How many adolescents are actually included in randomised trials primarily in adults with asthma? An analysis of recent trial reports

Article type
Authors
Annal O1, Clark K1, McNamara P2, Sinha I2
1University of Liverpool Medical School, United Kingdom
2Alder Hey Children's Hospital/University of Liverpool Medical School, United Kingdom
Abstract
Background: Adolescents are often included in randomised controlled trials (RCTs) of therapies for adults with asthma. People making treatment decisions about adolescents with asthma need to know the numbers of adolescents in the trial, and relevant safety and efficacy results for this subgroup. No studies have evaluated how fully these data are reported in published RCTs.
Objectives: To evaluate, in RCTs claiming to include both adults and adolescents with asthma:
1. whether authors report the proportion of adolescents included (and what this proportion is); and
2. whether results for adolescents are presented as a subgroup analysis.
Methods: One author searched Scopus and the Cochrane Database of Controlled Trials for RCTs published since 2012, claiming to include adults and adolescents with asthma. Three authors independently assessed each publication to assess whether the following were reported: proportion of all participants who were adolescents; the age of the youngest participant; whether adolescents were evaluated as a subgroup, and the results of this analysis.
Results: Thirty-three RCTs were included, each of which stated that both adolescents and adults were eligible for inclusion. In 27/33 (82%) this was claimed in the title or abstract. In 18/33 (55%) the age of the youngest participant was presented. In 12/33 (36%) the authors presented the proportion of participants in the trial who were adolescents. This ranged from 6% to 28%. In 9/33(27%) it was not possible to determine whether any adolescents were actually included in the trial. Only 2/33 (6%) presented results of a subgroup analysis in adolescents.
Conclusions: RCTs in asthma that claim to include both adults and adolescents are inadequately reported. Inclusion of adolescents within these trials is minimal, and this precludes any meaningful interpretation of results for this subgroup.