Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol

Diane Ashiru-Oredope; Shannon Guild; Caroline De Brún; Jack Duxbury Barber; Rhys Kingston; Jocelyn Elmes; Sarah Gerver; Matthew Inada-Kim; Colin Brown; Russell Hope; Assessing Risk factors and Inequalities for antimicrobial-resistant bloodstream infectionS in England. (ARISE) Systematic Review Protocol Group

doi:10.3310/nihropenres.14297.2

Home Browse Factors associated with inequalities in antimicrobial-resistant bloodstream...

ALL Metrics

Views

Downloads

Get PDF

Get XML

Export

▬

✚

Study Protocol

Revised

Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol

[version 2; peer review: 1 approved with reservations]

Previously titled: "Determinants of inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol"

Diane Ashiru-Oredope ^1,2, Shannon Guild¹, Caroline De Brún¹, [...] Jack Duxbury Barber³, Rhys Kingston¹, Jocelyn Elmes¹, Sarah Gerver¹, Matthew Inada-Kim^1,4, Colin Brown¹, Russell Hope¹, Assessing Risk factors and Inequalities for antimicrobial-resistant bloodstream infectionS in England. (ARISE) Systematic Review Protocol Group

Diane Ashiru-Oredope ^1,2, Shannon Guild¹, [...] Caroline De Brún¹, Jack Duxbury Barber³, Rhys Kingston¹, Jocelyn Elmes¹, Sarah Gerver¹, Matthew Inada-Kim^1,4, Colin Brown¹, Russell Hope¹, Assessing Risk factors and Inequalities for antimicrobial-resistant bloodstream infectionS in England. (ARISE) Systematic Review Protocol Group

PUBLISHED 28 Oct 2026

Author details Author details

¹ UK Health Security Agency, London, UK
² University of Nottingham School of Pharmacy, Nottingham, England, UK
³ University of Aberdeen School of Biological Sciences, Aberdeen, Scotland, UK
⁴ NHS England, Redditch, England, UK

Diane Ashiru-Oredope
Roles: Conceptualization, Data Curation, Methodology, Project Administration, Writing – Original Draft Preparation, Writing – Review & Editing

Shannon Guild
Roles: Data Curation, Formal Analysis, Project Administration, Writing – Original Draft Preparation, Writing – Review & Editing

Caroline De Brún
Roles: Data Curation, Methodology, Writing – Review & Editing

Jack Duxbury Barber
Roles: Methodology, Writing – Original Draft Preparation, Writing – Review & Editing

Rhys Kingston
Roles: Methodology, Project Administration, Writing – Review & Editing

Jocelyn Elmes
Roles: Methodology, Writing – Review & Editing

Sarah Gerver
Roles: Methodology, Writing – Review & Editing

Matthew Inada-Kim
Roles: Supervision, Writing – Review & Editing

Colin Brown
Roles: Supervision, Writing – Review & Editing

Russell Hope
Roles: Methodology, Supervision, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

Abstract

Introduction

Emerging evidence suggests that the burden of antimicrobial resistant (AMR) bloodstream infections (BSI) disproportionately affects populations with factors commonly associated with health inequalities, including high deprivation levels and ethnic minority groups. The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).

Methods and analysis

This protocol follows the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) and standards of the reporting equity-focused systematic reviews (PRISMA-E) guidelines. It is informed by the Cochrane methodology for prognosis and equity. We will search five electronic databases (Ovid Embase, Ovid MEDLINE, CINAHL, Scopus, Cochrane Central Register of Controlled Trials) using tailored search terms. Grey literature sources will also be screened. Peer-reviewed observational, intervention and mixed methods studies will be eligible if they report a) quantitative data on AMR BSI and b) factors associated with reported health disparities (e.g. deprivation score; education; ethnicity) in socioeconomically disadvantaged populations from Organisation for Economic Co-operation and Development member countries. No language or publication date limits will be applied. A team of reviewers will complete screening of allocated abstracts and titles for eligibility and a large language model will conduct a secondary screening. Full-text screening, data extraction and risk-of-bias assessment will be undertaken by trained members of the review team following predefined quality assurance procedures and validation exercises. A narrative synthesis will summarise findings, with meta-analysis conducted only where sufficient methodological and clinical homogeneity exist. This review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024601049).

Ethics and dissemination

We will identify factors associated with disparities in AMR BSI and use these findings to inform further modelling work, guide clinical practice, and support targeted public health interventions within HICs.

Plain Language Summary

Background

Some people are more likely than others to get serious bloodstream infections that do not respond to antibiotics. These antibiotic-resistant infections can be life-threatening. People living with social disadvantage — such as low income, poor housing, or limited access to healthcare — may face a higher risk. However, we do not yet know which factors matter most.

Methods

We will collect and review all studies from high-income countries that report on resistant bloodstream infections and social or economic factors such as income, housing, education, ethnicity or deprivation. We will check the quality of each study and summarise the findings. Where possible, we will combine results across studies.

Sharing the findings

We will share our results with healthcare professionals, policymakers, researchers and the public. Our aim is to help improve infection prevention and reduce unfair differences in health.

Keywords

Antimicrobial exposure/resistance, Bloodstream infections, Disparities, Equality, High-income countries, Inequalities, Risk factors, Sepsis, Socioeconomic deprivation, OECD

Corresponding author: Diane Ashiru-Oredope

Competing interests: No competing interests were disclosed.

Grant information: The project is funded by National Institute for Health and Care Research (NIHR) Senior Clinical and Practitioner Award (NIHR304553) to Diane Ashiru-Oredope. The views expressed are those of the contributors and not necessarily those of the National Institute for Health and Care Research or the Department of Health and Social Care.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2026 Ashiru-Oredope D et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Ashiru-Oredope D, Guild S, De Brún C et al. Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol [version 2; peer review: 1 approved with reservations]. NIHR Open Res 2026, 6:87 (https://doi.org/10.3310/nihropenres.14297.2) First published: 13 Jul 2026, 6:87 (https://doi.org/10.3310/nihropenres.14297.1) Latest published: 28 Oct 2026, 6:87 (https://doi.org/10.3310/nihropenres.14297.2)

Revised Amendments from Version 1

We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol. The key changes in the revision are:
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes: “Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants: “The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”
Additional rationale text added to clarify the unique focus of the current review: “While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate. “To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

See the authors' detailed response to the review by Razique Anwer

Strengths and limitations of this study

• To the best of our knowledge, whilst there have been reviews that have assessed social determinants of AMR, resistant infections more broadly and socioeconomic influences on AMR, none have focused on synthesising evidence on disparities in AMR BSI in socio-economically deprived populations in high-income countries.
• The large number of papers returned by the initial search limits ability to independently screen and review all papers by more than one human-reviewer.
• The use of an Large Language Model (LLM) as a second reviewer at the abstract screening stage has the potential to reduce risk of bias while improving efficiency and contributing to the wider evidence around artificial intelligence-assisted evidence synthesis.
• It may be difficult to employ meta-analysis and calculate valid measures of effect due to likely high heterogeneity (clinical, methodological, or statistical) across studies.

Introduction

While antimicrobials are considered a lynchpin of modern medicine, their unnecessary and inappropriate use has driven the emergence of resistant pathogens in humans, animals and the environment.¹ Antimicrobial resistance (AMR) is a global crisis, resulting in infections that are more challenging to treat and lead to greater morbidity and mortality. In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3 Within high-income countries (HICs) like the UK, such disparities in the incidence, prevalence and the consequences of drug-resistant infections are a product of avoidable, inequitable differences between various groups and communities.⁴ These health inequalities result from—and may in turn exacerbate—a complex intersection of systemic societal inequalities, including structural racism, gendered disparities and socioeconomic deprivation.^5,6

Evidence describing the effect of socioeconomic disadvantage—whether defined in relation to income, employment insecurity, inadequate housing or other indicators—on the burden of AMR remains limited.^7,8 Housing and living conditions have been associated with AMR in clinical and community settings within HICs,^9–13 and similar evidence exists for income status,^14,15 education level,^13,16,17 and deprivation index score, although findings relating to the latter are inconsistent.^10,18 Data exploring socioeconomic influences on antibiotic exposure are also sparse.

There is emerging evidence of both increased rates of infection and resistant infection,⁴ as well as higher levels of antimicrobial exposure, in individuals with factors commonly associated with health inequalities.¹⁹ For example, increased prevalence of hepatitis C/B and tuberculosis has been observed in homeless populations, individuals with substance use disorders, sex workers and individuals in contact with the justice system.²⁰ Furthermore, ethnic minorities, individuals experiencing deprivation and inclusion health groups–including vulnerable migrants such as asylum seekers, refugees, unaccompanied children, trafficked individuals and people with insecure immigration status–also face higher risk of infections including tuberculosis, sexually transmitted infections and MSSA.^21–24

Higher levels of antibiotic use have been reported among older adults, primarily those in residential care or living in more deprived areas.¹⁹ Furthermore, structural barriers to healthcare access mean that some vulnerable migrants rely on alternative routes of antibiotic supply, potentially increasing the likelihood of inappropriate antibiotic use.¹⁹ Data published in the 2022–2023 English Surveillance Programme for Antimicrobial Utilisation and Resistance (ESPAUR) report highlights the difference in AMR burden seen in populations that have several factors commonly associated with health inequalities, including variation by deprivation levels, ethnicity, age and region. Rates of AMR bloodstream infections (BSI) are 41.0% higher in the most deprived compared to the least deprived groups (33.0 versus 23.4 infections per 100,000 population).²⁵ Furthermore, this percentage difference between the most and least deprived groups in AMR BSI has increased by 7% from 2019 to 2022, while the difference in rate of overall BSI has remained constant over this time period, suggesting a specific widening of the burden of resistant infections between least and most deprived groups.²⁶

The 2022–2023 ESPAUR report also highlights a higher burden of AMR BSI within Asian and British Asian compared to White individuals (34.6 versus 18.7%). The rate of Carbapenemase-producing Gram-negative bacteria infection was also higher among Asian and British Asian ethnic groups (7.7 per 100,000 Asian population versus 4.1 per 100,000 White population). AMR burden was also shown to vary by age. The rate of AMR BSI was highest in >74-year-olds (157.0 per 100,000) and lowest in the 10–14-year-olds (1.8 per 100,000). Of note, while the rate of AMR BSI for children in all age groups between 1 and 14 was less than 5.0 per 100,000, the rate for children under one-year-old was comparatively high at 46.5 per 100,000. Finally, regional variation in AMR BSI was reported, with the London region reporting the highest burden (39.2 per 100,000 population) followed by the Northwest (32.9 per 100,000 population). The lowest burden was recorded in the Southwest (22.8 per 100,000 population). This data echoes other findings in the literature that demonstrate higher rates of resistant infections in minority ethnic populations,^27–30 with economic deprivation^31,32 and within inclusion health groups, such as vulnerable migrants.^31,33

Rationale

Although socioeconomic factors are increasingly recognised as important factors associated with AMR incidence and burden, existing systematic reviews have largely examined AMR as a whole rather than focusing on bloodstream infections, which carry some of the highest morbidity and mortality. Recent UKHSA-led reviews highlight this gap: one review mapping social determinants of AMR identified socioeconomic status, overcrowded housing and educational barriers as key contributors to resistant infection risk, but also noted limited evidence for specific disadvantaged groups and for upstream structural drivers.³⁴ A related scoping review in high-income countries found consistent associations between deprivation, ethnicity, age and higher rates of resistant infections—particularly MRSA and resistant E. coli—and showed that adjusting for socioeconomic factors substantially reduced apparent racial disparities.³⁵ Additional evidence demonstrating geographic clustering of AMR in the most deprived areas further reinforces the role of structural disadvantage in shaping AMR risk.³⁶ While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes. Given the substantial morbidity and mortality associated with AMR BSI, and evidence of widening disparities between the most and least deprived populations, a dedicated synthesis of the evidence is warranted. This review aims to address that gap by examining factors associated with socioeconomic inequalities in AMR BSI across high-income settings.

