Can primary care research be conducted more efficiently using routinely reported practice-level data: a cluster randomised controlled trial conducted in England?

Conducting randomised controlled trials (RCTs) in primary care is challenging; recruiting patients during time-limited or remote consultations can increase selection bias and physical access to patients' notes is costly and time-consuming. We investigated barriers and facilitators to running a more efficient design. An RCT aiming to reduce antibiotic prescribing among children presenting with acute cough and a respiratory tract infection (RTI) with a clinician-focused intervention, embedded at the practice level. By using aggregate level, routinely collected data for the coprimary outcomes, we removed the need to recruit individual participants. Primary care. Baseline data from general practitioner practices and interviews with individuals from Clinical Research Networks (CRNs) in England who helped recruit practices and Clinical Commission Groups (CCGs) who collected outcome data. The intervention included: (1) explicit elicitation of parental concerns, (2) a prognostic algorithm to identify children at low risk of hospitalisation and (3) provision of a printout for carers including safety-netting advice. For 0-9 years old-(1) Dispensing data for amoxicillin and macrolide antibiotics and (2) hospital admission rate for RTI. We recruited 294 of the intended 310 practices (95%) representing 336 496 registered 0-9 years old (5% of all 0-9 years old children). Included practices were slightly larger, had slightly lower baseline prescribing rates and were located in more deprived areas reflecting the national distribution. Engagement with CCGs and their understanding of their role in this research was variable. Engagement with CRNs and installation of the intervention was straight-forward although the impact of updates to practice IT systems and lack of familiarity required extended support in some practices. Data on the coprimary outcomes were almost 100%. The infrastructure for trials at the practice level using routinely collected data for primary outcomes is viable in England and should be promoted for primary care research where appropriate. ISRCTN11405239.

Blair PS ，Ingram J ，Clement C ，Young G ，Seume P ，Taylor J ，Cabral C ，Lucas PJ ，Beech E ，Horwood J ，Dixon P ，Gulliford MC ，Francis N ，Creavin ST ，Lane A ，Bevan S ，Hay AD ... -  《BMJ Open》

A multifaceted intervention to reduce antibiotic prescribing among CHIldren with acute COugh and respiratory tract infection: the CHICO cluster RCT.

Blair PS ，Young GJ ，Clement C ，Dixon P ，Seume P ，Ingram J ，Taylor J ，Horwood J ，Lucas PJ ，Cabral C ，Francis NA ，Beech E ，Gulliford M ，Creavin S ，Lane JA ，Bevan S ，Hay AD ... -  《-》

Protocol for an 'efficient design' cluster randomised controlled trial to evaluate a complex intervention to improve antibiotic prescribing for CHIldren presenting to primary care with acute COugh and respiratory tract infection: the CHICO study.

Respiratory tract infections (RTIs) in children are common and present major resource implications for primary care. Unnecessary use of antibiotics is associated with the development and proliferation of antimicrobial resistance. In 2016, the National Institute for Health Research (NIHR)-funded 'TARGET' programme developed a prognostic algorithm to identify children with acute cough and RTI at very low risk of 30-day hospitalisation and unlikely to need antibiotics. The intervention includes: (1) explicit elicitation of parental concerns, (2) the results of the prognostic algorithm accompanied by prescribing guidance and (3) provision of a printout for carers including safety netting advice. The CHIldren's COugh feasibility study suggested differential recruitment of healthier patients in control practices. This phase III 'efficiently designed' trial uses routinely collected data at the practice level, thus avoiding individual patient consent. The aim is to assess whether embedding a multifaceted intervention into general practitioner (GP) practice Information Technology (IT) systems will result in reductions of antibiotic prescribing without impacting on hospital attendance for RTI. The coprimary outcomes are (1) practice rate of dispensed amoxicillin and macrolide antibiotics, (2) hospital admission rate for RTI using routinely collected data by Clinical Commissioning Groups (CCGs). Data will be collected for children aged 0-9 years registered at 310 practices (155 intervention, 155 usual care) over a 12-month period. Recruitment and randomisation of practices (using the Egton Medical Information Systems web data management system) is conducted via each CCG stratified for children registered and baseline dispensing rates of each practice. Secondary outcomes will explore intervention effect modifiers. Qualitative interviews will explore intervention usage. The economic evaluation will be limited to a between-arm comparison in a cost-consequence analysis. Research ethics approval was given by London-Camden and Kings Cross Research Ethics Committee (ref:18/LO/0345). This manuscript refers to protocol V.4.0. Results will be disseminated through peer-reviewed journals and international conferences. ISRCTN11405239.

Seume P ，Bevan S ，Young G ，Ingram J ，Clement C ，Cabral C ，Lucas PJ ，Beech E ，Taylor J ，Horwood J ，Dixon P ，Gulliford MC ，Francis N ，Creavin ST ，Lane A ，Hay AD ，Blair PS ... -  《BMJ Open》

Multi-faceted intervention to improve management of antibiotics for children presenting to primary care with acute cough and respiratory tract infection (CHICO): efficient cluster randomised controlled trial.

