Improving referrals for glue ear from primary care: are multiple interventions better than one alone?

To evaluate the effect of a risk factor checklist and training video for general practitioners in reducing inter-practice variation and improving the appropriateness of referrals (assessed by their positive predictive value or PPV) of patients with suspected otitis media with effusion (OME or 'glue ear') to secondary care.

Fifty general practices (177 practitioners) from the NHS Trent region and the West of Scotland were cluster-randomised either to a control group (n = 12) or to one of three intervention groups (training video (n = 16), checklist (n = 11), or both (n = 11)). Data on all paediatric ear, nose and throat (ENT) referrals and diagnostic results at ENT clinics were collected for a one-year period pre- and post-intervention. Referral rates for OME and for closely related conditions were calculated for children aged 0-15 years, based on each practice's list size. PPV was defined as the proportion of referrals resulting in bilateral hearing loss > or = 20 dB at the ENT outpatient department.

There was a significant improvement in the PPV, adjusted for patients' waiting time between general practitioner (GP) referral and being seen at the ENT department. The improvement in PPV pre- and post-intervention was 15% (95% confidence interval, CI: -12.1% to 41.7%) for the practices receiving both interventions, compared with a degradation of 20% pre- and post-intervention (95% CI: -32.9% to -6.4%) for practices receiving only one intervention and a degradation of 34% for those receiving no intervention.

Disseminating a risk factor checklist and training video on glue ear to GPs using a multi-channel approach can improve the quality of referrals to ENT.