Taken together, these findings highlight the importance of identifying both modifiable and non-modifiable prognostic factors that contribute to disparities in AMR BSI and antibiotic exposure. A clearer understanding of these factors may support the development of targeted interventions to reduce the incidence of resistant infections and help clinicians better identify and manage patients at elevated risk.

Review objectives

The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among populations identified through socioeconomic indicators and related PROGRESS-Plus characteristics in HICs. The review will achieve this through the following objectives:

• Primary objectives:
- ○ To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.
- ○ To quantify incidence and burden of AMR BSIs among socioeconomically disadvantaged populations in HICs (dependent on quality of literature available)
- ○ To quantify estimates of association between identified risk factors and AMR BSI incidence and burden among socioeconomically disadvantaged populations in HICs using meta-analyses (dependent on quality of literature available)
• Secondary objectives:
- ○ To assess within socioeconomically disadvantaged populations in HICs:
  - ▪ which identified factors are potentially modifiable
  - ▪ health outcomes associated with AMR BSI, including mortality, morbidity and length of hospital stay
  - ▪ healthcare utilisation outcomes, including antimicrobial prescribing, healthcare access and healthcare costs associated with AMR BSI.
  - ▪ reported interventions to mitigate against AMR BSIs including preventative measures such as vaccination and healthcare infection, prevention and control measures

Methods

Design

This protocol has been prepared in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) with principles of PRISMA-Equity integrated.³⁷ The systematic review and meta-analysis will also be informed by the Cochrane methodology for prognosis and equity.^38,39 This review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number: CRD42024601049).⁴⁰ Registration was completed prior to completion of the literature search. This protocol is submitted after title and abstract screening; before completion of full text screening and data extraction to promote transparency and reduce the risk of selective reporting.

Any amendments in the protocol will be documented in the final review.

Electronic searches

Information sources

Five databases (Ovid Embase, Ovid MEDLINE, CINAHL, Scopus, Cochrane Central Register of Controlled Trials) were searched in December 2024 for all relevant studies and the search will be updated in 2026. Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases. National surveillance reports of eligible countries on AMR and antimicrobial consumption will be retrieved from manual online searches. Relevant surveillance reports and databases published by World Health Organization (WHO) and European Centre for Disease Prevention and Control (ECDC) will also be included. Snowballing of references and citations of included studies will be used to identify additional studies that meet the eligibility criteria. There will be no publication date or language restrictions.

Search terms

Search terms will be tailored for each database, using MeSH terms where applicable, and will be developed in collaboration with an information scientist (review team member) to ensure the search syntax is correct. A summary of search terms is included in Table 1. The full search strategy will be published in the with the results of the review.

Table 1. Summary of search strategy for electronic database search.

Concept	Examples of search terms*
Bloodstream infections	bloodstream infections OR bacteraemia OR sepsis OR Escherichia coli OR methicillin-resistant staphylococcus aureus (MRSA)
Antibiotic resistance	antimicrobial resistance OR antimicrobial consumption OR infection adj (rate or sensitivity) OR resistant adj (infection or organism)
Disparities	health care disparity OR inequity OR socioeconomic OR social determinants of health OR SES OR racial OR poverty OR deprivation OR lowest income OR religion OR age distribution OR disability OR vulnerable population
Risk factors	risk factors OR relative risk OR prevalence OR incidence OR prognosis OR morbidity OR mortality

* Not the complete list of search terms and exact search terms were adapted for each database.

Eligibility

Studies will be eligible when they report microbiologically confirmed bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance and provide data stratified by, or analytically associated with, socioeconomic indicators and/or relevant PROGRESS-Plus characteristics associated with health inequalities. The detailed inclusion and exclusion criteria for studies in this systematic review are outlined below.

Population

Socioeconomically disadvantaged populations in HICs⁴¹ with AMR BSI. HICs will be defined by membership of the Organisation for Economic Co-operation and Development (OECD) high-income status group, Figure 1.⁴¹ All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications were examined in relation to inequalities in AMR BSI and related outcomes. Studies conducted in low- or middle-income countries (as defined by non-membership of the OECD high-income status group) or that do not report data on related to socioeconomic disadvantaged populations will be excluded.

Figure 1. Map* of the OECD’s members by estimated population size in 2024.

The OECD’s 38 members are: Austria, Australia, Belgium, Canada, Chile, Colombia, Costa Rica, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Latvia, Lithuania, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. *Created using Microsoft Excel Map Chart tool.

Intervention(s) or exposure(s)

The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.

Comparator(s) or control(s)

Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.

Outcomes

The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance. Resistance definitions used by study authors will be extracted and considered when interpreting findings. This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.

Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.

Measures of association

Measures of association (e.g. odds ratios/risk differences) between incidence/burden of AMR BSI and identified risk factors for socioeconomically disadvantaged populations where these are directly reported by included studies.

Study design

Peer-reviewed observational studies conducted in populations living in HICs ( Figure 1) and national and international health surveillance reports of relevant countries/regions containing quantitative data on AMR BSI by socioeconomic status will be included. Interventional studies (randomised control trials) and mixed-method studies will also be included if they meet the above criteria. Grey literature from PhD and Master’s theses will also be included, provided above criteria are met. Case reports, commentaries, news articles, conference abstracts will be excluded.

Screening

Initial title and abstract screening

The title and abstract screening process is summarised in Figure 2. A team of reviewers will be involved in the study selection process via screening of abstracts and titles, and then full text. During initial registration of the study, it was intended that titles and abstracts from all studies from literature searches would be screened for inclusion independently by two authors. On an initial search and retrieval of >16,000 papers, the approach was updated such that a small sample of titles and abstracts (n = 100) will be screened by all reviewers in a validation screening for training and validation to ensure consistent study selection. Based on new internal capability to use large language models for screening, the methodology was adapted to include large language model (LLM) for screening.

Figure 2. Screening methodology of manual human reviewers and AI Abstract Screening Tool.

AI, artificial intelligence; LLM, large language model.

We will assess inter-rater reliability using Fleiss’ multi-rater Kappa where a test set of abstracts are independently assessed by three or more reviewers.^42,43 Fleiss’ multi-rater kappa can have a value of −1 to 1 where 0 indicates no agreement beyond random chance, 1 indicates perfect agreement and − 1 indicates perfect disagreement (i.e. agreement worse than chance).⁴⁴

Use of a large language model for screening

Following the completion of manual abstract screening, a secondary review stage will be undertaken using a large language model (LLM). This step has two purposes: first, to identify any abstracts that may have been inadvertently excluded during human review, and second, to contribute to a wider programme of work within UKHSA evaluating the role of LLMs in evidence synthesis.

The analysis will be conducted using a UKHSA internal tool developed to query LLMs. A free-text prompt will be drafted, asking the model to compare the abstract with the pre-defined criteria and assign a classification of either “include” or “exclude” based on the content. The Llama 3.3 70b model will be used for this task.

Prior to large-scale application, we will complete a prompt iteration process to optimise model performance. An initial set of 100 abstracts will be independently reviewed and labelled by multiple senior evidence reviewers. The majority label to include or exclude an abstract will be used as the ground truth for evaluation in prompt iteration. Across successive iterations, the LLM’s outputs will be compared to the majority expert labels, with precision, recall, and Cohen’s kappa calculated at each stage. The prompt will be refined until the model’s performance reaches satisfactory alignment with expert judgement. In line with best practice for systematic evidence review, a recall threshold of 95% is set as the primary target.

Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.

Full text screening

Before commencing the main full-text screening process, all reviewers involved in full-text assessment will independently screen a common validation set of studies. Discrepancies will be discussed and eligibility criteria refined where necessary to promote consistency of decision-making.

Following validation, each paper will be reviewed by a single reviewer and 10% of excluded papers will be independently assessed by a second reviewer. If more than 10% of excluded papers are reassigned to be included, all exclusions undertaken by that reviewer will be reassessed by another reviewer.

Data management and extraction

Prior to commencing data extraction, all reviewers involved in extraction will independently extract data from a common validation set of studies using the pre-specified extraction form. Discrepancies will be discussed and resolved through consensus to ensure consistency of extraction across reviewers.

Following validation, eligible studies will be allocated to individual reviewers for extraction. To minimise bias, reviewers will not extract data from studies that they previously screened during full-text review where feasible. Queries arising during extraction will be discussed within the review team and resolved through consensus.

Data will be managed using the EndNote reference manager, Rayyan⁴⁵ and Microsoft Excel Office software. Rayyan will be used to store extracted studies and record blinded decision-making throughout the review process. Both quantitative and qualitative data will be extracted from the included reviews to an excel workbook. Data extracted will include first author, year of publication, title, DOI, publication category (observational or interventional), socioeconomic metrics analysed (housing conditions, income status etc), primary and secondary outcomes measured (antibiotic resistance or antibiotic prescribing or both), metric used to measure any outcome (e.g. the prevalence of BSIs, the proportion of each bacteria causing BSIs and AMR rates of bacteria groups), country/region of study and whether the study examines any risk factors (and which one(s) and whether they are modifiable or non-modifiable). Interventions or preventative measures reported in study will also be extracted.

