Healthcare resource use per patient per 10 weeks | Video group (n=35) median (IQR) | Telephone group (n=24) median (IQR) | Control group (n=24) median (IQR) | p Value |

GP | 2.0 (1.0–3.2)^{A} | 3.9 (2.6–7.5)^{B} | 2.1 (0–3.3)^{A} | 0.002^{*} |

Admission to hospital | 0 (0–0)^{B} | 0 (0–0.96)^{A} | 0.5 (0–1.1)^{A} | 0.006^{*} |

Emergency department | 0 (0–0)^{B} | 0.22 (0–2.0)^{A B} | 0 (0–1.1)^{A} | 0.034^{*} |

Outpatient clinic | 1.0 (0.9–0.02) | 1.2 (0.91–2.0) | 2.0 (0.01–3.3) | 0.38^{*} |

Ward attender at paediatric cardiology department | 1.0 (0.91–0.15) | 1.2 (0.9–2.0) | 1.1 (1.0–2.4) | 0.68^{*} |

Cardiac liaison nurse contact | 1.8 (0.89–2.1)^{A} | 2.2 (1.1–3.1)^{A B} | 4.2 (2.2–6.2)^{B} | 0.001^{*} |

Total NHS episodes | Mean=7.7 ^{C} | Mean=13.3^{D} | Mean=12.9^{D} | <0.001^{†} |

SD=3.2 | SD=6.6 | SD=7.2 |

↵* Kruskall–Wallis.

↵† Analysis of variance: superscripts A and B denote groups which are not significantly different at the 5% level using Mann–Whitney U test and Bonferoni correction; superscripts C and D denote groups which are not significantly different at the 5% level using a post-hoc Duncan test.