The weighted effect size was small and not statistically significant (d = 0.33).
Quality of Life: The effect size for quality of life was small (d = 1.26) but not statistically significant.

Heterogeneity: Statistically significant heterogeneity was found for depression at posttreatment (Q = 4.88, p < .05) and follow-up (Q = 14.37, p < .01), as well as for quality of life (Q = 5.26, p < .05).

Moderation Analysis: Methodological quality was the only significant moderator variable, indicating a negative relationship between methodological quality and effect (higher quality was associated with smaller effects).

Discussion and Conclusions

Main Findings

The current study reviewed empirical evidence regarding the efficacy of face-to-face psychological interventions with parents of children with cancer. Fourteen studies met selection criteria and were included in the meta-analysis. Twelve studies focused on emotional symptoms and two on quality of life.

The findings indicate that most studies reported small to moderate effect sizes or found no significant effect of psychological intervention compared to the control group. This is consistent with previous literature indicating that effect sizes were small for comparisons of parental distress and parental adjustment. Parental distress and parental adjustment were the only outcomes showing an effect size significantly greater than zero at posttreatment (d = 1.19).

However, moderation analysis indicated this large effect size was due to the two studies with the lowest methodological quality. Indeed, the effect size at follow-up was considerably smaller (d = 0.31) because these two studies did not report follow-up assessment. Only one study included a long-term follow-up of one year.

Theoretical and Practical Implications

Parents frequently experience stress reactions after childhood cancer diagnosis, which is a normal response. According to Kazak, Schneider, Didonato, and Pai (2015), between one and two-thirds of parents are at low risk, approximately one-quarter presents medium risk, and up to 15% are at high risk and considered clinical population. Thus, it is crucial to identify which parents truly need psychological intervention due to severe, escalating, or persistent distress. This is consistent with Sheard and Maguire's recommendations about psychological interventions targeted at those identified as at risk of or suffering significant psychological distress. It is also essential to identify parents who can cope with symptoms, although they experience subclinical symptoms that should be monitored.

Given that emotional response to childhood cancer diagnosis is a normal response in parents, and most parents naturally recover during treatment, it seems more appropriate to direct psychological interventions to primary caregivers, which may include parents, grandparents, and legal guardians of the pediatric oncology patient.

Several limitations of this meta-analysis relate to the characteristics of the studies conducted and methodological aspects worth mentioning. First, most studies included a relatively small number of participants. Childhood cancer is considered a rare disease, and obtaining a large sample is not always an easy task. Second, the absence of control group and randomization were the reasons many studies obtained in the initial search were not included. However, this is a necessary criterion to attribute improvements to psychological intervention and not to the effect of time or other external variables.

Third, nine studies had more than 10% attrition rate, with nine showing dropout superior to 20%. Schulz and Grimes (2002) argue that loss to follow-up of 5% or less is usually of little clinical interest, whereas loss of 20% or greater can threaten trial validity. Fourth, nine of sixteen studies did not include intention-to-treat analysis, which could affect observed effect sizes. Fifth, only one study included long-term follow-up (1-year follow-up), while remaining studies conducted follow-up of maximum three months.

Future Directions

In conclusion, there is limited evidence for the efficacy of psychological interventions for parents of children with cancer according to randomized controlled trials. In this regard, it is relevant to highlight that single-group designs and non-randomized control trials tend to overestimate the intervention effect size. It is recommended that future studies: (a) analyze the effect of psychological interventions in more distressed parents, (b) develop assessments of family psychosocial risk to provide personalized attention, (c) conduct longer follow-ups, and (d) implement psychological interventions targeting primary caregivers.

Importance and Contribution

This meta-analysis represents the first systematic review focused specifically on parents and randomized controlled trials of parent-directed interventions. The greater homogeneity among meta-analyzed studies and rigorous assessment of methodological quality allow for more reliable evaluation of intervention effectiveness. Unlike previous meta-analyses, this approach provides greater homogeneity among analyzed studies, which increases internal validity of the meta-analysis and avoids potential bias introduced by within-group designs and non-randomized controlled trials.

The study contributes significantly to the field by: (1) providing an updated assessment of available empirical evidence in the area; (2) rigorously evaluating the efficacy of psychological interventions specifically in parents and adolescents with cancer in randomized controlled trials and face-to-face interventions; (3) identifying important limitations of published studies; and (4) recommending future research directions to establish efficacy of psychological interventions for parents of children with cancer. These findings have important clinical implications for developing more effective interventions and identifying populations that significantly benefit from services such as psychoeducation and family support.