To test usability of the data extraction form, we will conduct a pilot exercise using five randomly selected studies. Reviewers will independently extract data from the same studies, followed by a consensus meeting to calibrate extraction procedures and refine the extraction form where necessary.

Risk of bias (quality) assessment

Risk-of-bias assessments of each included study will be conducted independently by two reviewers. Disagreements will be resolved via discussion, and if that fails, by consulting a third reviewer. Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) will be used for cohort studies to comprehensively assess risk of bias across domains, Newcastle-Ottawa Scale (NOS) will be used for case-control studies. For Randomized Controlled Trials (RCTs) Cochrane Risk of Bias Tool (RoB 2) will be used.

For qualitative/mixed methods studies, reviewers will use the Mixed Methods Appraisal Tool (MMAT) to assess the quality of included studies.¹ It includes five core quality criteria for each of the following five categories of study designs: (a) qualitative, (b) randomized controlled, (c) nonrandomized, (d) quantitative descriptive, and (e) mixed methods. Finally, we will assess the risk of bias for prevalence studies using JBI Critical Appraisal Checklist for Prevalence Studies.

Plan for data synthesis

Given the anticipated heterogeneity in populations, socioeconomic measures, pathogens, resistance definitions and outcome measures, narrative synthesis is expected to be the primary method of evidence synthesis. Studies will be grouped according to outcome type, socioeconomic measure, pathogen and resistance phenotype where appropriate.

Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.

In the case that metanalysis is possible, we will report heterogeneity using two measures:

• Cochrane’s χ² test (Cochran’s Q), which examines the null hypothesis that all studies are evaluating the same effect but may not always accurately detect heterogeneity.
• Higgins’s I² which represents the percentage of variation between the sample estimates that is due to heterogeneity rather than to sampling error (tells us what proportion of the total variation across studies is beyond chance) will also be considered. It can take on values from 0 to 100%, with 100% being the maximum level of heterogeneity. Often, I² values below 25% are considered low, 25 to 50% moderate, and above 75% high heterogeneity.

For studies evaluating an intervention, a list of interventions present in the available scientific literature will be prepared, alongside the pooled effect estimates from a meta-analysis of quantitative outcomes where appropriate. These findings will be further discussed with all the relevant stakeholders to frame the appropriate recommendations.

If possible, we will conduct sub-group analyses informed by the PROGRESS-Plus framework, which identifies characteristics across which health opportunities and outcomes may be unequally distributed.^39,46 PROGRESS domains include Place of residence/regional variation, Race/ethnicity/culture/language, Occupation, Gender/sex, Religion, Education, Socio-economic status and Social capital. Other characteristics (‘Plus’) such as sexual orientation, age, disability and type of resistant infections will also be considered.

If appropriate we will conduct sensitivity analysis to assess the robustness of subgroup analyses and by excluding studies evaluated to have a high risk of bias. Funnel plots will be visually inspected for asymmetry, and Egger’s test will be used to assess publication bias if more than ten studies are included in the meta-analysis.

Patient and public involvement

Patients with lived experience and patient support team at Antibiotic Research UK (listed under the ARISE Systematic Review Protocol Group or in the acknowledgments) helped determine the relevance and potential benefit of the topic, the priority area as well as helped shape the scope. They also fully contributed to refining the review research question and identifying outcomes of relevance. Feedback from these individuals along with additional people with lived experience of serious infection and healthcare disadvantage will be sought following extraction and synthesis of the findings to inform the interpretation of findings and prioritisation of outcomes. Patients and the public will not be involved in extraction or analysis.

Ethics and dissemination

Evidence on risk factors associated with disparities in resistant BSIs is currently sparse. This is partly due to variability in recording and reporting across and within peer reviewed manuscripts and surveillance reports. This impedes comparability of results and utility of available data. We will utilize the collated list of risk factors from this review to investigate their impact within linked UK health databases to model disease burden and impact on severe adverse health outcomes such as mortality. These analyses will be used to inform clinical practice and ascertain intervention targets.

This systematic review is part of a broader project which will also aim to analyse pseudonymised primary care electronic health records, assess the contribution of Pharmaceutical Public Health (PPH) in reducing inequalities with resistant infections and antibiotic use/exposure and Co-design an intervention trial protocol and policy recommendations.

Data availability

Underlying data

No underlying data are associated with this article as it describes a study protocol rather than reporting study findings.

Extended data

Additional file supporting this article can be accessed via the Open Science Framework at https://doi.org/10.17605/OSF.IO/2D3VK.⁴⁷

The project contains the following file: PRISMA-P reporting checklist for protocols with principles of PRISMA-Equity integrated: https://doi.org/10.17605/OSF.IO/2D3VK.⁴⁷

Description of data: The completed PRISMA-P-Equity embedded Checklist, outlines how each item was addressed in the protocol.

Reporting guidelines

The completed PRISMA-P reporting checklist for protocols with principles of PRISMA-Equity integrated is available via https://doi.org/10.17605/OSF.IO/2D3VK.⁴⁷

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Acknowledgements

The Assessing Risk factors and Inequalities for antimicrobial-resistant bloodstream infectionS in England (ARISE). Systematic Review Protocol Group includes the authors as well as Christy Lee, Karina Micah, Orlagh Quinn, Faiza Ambreen and Miriam Onuwale (AMR and HCAI Division, Chief Medical Adviser Group, UK Health Security Agency); Xiaomin Zhong (Nuffield Department of Population Health, University of Oxford); Adeola Henry Ayeni (Public Health, Teesside University); Gbemisola Peter-Ayeni (Nottingham University Hospitals NHS Trust); Tomisin Owolabi (Imperial College London); Damilola Olaoye (University of Bristol ); Dee Grant, Aileen Bryson and Vanessa Carter (Patient representatives, London); and Ollie Higgins, Timothy Laurence, Felix Feldman, Toby Nonnenmacher, Joshua Harris and Leo Loman (Advanced Analytics Team, Chief Data Officer Group, UK Health Security Agency).

We acknowledge Fran Husson (patient representative, English Surveillance Programme for Antimicrobial Utilisation) for her contribution to discussions about research design that informed the final research questions and method.

