What are the effects of providing real-time population-based data on locally circulating microbes and syndromic presentations on clinician management of common infections in primary care? A quantitative systematic review.

Problem

The WHO describes antimicrobial resistance (AMR) as one of the greatest challenges to global public health. Over consumption of antibiotics is a major driver for the development of AMR. Clinician uncertainty has been identified as a key driver of antibiotic prescribing in primary care. Bayesian diagnostics asserts that clinicians first establish a pre-test diagnostic probability, which is modified (post-test probability) as further clinical evidence becomes available. The aim of this study is to investigate if interventions to inform clinicians about locally relevant population-based microbiological and syndromic surveillance reduce diagnostic uncertainty (with or without evidence of changing diagnostic probabilities) and improve antibiotic prescribing.

Approach

We are systematically reviewing studies that evaluate provision of population-based, locally relevant, microbiological or syndromic surveillance data to primary care clinicians. Studies are eligible if investigating effects on primary care clinician management of common respiratory and gastrointestinal infections in developed countries where the intervention includes dissemination of real-time, population-based data on locally relevant microbes or syndromic presentations. The main outcomes of interest are antibiotic prescribing and secondary care referral rates. Studies are ineligible if investigating effects of information regarding HIV, malaria, tuberculosis and sexually transmitted infections.The eligibility criteria for retrieval of full-text articles include a range of study designs; randomised and non-randomised trials, prospective and retrospective cohort studies, unpublished studies and informal systems where surveillance data is being distributed to primary care clinicians. Databases searched were Medline, EMBASE, CINAHL, and Web of Science, from database inception to February 2016, using MeSH terms for the following categories: dissemination of information, population-based data, common infections and primary care. Grey literature sources, thesis databases and trial registries are being searched for unpublished studies, with additional studies being identified through screening references and contacting experts. Quality assessments are specific to study design: we are using the Cochrane Risk of Bias tool for RCTs; and the ROBINS-I tool for non-randomised intervention studies.

Findings

Electronic searches identified 8370 hits, of which 2737 were duplicates, leaving 5633 potentially eligible studies. Findings will be presented on effect sizes (antibiotic prescribing and secondary care referrals), quality and validity of studies, with meta-analyses conducted if appropriate. Where appropriate, we will assess the development, execution and evaluation of the interventions in line with the MRC guidelines on complex interventions, as well as any use of behaviour change models. A strength of evidence assessment will be undertaken using the GRADE tool. All results will be reported according to PRISMA guidelines.

Consequences

Thorough evaluation of the included studies will determine the effectiveness and efficacy of this type of intervention. This systematic review will provide an essential contribution to inform design of future interventions and guide the design of ethical and robust studies seeking to evaluate complex interventions in a primary care setting.