Effectiveness of an unguided modular online intervention for highly anxious parents in preventing anxiety in their children: a parallel group randomised controlled trial

Background: Children whose parents have anxiety problems are at increased risk of developing anxiety themselves. Parenting behaviors are a contributing factor to intergenerational transmission. Interventions which seek to limit anxiogenic parenting behaviors have shown potential in reducing anxiety in offspring but are not widely accessible. This prevention trial aimed to establish the effectiveness of an unguided modular online intervention for highly anxious parents in preventing anxiety in their children. Methods: A parallel group, block-randomised controlled trial of unblinded participants in a 1:1 ratio was conducted to compare efficacy of the online course compared to a no-intervention control. The intervention comprised 8 modules, of approximately 20 min each, and participants progressed through the course at their own pace. The study was conducted entirely online with a self-referred UK-based community sample of parents (child 2–11 years) with substantial anxiety. The primary outcome measure was change in parent-reported child anxiety, as measured by the Spence Children's Anxiety Scale–Parent Report (SCAS-P) or Spence Pre-School Anxiety Scale–Parent Report (Preschool SCAS). Secondary outcomes were child internalising, externalising, and attentional symptoms (Pediatric Symptom Checklist), and parent anxiety (SCARED-Adult). Analyses using complete case analysis following intention to treat principles investigated intervention effects at 6 months (primary analysis) and additionally at 9 to 25-months’ follow-up. Trial registration: ClinicalTrials.Gov NCT04755933, https://clinicaltrials.gov/ct2/show/NCT04755933. Findings: 1811 participants (intervention = 900; control = 911; 92.7% (1677/1810) female; 85.3% (1535/1800) White-British; 66.8% (1201/1799) university educated). Participant retention (based on primary outcome completion) at T2 (6-months post consent) was 67.6% overall (n = 1224) and substantially lower in the intervention arm 57.3% ((516/900) control = 77.8% (708/910)). Child anxiety was lower in the intervention group compared to control at 6-month follow-up (adjusted effect size estimate −0.15 (95% CI: −0.23 to −0.08, p < 0.001). There was very strong evidence that those in the intervention arm had lowered child anxiety (standardised SCAS score) compared to the control arm, with an effect size (Cohen's d) of −0.16 (95% CI: −0.23 to −0.08, p < 0.001). The difference in standardised Spence Child Anxiety Scale score between the arms was −0.15 standard deviations. On the original scales for SCAS-P (0–114) and Preschool SCAS (0–112), this corresponds to a reduction of −2.38 (95% CI: −3.59 to −1.16) and −2.68 (95% CI: −4.05 to −1.31), respectively. No reported harms. Interpretation: A clinically unsupported online intervention designed for parents with high levels of anxiety is effective in reducing anxiety and internalising symptoms in their children, and also anxiety in parents. Given the low resource intensity of this intervention, and the positive effects reported here, these findings suggest it has promise in limiting the intergenerational transmission of severe anxiety. Funding: This work was supported by Kavli Trust (grant 38/19).