References

1. Okeke IN, de Kraker MEA , Van Boeckel TP, et al.: The scope of the antimicrobial resistance challenge. Lancet. 2024; 403(10442): 2426–2438. Publisher Full Text
2. Antimicrobial Resistance C: Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022; 399(10325): 629–655. Publisher Full Text
3. The Lancet: Antimicrobial resistance: an agenda for all. Lancet. Jun 1 2024; 403(10442): 2349. Publisher Full Text
4. Ayorinde A, Ghosh I, Ali I, et al.: Health inequalities in infectious diseases: a systematic overview of reviews. BMJ Open. 2023; 13(4): e067429. Publisher Full Text
5. CSDH: Closing the gap in a generation: health equity through action on the social determinants of health. Final Report of the Commission on Social Determinants of Health. Geneva: World Health Organization; 2008.
6. Marmot M: Fair Society, Healthy Lives: Strategic review of health inequalities in England post-2010.2010.
7. Alividza V, Mariano V, Ahmad R, et al.: Investigating the impact of poverty on colonization and infection with drug-resistant organisms in humans: a systematic review. Infect. Dis. Poverty. 2018; 7: 76. PubMed Abstract | Publisher Full Text | Free Full Text
8. Tony Atkinson BC, Marlier E, Nolan B: Social Indicators: The EU and Social Inclusion.2002.
9. Jourdain S, Smeesters PR, Denis O, et al.: Differences in nasopharyngeal bacterial carriage in preschool children from different socio-economic origins. Clin. Microbiol. Infect. 2010; 17.
10. Bagger JP, Taylor KM: Postoperative infection with meticillin-resistant Staphylococcus aureus and socioeconomic background. Lancet. 2004; 363.
11. Young DM, Charlebois ED, Chambers H, et al.: An Epidemic of Methicillin-Resistant Staphylococcus aureus Soft Tissue Infections Among Medically Underserved Patients. Arch. Surg. 2004; 139.
12. Bratu S, Landman D, Gupta J, et al.: A population-based study examining the emergence of community-associated methicillin-resistant Staphylococcus aureus USA300 in New York City. Ann. Clin. Microbiol. Antimicrob. 2006; 5: 29. Publisher Full Text
13. Nomamiukor BO, Horner C, Kirby A, et al.: Living conditions are associated with increased antibiotic resistance in community isolates of Escherichia coli. J. Antimicrob. Chemother. 2015; 70(11): 3154–3158. PubMed Abstract | Publisher Full Text
14. McMullen KM, Warren DK, Woeltje KF: The changing susceptibilities of methicillin-resistant Staphylococcus aureus at a midwestern hospital: the emergence of “community-associated” MRSA. Am. J. Infect. Control. 2009; 37(6): 454–457. Publisher Full Text
15. Chen FM, Farley M, Plikaytis B, et al.: Geocoding and Linking Data from Population-based Surveillance and the US Census to Evaluate the Impact of Median Household Income on the Epidemiology of Invasive Streptococcus pneumoniae Infections. Am. J. Epidemiol. 1998; 148.
16. Huang SS, Finkelstein JA, Rifas-Shiman SL, et al.: Community-level predictors of pneumococcal carriage and resistance in young children. Am. J. Epidemiol. Apr 1 2004; 159(7): 645–654. Publisher Full Text
17. Garcia-Rey C, Fenoll A, Aguilar L, et al.: Effect of social and climatological factors on antimicrobial use and Streptococcus pneumoniae resistance in different provinces in Spain. J. Antimicrob. Chemother. Aug 2004; 54(2): 465–471. PubMed Abstract | Publisher Full Text
18. Parsons HK, Carter MJ, Sanders DS, et al.: Helicobacter pylori antimicrobial resistance in the United Kingdom: the effect of age, sex and socio-economic status. Aliment. Pharmacol. Ther. Sep 2001; 15(9): 1473–1478. Publisher Full Text
19. Harvey EJ, De Brun C, Casale E, et al.: Influence of factors commonly known to be associated with health inequalities on antibiotic use in high-income countries: a systematic scoping review. J. Antimicrob. Chemother. Apr 3 2023; 78(4): 861–870. Publisher Full Text
20. Aldridge RW, Story A, Hwang SW, et al.: Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis. Lancet. Jan 20 2018; 391(10117): 241–250. PubMed Abstract | Publisher Full Text | Free Full Text
21. Crichton J, Hickman M, Campbell R, et al.: Prevalence of chlamydia in young adulthood and association with life course socioeconomic position: birth cohort study. PLoS One. 2014; 9(8): e104943. PubMed Abstract | Publisher Full Text | Free Full Text
22. Oren E, Koepsell T, Leroux BG, et al.: Area-based socio-economic disadvantage and tuberculosis incidence. Int. J. Tuberc. Lung Dis. Jul 2012; 16(7): 880–885. PubMed Abstract | Publisher Full Text
23. Pini A, Stenbeck M, Galanis I, et al.: Socioeconomic disparities associated with 29 common infectious diseases in Sweden, 2005-14: an individually matched case-control study. Lancet Infect. Dis. Feb 2019; 19(2): 165–176. PubMed Abstract | Publisher Full Text
24. Mancuso JD, Diffenderfer JM, Ghassemieh BJ, et al.: The Prevalence of Latent Tuberculosis Infection in the United States. Am. J. Respir. Crit. Care Med. Aug 15 2016; 194(4): 501–509. Publisher Full Text
25. UK Health Security Agency: English surveillance programme for antimicrobial utilisation and resistance (ESPAUR) Report 2022 to 2023.2023.
26. Gilham EL, Higgins H, Mason E, et al.: National reporting on factors associated with health inequalities and infection, antimicrobial resistance and consumption rates in England: 2018–2022. Lancet. 2024; 404. Publisher Full Text
27. Ray GT, Suaya JA, Baxter R: Trends and characteristics of culture-confirmed Staphylococcus aureus infections in a large U.S. integrated health care organization. J. Clin. Microbiol. Jun 2012; 50(6): 1950–1957. PubMed Abstract | Publisher Full Text | Free Full Text
28. Turnidge SYCTSJvHLEBJCJD: Impact of ethnicity and socio-economic status on Staphylococcus aureus bacteremia incidence and mortality: a heavy burden in Indigenous Australians. BMC Infect. Dis. 2012; 12.
29. Gualandi N, Mu Y, Bamberg WM, et al.: Racial Disparities in Invasive Methicillin-resistant Staphylococcus aureus Infections, 2005-2014. Clin. Infect. Dis. Sep 28 2018; 67(8): 1175–1181. Publisher Full Text
30. See I, Wesson P, Gualandi N, et al.: Socioeconomic Factors Explain Racial Disparities in Invasive Community-Associated Methicillin-Resistant Staphylococcus aureus Disease Rates. Clin. Infect. Dis. Mar 1 2017; 64(5): 597–604. Publisher Full Text
31. Tosas Auguet O, Betley JR, Stabler RA, et al.: Evidence for Community Transmission of Community-Associated but Not Health-Care-Associated Methicillin-Resistant Staphylococcus Aureus Strains Linked to Social and Material Deprivation: Spatial Analysis of Cross-sectional Data. PLoS Med. Jan 2016; 13(1): e1001944. PubMed Abstract | Publisher Full Text | Free Full Text
32. Matthew R, Blakiston JTF: Population-level exposures associated with MRSA and ESBL-E. coli infection across district health boards in Aotearoa New Zealand: an ecological study. NZMJ. 2020; 133.
33. Piper Jenks N, Pardos de la Gandara M, D’Orazio BM, et al.: Differences in prevalence of community-associated MRSA and MSSA among U.S. and non-U.S. born populations in six New York Community Health Centers. Travel Med. Infect. Dis. Nov–Dec 2016; 14(6): 551–560. Publisher Full Text
34. Shutt AE, Ashiru-Oredope D, Price J, et al.: The intersection of the social determinants of health and antimicrobial resistance in human populations: a systematic review. BMJ Glob. Health. May 30 2025; 10(5). Publisher Full Text
35. Ghataure A, Gilham EL, Casale E, et al.: Association of factors linked to health inequalities and the risk of antibiotic-resistant infection in high-income countries: a systematic scoping review. JAC Antimicrob. Resist. 2025; 7. PubMed Abstract | Publisher Full Text | Free Full Text
36. Levitch E, Matthews L, Choi E, et al.: The impact of socioeconomic status on the prevalence of antimicrobial resistance in high-income nations: a systematic review. Antimicrob. Steward. Healthc. Epidemiol. 2025; 5(1): e264. PubMed Abstract | Publisher Full Text | Free Full Text
37. Moher D, Shamseer L, Clarke M, et al.: Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst. Rev. 2015; 4.
38. Welch VA, Jull J, Hartling L, et al.: Cochrane Handbook for Systematic Reviews of Interventions - Chapter 16: Equity and specific populations. Cochrane Handbook for Systematic Reviews of Interventions. 2024.
39. Welch V, Petticrew M, Tugwell P, et al.: PRISMA-Equity 2012 extension: reporting guidelines for systematic reviews with a focus on health equity. PLoS Med. 2012; 9(10): e1001333. PubMed Abstract | Publisher Full Text | Free Full Text
40. Ashiru-Oredope D: Data from: Risk factors associated with disparities in antibiotic-resistant bloodstream infections among socioeconomically disadvantaged groups in High-Income Countries. PROSPERO; 2024. Reference Source
41. Bank TW: Population, total - OECD members. Accessed 17/11/2025. Reference Source
42. Welch V, Brand K, Kristjansson E, et al.: Systematic reviews need to consider applicability to disadvantaged populations: inter-rater agreement for a health equity plausibility algorithm. BMC Med. Res. Methodol. 2012; 12. PubMed Abstract | Publisher Full Text | Free Full Text
43. Wu G, Eastwood C, Sapiro N, et al.: Achieving high inter-rater reliability in establishing data labels: a retrospective chart review study. BMJ Open Qual. 2024; 13(2). Publisher Full Text
44. Randolph JJ: Free-Marginal Multirater Kappa (multirater κfree): An Alternative to Fleiss’ FixedMarginal Multirater Kappa. Joensuu Learning and Instruction Symposium 2005. 2005.
45. Ouzzani M, Hammady H, Fedorowicz Z, et al.: Rayyan-a web and mobile app for systematic reviews. Syst. Rev. Dec 5 2016; 5. PubMed Abstract | Publisher Full Text | Free Full Text
46. O’Neill J, Tabish H, Welch V, et al.: Applying an equity lens to interventions: using PROGRESS ensures consideration of socially stratifying factors to illuminate inequities in health. J. Clin. Epidemiol. Jan 2014; 67(1): 56–64. PubMed Abstract | Publisher Full Text
47. Ashiru-Oredope D: Data from: PRISMA-P reporting checklist for protocols with principles of PRISMA-Equity integrated. Open Science Framework (OSF). 2026. Publisher Full Text

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 13 Jul 2026

Author details Author details

Diane Ashiru-Oredope
Roles: Conceptualization, Data Curation, Methodology, Project Administration, Writing – Original Draft Preparation, Writing – Review & Editing

Shannon Guild
Roles: Data Curation, Formal Analysis, Project Administration, Writing – Original Draft Preparation, Writing – Review & Editing

Caroline De Brún
Roles: Data Curation, Methodology, Writing – Review & Editing

Jack Duxbury Barber
Roles: Methodology, Writing – Original Draft Preparation, Writing – Review & Editing

Rhys Kingston
Roles: Methodology, Project Administration, Writing – Review & Editing

Jocelyn Elmes
Roles: Methodology, Writing – Review & Editing

Sarah Gerver
Roles: Methodology, Writing – Review & Editing

Matthew Inada-Kim
Roles: Supervision, Writing – Review & Editing

Colin Brown
Roles: Supervision, Writing – Review & Editing

Russell Hope
Roles: Methodology, Supervision, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

The project is funded by National Institute for Health and Care Research (NIHR) Senior Clinical and Practitioner Award (NIHR304553) to Diane Ashiru-Oredope. The views expressed are those of the contributors and not necessarily those of the National Institute for Health and Care Research or the Department of Health and Social Care.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Article Versions (2)

version 2

Revised

Published: 28 Oct 2026, 6:87

https://doi.org/10.3310/nihropenres.14297.2

version 1

Published: 13 Jul 2026, 6:87

https://doi.org/10.3310/nihropenres.14297.1

© 2026 Ashiru-Oredope D et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

VIEWS

243

downloads

Citations

SEE MORE DETAILS

CITE

how to cite this article

Ashiru-Oredope D, Guild S, De Brún C et al. Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol [version 2; peer review: 1 approved with reservations]. NIHR Open Res 2026, 6:87 (https://doi.org/10.3310/nihropenres.14297.2)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Version 1

VERSION 1

PUBLISHED 13 Jul 2026

Views

Reviewer Report 28 Jul 2026

Razique Anwer, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia

Approved with Reservations

https://doi.org/10.3310/nihropenres.15574.r41069

This protocol addresses an important and clinically relevant question. Inequalities in antimicrobial-resistant bloodstream infections remain considerably less well characterised than inequalities in infection or antimicrobial use more generally. The involvement of an information specialist, the inclusion of patient and public contributors, the intention to include non-English evidence, and the attempt to evaluate an LLM-supported screening workflow are notable strengths.

Major concerns_
1. Title and objective: “determinants” may imply causality that the review cannot establish
Page 1, Title:
“Determinants of inequalities in antimicrobial-resistant bloodstream infections…”
Page 5, Review objectives:
“The aim of this review is to ascertain the drivers and risk factors contributing to disparities…”
Most eligible studies are expected to be observational, ecological or surveillance based. Such studies may identify associations between social factors and AMR BSI, but it will not necessarily establish that these factors are causal “determinants” or “drivers.”
So, I suggest the authors to rewrite the “Title” _
Social and socioeconomic factors associated with inequalities in antimicrobial-resistant bloodstream infections in high-income countries: a systematic review protocol.

In Abstract, authors stated_
“Studies will be eligible if they report a) quantitative data on AMR BSI and b) factors associated with reported health disparities…”
However, Page 7, Study design states:
“…surveillance reports…containing quantitative data on AMR BSI or antibiotic prescribing by socioeconomic status will be included.”
The use of “or” permits inclusion of studies that report antimicrobial prescribing but contain no BSI or resistance outcome. Such studies cannot directly answer the primary question concerning inequalities in AMR BSI.
The authors should either exclude prescribing-only studies or define a separate secondary review question.
Page 4, “Introduction section”
“Over five million deaths annually are associated with AMR bacterial infections…”
Suggested revision:
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.”

The distinction between deaths “associated with” and “attributable to” AMR should be retained throughout the manuscript.