To assess whether an easy-to-use multifaceted intervention for children presenting to primary care with respiratory tract infections would reduce antibiotic dispensing, without increasing hospital admissions for respiratory tract infection. Two arm randomised controlled trial clustered by general practice, using routine outcome data, with qualitative and economic evaluations. English primary care practices using the EMIS electronic medical record system. Children aged 0-9 years presenting with respiratory tract infection at 294 general practices, before and during the covid-19 pandemic. Elicitation of parental concerns during consultation; a clinician focused prognostic algorithm to identify children at very low, normal, or elevated 30 day risk of hospital admission accompanied by antibiotic prescribing guidance; and a leaflet for carers including safety netting advice. Rate of dispensed amoxicillin and macrolide antibiotics (superiority comparison) and hospital admissions for respiratory tract infection (non-inferiority comparison) for children aged 0-9 years over 12 months (same age practice list size as denominator). Of 310 practices needed, 294 (95%) were randomised (144 intervention and 150 controls) representing 5% of all registered 0-9 year olds in England. Of these, 12 (4%) subsequently withdrew (six owing to the pandemic). Median intervention use per practice was 70 (by a median of 9 clinicians). No evidence was found that antibiotic dispensing differed between intervention practices (155 (95% confidence interval 138 to 174) items/year/1000 children) and control practices (157 (140 to 176) items/year/1000 children) (rate ratio 1.011, 95% confidence interval 0.992 to 1.029; P=0.25). Pre-specified subgroup analyses suggested reduced dispensing in intervention practices with fewer prescribing nurses, in single site (compared with multisite) practices, and in practices located in areas of lower socioeconomic deprivation, which may warrant future investigation. Pre-specified sensitivity analysis suggested reduced dispensing among older children in the intervention arm (P=0.03). A post hoc sensitivity analysis suggested less dispensing in intervention practices before the pandemic (rate ratio 0.967, 0.946 to 0.989; P=0.003). The rate of hospital admission for respiratory tract infections in the intervention practices (13 (95% confidence interval 10 to 18) admissions/1000 children) was non-inferior compared with control practices (15 (12 to 20) admissions/1000 children) (rate ratio 0.952, 0.905 to 1.003). This multifaceted antibiotic stewardship intervention for children with respiratory tract infections did not reduce overall antibiotic dispensing or increase respiratory tract infection related hospital admissions. Evidence suggested that in some subgroups and situations (for example, under non-pandemic conditions) the intervention slightly reduced prescribing rates but not in a clinically relevant way. ISRCTN11405239ISRCTN registry ISRCTN11405239.

Blair PS ，Young G ，Clement C ，Dixon P ，Seume P ，Ingram J ，Taylor J ，Cabral C ，Lucas PJ ，Beech E ，Horwood J ，Gulliford M ，Francis NA ，Creavin S ，Lane JA ，Bevan S ，Hay AD ... -  《BMJ-British Medical Journal》

Feasibility cluster randomised controlled trial of a within-consultation intervention to reduce antibiotic prescribing for children presenting to primary care with acute respiratory tract infection and cough.

To investigate recruitment and retention, data collection methods and the acceptability of a 'within-consultation' complex intervention designed to reduce antibiotic prescribing. Primary care feasibility cluster randomised controlled trial. 32 general practices in South West England recruiting children from October 2014 to April 2015. Children (aged 3 months to <12 years) with acute cough and respiratory tract infection (RTI). A web-based clinician-focussed clinical rule to predict risk of future hospitalisation and a printed leaflet with individualised child health information for carers, safety-netting advice and a treatment decision record. Usual practice, with clinicians recording data on symptoms, signs and treatment decisions. Of 542 children invited, 501 (92.4%) consented to participate, a month ahead of schedule. Antibiotic prescribing data were collected for all children, follow-up data for 495 (98.8%) and the National Health Service resource use data for 494 (98.6%). The overall antibiotic prescribing rates for children's RTIs were 25% and 15.8% (p=0.018) in intervention and control groups, respectively. We found evidence of postrandomisation differential recruitment: the number of children recruited to the intervention arm was higher (292 vs 209); over half were recruited by prescribing nurses compared with less than a third in the control arm; children in the intervention arm were younger (median age 2 vs 3 years controls, p=0.03) and appeared to be more unwell than those in the control arm with higher respiratory rates (p<0.0001), wheeze prevalence (p=0.007) and global illness severity scores assessed by carers (p=0.045) and clinicians (p=0.01). Interviews with clinicians confirmed preferential recruitment of less unwell children to the trial, more so in the control arm. Differential recruitment may explain the paradoxical antibiotic prescribing rates. Future cluster level studies should consider designs which remove the need for individual consent postrandomisation and embed the intervention within electronic primary care records. ISRCTN 23547970 UKCRN STUDY ID: 16891.

Blair PS ，Turnbull S ，Ingram J ，Redmond N ，Lucas PJ ，Cabral C ，Hollinghurst S ，Dixon P ，Peters T ，Horwood J ，Little P ，Francis NA ，Gilbertson A ，Jameson C ，Hay AD ... -  《BMJ Open》