Evaluation of an aid to diagnosis for congenital dysplasia of the hip in general practice: controlled trial randomised by practice

Background: In the UK, a compulsory 6-week hip check is performed in primary care for the detection of developmental dysplasia of the hip (DDH). However, missed diagnoses and infants incorrectly labelled with DDH remain an important problem, potentially leading to adverse consequences for infants, their families and the NHS. National policy states that infants should be referred to hospital if the 6-week check "suggests DDH", though there is no available tool to aid examination or offer guidelines for referral. We developed standardised diagnostic criteria for DDH, based on international Delphi consensus, and a 9-item Checklist that has the potential to enable non-experts to diagnose DDH in a manner approaching that of experts. Aims: 1. To elicit the influencing factors leading to inadequate diagnoses of DDH in primary care according to the Behaviour Change Wheel Framework. 2. To refine existing consensus-based standardised diagnostic criteria for DDH into a diagnostic aid tailored specifically for the 6-week check. 3. To determine if complementing the 6-week check with this aid is cost effective and reduces the number of mis-diagnoses. 4. To assess context, mechanism and implementation of the aid, and understand how the aid becomes embedded in practice. Setting: Primary care; GP practices and 15 Clinical Research Networks in England, with follow-up in secondary care. Participants: Primary care clinicians conducting 6-week infant hip checks and parents/carers of infants. Design: Mixed-method design for evaluation of complex interventions, conducted in three sequential phases, which will include qualitative and behaviour change studies; decision analysis; feasibility study and controlled trial randomised by practice with 2-year follow-up including clinical, health economic and qualitative process evaluation. Methods: First (10 months) In order to develop the most effective format, delivery mode and implementation strategy for an optimal diagnostic aid, we will undertake a rapid qualitative scoping study to determine the range of current organisational practices and explore the views of health professionals/parents about the 6-week hip check by interviewing primary care clinicians and carers from a purposive sample of practices in England, and shadowing them through the process of the 6-week check. We will use decision analysis to model the use of this aid in general practice. These studies will elicit potential barriers to appropriate diagnosis and referrals (primary behaviour for change), which we will evaluate within a Behaviour Change Framework (health professional s capability, opportunity and motivation). In parallel, we will draw on normalization process theory to elicit factors related to the embedding of the intervention in practice. This work will inform all aspects of the intervention and trial design. Second (8 months) We will test the feasibility of the intervention in a study of 6 established GP practices in North London. Findings of this pilot will refine the intervention where needed, and finalise all endpoints of the trial. Third (48 months) Over 16 months, we will recruit 152 GP practices in England to be randomly allocated to intervention or standard care when performing the 6-week check. We aim to detect a reduction, on average, from 3 to 2 referrals per practice as a consequence of primary care professionals using the aid. We hypothesise the intervention will lead to more accurate diagnoses and decisions at the 6-week check, i.e. fewer unnecessary referrals (measured in behavioural terms – i.e. number of appropriate referrals ) and fewer false negative diagnoses (measured as a non-behaviour – i.e. lack of referral for those needing it). While it is unfeasible to blind clinicians to interventions, an independent researcher will collect all outcomes (blinded to treatment) in secondary care for those referred at the 6-week check. In order to capture all false negati