Page 5 “To identify risk factors considered to be associated with disparities in the incidence and burden of AMR BSI…”
The terms “incidence” and “burden” currently encompass several distinct outcomes. A socioeconomic factor may increase the overall risk of bloodstream infection without increasing the probability that the infecting organism is resistant. Conversely, a factor may be associated with resistance among patients who already have BSI without materially changing the population incidence of BSI.
The review should therefore separate at least three questions:

Population-level question: Among people living in eligible countries, is socioeconomic or social disadvantage associated with the incidence rate of laboratory-confirmed AMR BSI?
Resistance-conditional question: Among patients with laboratory-confirmed BSI, is socioeconomic or social disadvantage associated with the probability that the causative organism is antimicrobial resistant?
Prognostic question: Among patients with established AMR BSI, is socioeconomic or social disadvantage associated with mortality, morbidity, length of stay or other clinical outcomes?

These should not be combined in one meta-analysis because their denominators are respectively the general population, all BSI episodes, and patients with AMR BSI.
Antimicrobial prescribing should be treated separately as a possible exposure, mediator or contextual outcome rather than being incorporated into the definition of AMR BSI burden.
Page 5, Rationale, references 34–36:
The protocol cites three recent systematic or scoping reviews addressing social determinants, health inequalities and AMR in high-income settings. The authors state that none focuses specifically on AMR BSI, which may justify the present review, but the degree of overlap is not described.
A brief comparison table would strengthen the rationale, covering:

population;
infection type;
AMR outcome;
socioeconomic exposures;
geographical scope;
search dates;
and number of relevant BSI studies identified.

This is particularly important because the authors of the current protocol overlap with authors of some of the cited reviews. The manuscript should demonstrate clearly that the present review addresses a distinct question rather than subdividing an existing evidence base without sufficient clinical justification.
The protocol should also explain how primary studies and surveillance datasets already identified in those reviews will be used—for example, for citation chasing—without treating the reviews themselves as primary evidence.
Page 5, Secondary objectives:
“reported interventions to mitigate against AMR BSIs including preventative measures such as vaccination and healthcare infection, prevention and control measures”
An intervention that reduces AMR BSI overall does not necessarily reduce inequality. To address the stated equity question, an intervention study would need to report:

differential effectiveness across socioeconomic groups;
an interaction between intervention effect and a social stratifier;
or a change in an absolute or relative inequality measure.

Otherwise, the intervention component becomes a separate review of AMR BSI prevention.
The authors should either remove this secondary objective or restrict it to studies that evaluate whether an intervention changes inequalities in AMR BSI.
Suggested revision:
To identify interventions evaluated for their effect on absolute or relative socioeconomic inequalities in AMR BSI incidence or outcomes.
Pages 6–7 “Socioeconomically disadvantaged populations in HICs with AMR BSI.”
“The primary outcome of interest…is incidence/burden of AMR BSI.”
The protocol does not define an eligible AMR BSI episode. Given the absence of publication-date restrictions and the expected diversity of surveillance systems, definitions are likely to vary considerably.
The protocol should specify:

Whether BSI must be confirmed by blood culture
Whether molecular detection without culture is eligible
How common blood-culture contaminants will be handled
Whether resistance must be phenotypically confirmed or whether resistance genes alone are sufficient
Whether “intermediate” susceptibility will be classified as resistant
Whether study-defined resistance is acceptable when CLSI or EUCAST criteria are not reported
How recurrent isolates from the same patient will be handled
Minimum interval defining a new BSI episode
How polymicrobial BSI will be classified, and
Whether community-onset and healthcare-onset BSI will be analysed separately.

“Sepsis” should not independently satisfy the BSI eligibility criterion because sepsis may occur without bloodstream infection, while bacteraemia may occur without sepsis.
A possible eligibility statement would be:
Studies will be eligible when they report microbiologically confirmed bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance and provide data stratified by, or analytically associated with, at least one prespecified social or socioeconomic factor.
Page 7 “Exposures will include factors…such as healthcare access, antibiotic misuse, disability and comorbidity…”
This list combines socioeconomic exposures, structural determinants, clinical risk factors, potential mediators and possible confounders. For example:

Socioeconomic deprivation may be the principal exposure
Healthcare access and antimicrobial use may lie on the causal pathway
Age and comorbidity may confound or modify the association;
Ethnicity may represent a social stratifier rather than a biological exposure;
Disability may be an exposure, a mediator, or a consequence, depending on the analysis.

The protocol would benefit from a conceptual framework or directed acyclic graph identifying the principal exposure, outcome, confounders, mediators and effect modifiers. Without this, the review may pool adjusted estimates that answer different causal questions.
This is particularly relevant to prior antimicrobial exposure. If the intended question is the total association between socioeconomic disadvantage and AMR BSI, adjusting for antimicrobial use may constitute overadjustment when antimicrobial use is part of the causal pathway. Conversely, if the authors seek a direct effect independent of antimicrobial exposure, that should be stated explicitly.
The authors should prespecify which adjusted estimate will be extracted when a study reports several models and should record the covariates included in each model.
Page 6, Table 1: “bloodstream infections OR bacteraemia OR sepsis OR Escherichia coli OR methicillin-resistant Staphylococcus aureus…”
The protocol does not state whether all bacterial pathogens and resistance phenotypes will be considered together. The epidemiology and social distribution of community-onset ESBL-producing E. coli, MRSA bacteraemia, carbapenem-resistant Enterobacterales, VRE, resistant Pseudomonas aeruginosa and resistant Acinetobacter baumannii are unlikely to be directly comparable.
The authors should prespecify clinically meaningful organism–resistance categories, for example:

third-generation cephalosporin-resistant or ESBL-producing Enterobacterales;
carbapenem-resistant Enterobacterales;
methicillin-resistant Staphylococcus aureus;
vancomycin-resistant enterococci;
carbapenem-resistant P. aeruginosa;
and carbapenem-resistant A. baumannii.

An overall pooled “AMR BSI” estimate should be avoided unless the authors can demonstrate that the pathogens, resistance definitions and epidemiological contexts are sufficiently comparable.
The title uses “antimicrobial-resistant,” but the search strategy and rationale appear to concern bacterial antibiotic resistance. The authors should clarify whether fungal bloodstream infections with antifungal resistance are eligible. If they are not, “antibiotic-resistant bacterial bloodstream infections” may be more precise.
The table provides only selected examples. It does not include a complete line-by-line strategy for any database. The extended-data section indicates that the OSF repository contains the PRISMA-P checklist, but it does not state that the full database strategies are available.
PRISMA-P requires a draft search strategy for at least one electronic database to be reported in sufficient detail for replication.

Page 6, “There will be no publication date…restrictions.”
Resistance classification has changed over time because susceptibility breakpoints, laboratory methods and definitions such as MDR, XDR, ESBL and carbapenem resistance have evolved. A microorganism categorised as susceptible under an older breakpoint may be classified differently under a later standard.
The data-extraction form should therefore include:

susceptibility-testing method;
CLSI, EUCAST or other interpretive standard;
version or year of the breakpoint;
whether “resistant” or “non-susceptible” was reported;
and the study’s definition of multidrug resistance.

Studies that do not provide a reproducible definition of resistance should not be combined quantitatively with those that use recognised standards. Sensitivity analysis by breakpoint era or resistance-definition quality may be required.
This is particularly important because unrestricted publication

Page 6, Information sources:
“Dissertations and theses will be searched using Core, Researchgate and Academia databases.”
ResearchGate and Academia.edu are not structured bibliographic databases with stable and fully reproducible search functions. The authors should avoid referring to them simply as databases and provide the exact procedure used, including: search terms, search date, account or access restrictions, sorting method, number of results screened and any stopping rule.
Page 7, Comparator(s):
“Populations not experiencing socioeconomic disadvantage within the same HICs.”
Many relevant studies will report deprivation deciles or quintiles, a continuous deprivation score, household income categories or educational gradients rather than a binary disadvantaged/non-disadvantaged comparison.
The comparator criterion should therefore permit:

comparisons across ordered socioeconomic categories
continuous associations
lowest-versus-highest category comparisons, and
within-study social gradients.

The authors should prespecify how differently scaled deprivation measures will be harmonised. Area-level indices such as the English Index of Multiple Deprivation, the US Area Deprivation Index and the Australian SEIFA indices are country-specific relative measures and should not be treated as directly equivalent absolute exposures.
Where data permit, the authors could distinguish:

extreme-group comparisons
effect per unit or standard-deviation increase
relative index of inequality, and
slope index of inequality.

In Abstract, it stated that
“Thirteen reviewers will complete screening…”
And here at Page 8 Initial title and abstract screening:
“Ten reviewers will be involved in the study selection process…”
Figure 2 also refers more generally to “all reviewers.”
The exact number of human reviewers participating in title and abstract screening, full-text screening, data extraction and risk-of-bias assessment should be reported consistently. If different numbers are involved at different stages, this should be clearly stated rather than using one overall number.

Page 8, Use of a large language model for screening:
“An initial set of 100 abstracts will be independently reviewed…”. “a recall threshold of 95% is set as the primary target.”
Recall is calculated among genuinely eligible records. Therefore, the precision of the recall estimate depends on the number of eligible abstracts in the validation set, not merely on a total sample of 100 abstracts.
If only five or ten of the 100 abstracts are eligible, an observed recall of 100% would still have considerable statistical uncertainty. If the authors intend to demonstrate, rather than merely observe, recall of at least 95%, approximately 59 genuinely eligible abstracts with zero false-negative classifications would be needed for a one-sided 95% lower confidence bound close to 95%.
The validation sample should therefore be enriched with known eligible and near-eligible records while retaining a representative random component. The authors should report:

the number of eligible records in each dataset;
recall with a confidence interval;
the number and characteristics of false negatives;
precision and specificity;
and performance for difficult exclusion categories.

The clinically important issue is not only the overall recall but whether the model disproportionately misses studies with unusual terminology, non-English abstracts, ecological designs or non-standard resistance definitions.
“Across successive iterations, the LLM’s outputs will be compared to the majority expert labels…”
The text suggests that the same 100 abstracts may be used repeatedly for prompt refinement and performance assessment. If the prompt is optimised against this dataset, performance measured on the same records will be optimistic.
Figure 2 appears to show an “additional set” for final evaluation, but the corresponding text does not describe this independent set adequately.
The authors should clearly separate:

a development dataset used for prompt refinement
an independent, locked validation dataset used only once after the prompt is finalised, and
the complete dataset to which the validated procedure will be applied.

Page 8, “The Llama 3.3 70b model will be used…”
“A free-text prompt will be drafted…”
The protocol should also clarify the screening logic. It appears that all human-included records will progress to full text and that LLM-included records will be added even when the human reviewer excluded them. This important safety rule should be stated explicitly.
My suggestion_
Protocol should report:

Exact model identifier and version
Complete final prompt
System instructions (if any)
Temperature and other sampling parameters
How non-binary, missing or malformed responses will be handled
Date on which the model was run
Whether the internal UKHSA tool applies any preprocessing or post-processing.

Page 9, “ROBINS-I will be used for cohort studies…”
ROBINS-I is intended for non-randomised studies evaluating interventions, whereas most studies in this review will evaluate the effects of socioeconomic or social exposures. The current ROBINS-E framework is specifically designed for non-randomised studies of exposure effects.
A single tool should not be assigned solely based on a “cohort” label. The appropriate tool depends on the study question and result being evaluated.
The authors should provide a design-to-tool matrix for:

cohort studies of exposure effects;
case-control studies;
analytical cross-sectional studies;
prevalence studies;
ecological or spatial studies;
surveillance reports;
randomised intervention studies;
and mixed-method studies.

A further difficulty is that national surveillance reports may not be adequately assessed by NOS, ROBINS-E or the JBI prevalence checklist alone. Important surveillance-specific domains include:

completeness of laboratory participation;
consistency of blood-culture sampling;
episode deduplication;
changes in resistance definitions;
completeness of population denominators;
linkage success for socioeconomic variables;
and differential missingness.

These domains should be incorporated explicitly.
Page 10, “PROGRESS domains include Place of residence…Race/ethnicity…Gender/sex, Religion, Education, Socio-economic status…”
The title and population criteria imply that an eligible study must include a measure of socioeconomic disadvantage. However, the Introduction and proposed subgroup analyses also include ethnicity, age, sex, disability, migration status, religion and geographical region.
The eligibility implications are unclear. For example, would a study comparing AMR BSI rates between ethnic groups, but reporting no income, education, occupation or deprivation measure, be eligible?
The authors should adopt one of two approaches:

retain the socioeconomic focus and require every included study to report an individual-level or area-level socioeconomic indicator; or
broaden the review to social inequalities and revise the title, objectives and eligibility criteria accordingly.

If the broader approach is selected, a more accurate title may be:
Social and socioeconomic inequalities in antimicrobial-resistant bloodstream infections in high-income countries: a systematic review protocol.
Age, sex and comorbidity should not be described automatically as indicators of socioeconomic disadvantage. They may be examined as stratifiers or effect modifiers, but they should be conceptually separated from socioeconomic position.

Is the rationale for, and objectives of, the study clearly described?

Partly
Is the study design appropriate for the research question?

Partly
Are sufficient details of the methods provided to allow replication by others?

Partly
Are the datasets clearly presented in a useable and accessible format?

Not applicable

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Clinical Microbiology; Antimicrobial Resistance; Antimicrobial Stewardship; Healthcare-Associated Infections; Bloodstream Infections; Infectious Disease Epidemiology; Bacterial Pathogenesis; Systematic Reviews and Meta-analysis.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Author Response 28 Oct 2026

Diane Ashiru-Oredope, UK Health Security Agency, London, UK

28 Oct 2026

Author Response

We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title ... Continue reading We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title and objective use terms such as “determinants” and “drivers” which imply causality that observational studies cannot establish.
We agree that the review is expected to synthesise predominantly observational evidence and therefore cannot establish causal relationships. The review aims to identify factors associated with observed inequalities rather than determine causal effects.
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.
“Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”

Page 2 (Abstract)
Wording such as “drivers and risk factors contributing to disparities” implies causal effects.
We agree and have revised wording throughout the manuscript to reflect associations rather than causation.
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants.
“The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”

Page 2 (Abstract)
Reviewer numbers are inconsistently reported.
We thank the reviewer for identifying this inconsistency. Reviewer availability continues to evolve during protocol implementation.
Reviewer numbers removed and replaced with descriptions of review-team procedures and quality assurance processes.

“A team of reviewers will complete screening of allocated abstracts and titles for eligibility...”

Page 4 (Introduction)
Distinguish deaths associated with AMR from deaths attributable to AMR.
We thank the reviewer for this clarification and agree.
Introduction revised to distinguish AMR-associated and AMR-attributable mortality.
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs for antimicrobial resistant bacterial infections are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3”

Page 5 (Rationale)
Existing reviews overlap with the review topic and the rationale requires clarification.
We thank the reviewer for highlighting this issue. Existing reviews have examined AMR broadly, resistant infections more generally and socioeconomic influences on AMR, but none focus specifically on AMR BSI or synthesise factors associated with inequalities in AMR BSI and related outcomes.
Additional rationale text added to clarify the unique focus of the current review.

“While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”

Page 5-6 (Rationale)
Previous reviews should be used for citation chasing rather than as primary evidence.
We agree. Existing reviews are used to establish rationale and identify potentially relevant primary studies.
Reviews will not be treated as primary evidence during synthesis.

Page 6-7 (Objectives)
Intervention objective does not necessarily address inequalities.
We agree and have refined the objective accordingly.
Intervention objective revised to focus specifically on interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.

“...interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.”

Page 6 (Objectives)
Incidence, resistance among BSI cases and outcomes following AMR BSI represent different epidemiological questions.
We agree these are distinct outcome domains and should not automatically be combined.
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate.

“To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”

Page 7 (Information Sources)
ResearchGate and Academia.edu are not bibliographic databases.
We agree and have revised the wording accordingly.
ResearchGate and Academia.edu are now described as supplementary grey-literature sources rather than bibliographic databases.

" Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases.”

Page 7-8 (Search Strategy)
Full database search strategy is not provided.
We agree transparency and reproducibility are important. Full search strategies have been developed and executed and will be reported alongside the completed systematic review.
No change: Summary search concepts are reported in the protocol. Full search strategies will accompany the completed review publication.

Pages 8 (Eligibility & Population)
Eligibility implications are unclear. It is not clear whether studies must include a socioeconomic measure or whether studies reporting PROGRESS characteristics alone would be eligible.
We thank the reviewer for highlighting this issue. The review adopts an equity-focused approach informed by the PROGRESS-Plus framework. Relevant PROGRESS-Plus characteristics may be used as indicators of populations experiencing socioeconomic disadvantage and health inequalities.
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.

“All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications where examined in relation to inequalities in AMR BSI and related outcomes.”

Page 8–9 (Eligibility & outcomes)
Eligible AMR BSI episodes require clearer definition.
We agree greater clarity is beneficial.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance.

“The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance.

Page 9 (Eligibility & outcomes)
Additional detail requested regarding contaminants, recurrence, polymicrobial infection, episode definitions and deduplication.
We appreciate these recommendations. Given the anticipated heterogeneity of surveillance systems and study designs, the review will accept study-defined BSI and resistance definitions and extract these where reported.
Definitions used by study authors will be extracted and considered during interpretation. No additional surveillance-specific eligibility criteria introduced.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Page 9 (Eligibility & outcomes)
Clarify whether fungal bloodstream infections are eligible.
We agree clarification is important.
Eligibility criteria revised to clarify that the review focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

“This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.”

Page 9 (Eligibiliy & Exposures)
Exposure framework combines exposures, mediators, confounders and effect modifiers.
We acknowledge this issue. The review is intended to identify factors associated with inequalities rather than estimate causal effects.
Protocol revised to explicitly state that it does not seek to estimate causal effects. Study-specific analytical approaches and adjustment variables will be extracted and reported.

“The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.”

Page 9 (Exposure Section)
A conceptual framework or DAG should be developed.
We acknowledge the value of causal frameworks when estimating causal effects. However, this review is descriptive and exploratory rather than causal in nature.
No change. Clarification added regarding the purpose of the review (see above). The inability to estimate causal effects due to the heterogeneity of available studies will be included in the discussion on limitations.

Page 9 (Eligibility & Comparator Section)
Comparator should allow gradients and continuous measures.
We agree.
Comparator section revised to include deprivation gradients, ordered categories and continuous measures of socioeconomic disadvantage.

“Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.”

Page 9 (Eligibility & Comparator Section)
Socioeconomic measures across studies should be harmonised.
We agree that socioeconomic measures vary substantially across countries and settings. Given the expected heterogeneity, narrative synthesis is anticipated.
Socioeconomic measures will generally be reported as defined by study authors and interpreted narratively.

“Comparators will be reported as defined by study authors.”

Page 9 (Eligibility & Outcomes)
Prescribing-only studies should not be eligible unless linked to AMR BSI. Antimicrobial prescribing should be treated separately from AMR BSI burden.
We agree that the review should remain focused on AMR BSI and related outcomes.
Antimicrobial prescribing is treated as a healthcare utilisation outcome or explanatory factor relevant to AMR BSI inequalities rather than a standalone primary outcome.

“Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.”

Page 9 (Eligibility & Outcomes)
No publication-date restrictions may result in heterogeneous resistance definitions.

We agree these factors may contribute to heterogeneity. However, extensive harmonisation may not be feasible given the anticipated diversity of included studies.

Data extraction expanded to capture resistance definitions used by study authors. These factors will be considered during interpretation of findings. No commitment made to breakpoint-era sensitivity analyses.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Pages 10-11 (Screening)
LLM validation sample may be insufficient for robust recall estimation.
We acknowledge this limitation. The LLM component is exploratory and intended to support methodological evaluation rather than replace human review.
No change: Human reviewers remain responsible for study selection decisions. No substantive protocol amendment required.

Pages 10-11 (Screening)
Development and validation datasets should be separated.
We agree that development and evaluation are conceptually distinct.
Clarified that prompt refinement and performance evaluation occur as separate stages within the LLM evaluation process.

“Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.”

Pages 10-11 (Screening)
More detail should be reported regarding prompts, model parameters and implementation settings.
We agree transparency is important. However, detailed evaluation of LLM performance forms part of a separate methodological programme of work and will be reported independently.
Protocol specifies the model used and overall screening procedure. Additional technical evaluation will be reported separately. (see above)

Pages 10-11 (Screening)
LLM may miss unusual study types, non-English studies or atypical terminology.
We agree this is an important consideration. However, the LLM functions as a supplementary rather than replacement screening method.
Human review remains the primary study-selection process. Further methodological evaluation will be reported separately.(see above)

Pages 10-11 (Screening)
Screening logic should explicitly state that AI-included studies progress to full-text screening.
We agree and thank the reviewer for this suggestion.
Protocol revised to clarify that all records identified for inclusion by the AI tool progress to full-text review.

“An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening”

Page 12 (Risk of Bias)
ROBINS-I may not be appropriate for exposure studies.
We thank the reviewer for this suggestion. We acknowledge the availability of ROBINS-E and will consider the most appropriate risk-of-bias approach based on included study designs.
No change: Any deviations from protocol will be transparently reported in the final review.

Page 12 (Risk of Bias)
Surveillance studies may require surveillance-specific risk-of-bias domains.
We agree surveillance studies have unique methodological considerations.
No change: No formal surveillance-specific framework introduced. These issues will be considered during interpretation of findings.

Page 12 (Plan for Data Synthesis)
Different pathogens and resistance phenotypes should not automatically be pooled. The protocol places too much emphasis on quantitative synthesis.
We agree substantial heterogeneity is anticipated across populations, socioeconomic measures, pathogens, resistance definitions and outcomes.
Studies will be grouped according to pathogen and resistance phenotype where appropriate. Meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists.

“Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity (including pathogens, resistance phenotypes) exist. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.”

Page 12 (Plan for Data Synthesis)
Detailed publication-bias, subgroup and heterogeneity procedures imply a complex meta-analysis.
We acknowledge this concern. The review is expected to rely primarily on narrative synthesis, although quantitative synthesis remains possible where appropriate.
Meta-analysis methods have been retained but reframed as contingent on sufficient methodological and clinical homogeneity rather than routine analyses (see above).
We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title and objective use terms such as “determinants” and “drivers” which imply causality that observational studies cannot establish.
We agree that the review is expected to synthesise predominantly observational evidence and therefore cannot establish causal relationships. The review aims to identify factors associated with observed inequalities rather than determine causal effects.
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.
“Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”

Page 2 (Abstract)
Wording such as “drivers and risk factors contributing to disparities” implies causal effects.
We agree and have revised wording throughout the manuscript to reflect associations rather than causation.
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants.
“The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”

Page 2 (Abstract)
Reviewer numbers are inconsistently reported.
We thank the reviewer for identifying this inconsistency. Reviewer availability continues to evolve during protocol implementation.
Reviewer numbers removed and replaced with descriptions of review-team procedures and quality assurance processes.

“A team of reviewers will complete screening of allocated abstracts and titles for eligibility...”

Page 4 (Introduction)
Distinguish deaths associated with AMR from deaths attributable to AMR.
We thank the reviewer for this clarification and agree.
Introduction revised to distinguish AMR-associated and AMR-attributable mortality.
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs for antimicrobial resistant bacterial infections are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3”

Page 5 (Rationale)
Existing reviews overlap with the review topic and the rationale requires clarification.
We thank the reviewer for highlighting this issue. Existing reviews have examined AMR broadly, resistant infections more generally and socioeconomic influences on AMR, but none focus specifically on AMR BSI or synthesise factors associated with inequalities in AMR BSI and related outcomes.
Additional rationale text added to clarify the unique focus of the current review.

“While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”

Page 5-6 (Rationale)
Previous reviews should be used for citation chasing rather than as primary evidence.
We agree. Existing reviews are used to establish rationale and identify potentially relevant primary studies.
Reviews will not be treated as primary evidence during synthesis.

Page 6-7 (Objectives)
Intervention objective does not necessarily address inequalities.
We agree and have refined the objective accordingly.
Intervention objective revised to focus specifically on interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.

“...interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.”

Page 6 (Objectives)
Incidence, resistance among BSI cases and outcomes following AMR BSI represent different epidemiological questions.
We agree these are distinct outcome domains and should not automatically be combined.
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate.

“To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”

Page 7 (Information Sources)
ResearchGate and Academia.edu are not bibliographic databases.
We agree and have revised the wording accordingly.
ResearchGate and Academia.edu are now described as supplementary grey-literature sources rather than bibliographic databases.

" Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases.”

Page 7-8 (Search Strategy)
Full database search strategy is not provided.
We agree transparency and reproducibility are important. Full search strategies have been developed and executed and will be reported alongside the completed systematic review.
No change: Summary search concepts are reported in the protocol. Full search strategies will accompany the completed review publication.

Pages 8 (Eligibility & Population)
Eligibility implications are unclear. It is not clear whether studies must include a socioeconomic measure or whether studies reporting PROGRESS characteristics alone would be eligible.
We thank the reviewer for highlighting this issue. The review adopts an equity-focused approach informed by the PROGRESS-Plus framework. Relevant PROGRESS-Plus characteristics may be used as indicators of populations experiencing socioeconomic disadvantage and health inequalities.
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.

“All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications where examined in relation to inequalities in AMR BSI and related outcomes.”

Page 8–9 (Eligibility & outcomes)
Eligible AMR BSI episodes require clearer definition.
We agree greater clarity is beneficial.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance.

“The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance.

Page 9 (Eligibility & outcomes)
Additional detail requested regarding contaminants, recurrence, polymicrobial infection, episode definitions and deduplication.
We appreciate these recommendations. Given the anticipated heterogeneity of surveillance systems and study designs, the review will accept study-defined BSI and resistance definitions and extract these where reported.
Definitions used by study authors will be extracted and considered during interpretation. No additional surveillance-specific eligibility criteria introduced.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Page 9 (Eligibility & outcomes)
Clarify whether fungal bloodstream infections are eligible.
We agree clarification is important.
Eligibility criteria revised to clarify that the review focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

“This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.”

Page 9 (Eligibiliy & Exposures)
Exposure framework combines exposures, mediators, confounders and effect modifiers.
We acknowledge this issue. The review is intended to identify factors associated with inequalities rather than estimate causal effects.
Protocol revised to explicitly state that it does not seek to estimate causal effects. Study-specific analytical approaches and adjustment variables will be extracted and reported.

“The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.”

Page 9 (Exposure Section)
A conceptual framework or DAG should be developed.
We acknowledge the value of causal frameworks when estimating causal effects. However, this review is descriptive and exploratory rather than causal in nature.
No change. Clarification added regarding the purpose of the review (see above). The inability to estimate causal effects due to the heterogeneity of available studies will be included in the discussion on limitations.

Page 9 (Eligibility & Comparator Section)
Comparator should allow gradients and continuous measures.
We agree.
Comparator section revised to include deprivation gradients, ordered categories and continuous measures of socioeconomic disadvantage.

“Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.”

Page 9 (Eligibility & Comparator Section)
Socioeconomic measures across studies should be harmonised.
We agree that socioeconomic measures vary substantially across countries and settings. Given the expected heterogeneity, narrative synthesis is anticipated.
Socioeconomic measures will generally be reported as defined by study authors and interpreted narratively.

“Comparators will be reported as defined by study authors.”

Page 9 (Eligibility & Outcomes)
Prescribing-only studies should not be eligible unless linked to AMR BSI. Antimicrobial prescribing should be treated separately from AMR BSI burden.
We agree that the review should remain focused on AMR BSI and related outcomes.
Antimicrobial prescribing is treated as a healthcare utilisation outcome or explanatory factor relevant to AMR BSI inequalities rather than a standalone primary outcome.

“Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.”

Page 9 (Eligibility & Outcomes)
No publication-date restrictions may result in heterogeneous resistance definitions.

We agree these factors may contribute to heterogeneity. However, extensive harmonisation may not be feasible given the anticipated diversity of included studies.

Data extraction expanded to capture resistance definitions used by study authors. These factors will be considered during interpretation of findings. No commitment made to breakpoint-era sensitivity analyses.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Pages 10-11 (Screening)
LLM validation sample may be insufficient for robust recall estimation.
We acknowledge this limitation. The LLM component is exploratory and intended to support methodological evaluation rather than replace human review.
No change: Human reviewers remain responsible for study selection decisions. No substantive protocol amendment required.

Pages 10-11 (Screening)
Development and validation datasets should be separated.
We agree that development and evaluation are conceptually distinct.
Clarified that prompt refinement and performance evaluation occur as separate stages within the LLM evaluation process.

“Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.”

Pages 10-11 (Screening)
More detail should be reported regarding prompts, model parameters and implementation settings.
We agree transparency is important. However, detailed evaluation of LLM performance forms part of a separate methodological programme of work and will be reported independently.
Protocol specifies the model used and overall screening procedure. Additional technical evaluation will be reported separately. (see above)

Pages 10-11 (Screening)
LLM may miss unusual study types, non-English studies or atypical terminology.
We agree this is an important consideration. However, the LLM functions as a supplementary rather than replacement screening method.
Human review remains the primary study-selection process. Further methodological evaluation will be reported separately.(see above)

Pages 10-11 (Screening)
Screening logic should explicitly state that AI-included studies progress to full-text screening.
We agree and thank the reviewer for this suggestion.
Protocol revised to clarify that all records identified for inclusion by the AI tool progress to full-text review.

“An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening”

Page 12 (Risk of Bias)
ROBINS-I may not be appropriate for exposure studies.
We thank the reviewer for this suggestion. We acknowledge the availability of ROBINS-E and will consider the most appropriate risk-of-bias approach based on included study designs.
No change: Any deviations from protocol will be transparently reported in the final review.

Page 12 (Risk of Bias)
Surveillance studies may require surveillance-specific risk-of-bias domains.
We agree surveillance studies have unique methodological considerations.
No change: No formal surveillance-specific framework introduced. These issues will be considered during interpretation of findings.

Page 12 (Plan for Data Synthesis)
Different pathogens and resistance phenotypes should not automatically be pooled. The protocol places too much emphasis on quantitative synthesis.
We agree substantial heterogeneity is anticipated across populations, socioeconomic measures, pathogens, resistance definitions and outcomes.
Studies will be grouped according to pathogen and resistance phenotype where appropriate. Meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists.

“Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity (including pathogens, resistance phenotypes) exist. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.”

Page 12 (Plan for Data Synthesis)
Detailed publication-bias, subgroup and heterogeneity procedures imply a complex meta-analysis.
We acknowledge this concern. The review is expected to rely primarily on narrative synthesis, although quantitative synthesis remains possible where appropriate.
Meta-analysis methods have been retained but reframed as contingent on sufficient methodological and clinical homogeneity rather than routine analyses (see above).
Competing Interests: None Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 28 Oct 2026

Diane Ashiru-Oredope, UK Health Security Agency, London, UK

28 Oct 2026

Author Response

We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title ... Continue reading We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title and objective use terms such as “determinants” and “drivers” which imply causality that observational studies cannot establish.
We agree that the review is expected to synthesise predominantly observational evidence and therefore cannot establish causal relationships. The review aims to identify factors associated with observed inequalities rather than determine causal effects.
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.
“Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”

Page 2 (Abstract)
Wording such as “drivers and risk factors contributing to disparities” implies causal effects.
We agree and have revised wording throughout the manuscript to reflect associations rather than causation.
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants.
“The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”

Page 2 (Abstract)
Reviewer numbers are inconsistently reported.
We thank the reviewer for identifying this inconsistency. Reviewer availability continues to evolve during protocol implementation.
Reviewer numbers removed and replaced with descriptions of review-team procedures and quality assurance processes.

“A team of reviewers will complete screening of allocated abstracts and titles for eligibility...”

Page 4 (Introduction)
Distinguish deaths associated with AMR from deaths attributable to AMR.
We thank the reviewer for this clarification and agree.
Introduction revised to distinguish AMR-associated and AMR-attributable mortality.
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs for antimicrobial resistant bacterial infections are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3”

Page 5 (Rationale)
Existing reviews overlap with the review topic and the rationale requires clarification.
We thank the reviewer for highlighting this issue. Existing reviews have examined AMR broadly, resistant infections more generally and socioeconomic influences on AMR, but none focus specifically on AMR BSI or synthesise factors associated with inequalities in AMR BSI and related outcomes.
Additional rationale text added to clarify the unique focus of the current review.

“While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”

Page 5-6 (Rationale)
Previous reviews should be used for citation chasing rather than as primary evidence.
We agree. Existing reviews are used to establish rationale and identify potentially relevant primary studies.
Reviews will not be treated as primary evidence during synthesis.

Page 6-7 (Objectives)
Intervention objective does not necessarily address inequalities.
We agree and have refined the objective accordingly.
Intervention objective revised to focus specifically on interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.

“...interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.”

Page 6 (Objectives)
Incidence, resistance among BSI cases and outcomes following AMR BSI represent different epidemiological questions.
We agree these are distinct outcome domains and should not automatically be combined.
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate.

“To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”

Page 7 (Information Sources)
ResearchGate and Academia.edu are not bibliographic databases.
We agree and have revised the wording accordingly.
ResearchGate and Academia.edu are now described as supplementary grey-literature sources rather than bibliographic databases.

" Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases.”

Page 7-8 (Search Strategy)
Full database search strategy is not provided.
We agree transparency and reproducibility are important. Full search strategies have been developed and executed and will be reported alongside the completed systematic review.
No change: Summary search concepts are reported in the protocol. Full search strategies will accompany the completed review publication.

Pages 8 (Eligibility & Population)
Eligibility implications are unclear. It is not clear whether studies must include a socioeconomic measure or whether studies reporting PROGRESS characteristics alone would be eligible.
We thank the reviewer for highlighting this issue. The review adopts an equity-focused approach informed by the PROGRESS-Plus framework. Relevant PROGRESS-Plus characteristics may be used as indicators of populations experiencing socioeconomic disadvantage and health inequalities.
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.

“All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications where examined in relation to inequalities in AMR BSI and related outcomes.”

Page 8–9 (Eligibility & outcomes)
Eligible AMR BSI episodes require clearer definition.
We agree greater clarity is beneficial.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance.

“The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance.

Page 9 (Eligibility & outcomes)
Additional detail requested regarding contaminants, recurrence, polymicrobial infection, episode definitions and deduplication.
We appreciate these recommendations. Given the anticipated heterogeneity of surveillance systems and study designs, the review will accept study-defined BSI and resistance definitions and extract these where reported.
Definitions used by study authors will be extracted and considered during interpretation. No additional surveillance-specific eligibility criteria introduced.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Page 9 (Eligibility & outcomes)
Clarify whether fungal bloodstream infections are eligible.
We agree clarification is important.
Eligibility criteria revised to clarify that the review focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

“This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.”

Page 9 (Eligibiliy & Exposures)
Exposure framework combines exposures, mediators, confounders and effect modifiers.
We acknowledge this issue. The review is intended to identify factors associated with inequalities rather than estimate causal effects.
Protocol revised to explicitly state that it does not seek to estimate causal effects. Study-specific analytical approaches and adjustment variables will be extracted and reported.

“The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.”

Page 9 (Exposure Section)
A conceptual framework or DAG should be developed.
We acknowledge the value of causal frameworks when estimating causal effects. However, this review is descriptive and exploratory rather than causal in nature.
No change. Clarification added regarding the purpose of the review (see above). The inability to estimate causal effects due to the heterogeneity of available studies will be included in the discussion on limitations.

Page 9 (Eligibility & Comparator Section)
Comparator should allow gradients and continuous measures.
We agree.
Comparator section revised to include deprivation gradients, ordered categories and continuous measures of socioeconomic disadvantage.

“Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.”

Page 9 (Eligibility & Comparator Section)
Socioeconomic measures across studies should be harmonised.
We agree that socioeconomic measures vary substantially across countries and settings. Given the expected heterogeneity, narrative synthesis is anticipated.
Socioeconomic measures will generally be reported as defined by study authors and interpreted narratively.

“Comparators will be reported as defined by study authors.”

Page 9 (Eligibility & Outcomes)
Prescribing-only studies should not be eligible unless linked to AMR BSI. Antimicrobial prescribing should be treated separately from AMR BSI burden.
We agree that the review should remain focused on AMR BSI and related outcomes.
Antimicrobial prescribing is treated as a healthcare utilisation outcome or explanatory factor relevant to AMR BSI inequalities rather than a standalone primary outcome.

“Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.”

Page 9 (Eligibility & Outcomes)
No publication-date restrictions may result in heterogeneous resistance definitions.

We agree these factors may contribute to heterogeneity. However, extensive harmonisation may not be feasible given the anticipated diversity of included studies.

Data extraction expanded to capture resistance definitions used by study authors. These factors will be considered during interpretation of findings. No commitment made to breakpoint-era sensitivity analyses.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Pages 10-11 (Screening)
LLM validation sample may be insufficient for robust recall estimation.
We acknowledge this limitation. The LLM component is exploratory and intended to support methodological evaluation rather than replace human review.
No change: Human reviewers remain responsible for study selection decisions. No substantive protocol amendment required.

Pages 10-11 (Screening)
Development and validation datasets should be separated.
We agree that development and evaluation are conceptually distinct.
Clarified that prompt refinement and performance evaluation occur as separate stages within the LLM evaluation process.

“Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.”

Pages 10-11 (Screening)
More detail should be reported regarding prompts, model parameters and implementation settings.
We agree transparency is important. However, detailed evaluation of LLM performance forms part of a separate methodological programme of work and will be reported independently.
Protocol specifies the model used and overall screening procedure. Additional technical evaluation will be reported separately. (see above)

Pages 10-11 (Screening)
LLM may miss unusual study types, non-English studies or atypical terminology.
We agree this is an important consideration. However, the LLM functions as a supplementary rather than replacement screening method.
Human review remains the primary study-selection process. Further methodological evaluation will be reported separately.(see above)

Pages 10-11 (Screening)
Screening logic should explicitly state that AI-included studies progress to full-text screening.
We agree and thank the reviewer for this suggestion.
Protocol revised to clarify that all records identified for inclusion by the AI tool progress to full-text review.

“An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening”

Page 12 (Risk of Bias)
ROBINS-I may not be appropriate for exposure studies.
We thank the reviewer for this suggestion. We acknowledge the availability of ROBINS-E and will consider the most appropriate risk-of-bias approach based on included study designs.
No change: Any deviations from protocol will be transparently reported in the final review.

Page 12 (Risk of Bias)
Surveillance studies may require surveillance-specific risk-of-bias domains.
We agree surveillance studies have unique methodological considerations.
No change: No formal surveillance-specific framework introduced. These issues will be considered during interpretation of findings.

Page 12 (Plan for Data Synthesis)
Different pathogens and resistance phenotypes should not automatically be pooled. The protocol places too much emphasis on quantitative synthesis.
We agree substantial heterogeneity is anticipated across populations, socioeconomic measures, pathogens, resistance definitions and outcomes.
Studies will be grouped according to pathogen and resistance phenotype where appropriate. Meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists.

“Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity (including pathogens, resistance phenotypes) exist. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.”

Page 12 (Plan for Data Synthesis)
Detailed publication-bias, subgroup and heterogeneity procedures imply a complex meta-analysis.
We acknowledge this concern. The review is expected to rely primarily on narrative synthesis, although quantitative synthesis remains possible where appropriate.
Meta-analysis methods have been retained but reframed as contingent on sufficient methodological and clinical homogeneity rather than routine analyses (see above).
We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title and objective use terms such as “determinants” and “drivers” which imply causality that observational studies cannot establish.
We agree that the review is expected to synthesise predominantly observational evidence and therefore cannot establish causal relationships. The review aims to identify factors associated with observed inequalities rather than determine causal effects.
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.
“Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”

Page 2 (Abstract)
Wording such as “drivers and risk factors contributing to disparities” implies causal effects.
We agree and have revised wording throughout the manuscript to reflect associations rather than causation.
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants.
“The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”

Page 2 (Abstract)
Reviewer numbers are inconsistently reported.
We thank the reviewer for identifying this inconsistency. Reviewer availability continues to evolve during protocol implementation.
Reviewer numbers removed and replaced with descriptions of review-team procedures and quality assurance processes.

“A team of reviewers will complete screening of allocated abstracts and titles for eligibility...”

Page 4 (Introduction)
Distinguish deaths associated with AMR from deaths attributable to AMR.
We thank the reviewer for this clarification and agree.
Introduction revised to distinguish AMR-associated and AMR-attributable mortality.
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs for antimicrobial resistant bacterial infections are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3”

Page 5 (Rationale)
Existing reviews overlap with the review topic and the rationale requires clarification.
We thank the reviewer for highlighting this issue. Existing reviews have examined AMR broadly, resistant infections more generally and socioeconomic influences on AMR, but none focus specifically on AMR BSI or synthesise factors associated with inequalities in AMR BSI and related outcomes.
Additional rationale text added to clarify the unique focus of the current review.

“While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”

Page 5-6 (Rationale)
Previous reviews should be used for citation chasing rather than as primary evidence.
We agree. Existing reviews are used to establish rationale and identify potentially relevant primary studies.
Reviews will not be treated as primary evidence during synthesis.

Page 6-7 (Objectives)
Intervention objective does not necessarily address inequalities.
We agree and have refined the objective accordingly.
Intervention objective revised to focus specifically on interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.

“...interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.”

Page 6 (Objectives)
Incidence, resistance among BSI cases and outcomes following AMR BSI represent different epidemiological questions.
We agree these are distinct outcome domains and should not automatically be combined.
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate.

“To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”

Page 7 (Information Sources)
ResearchGate and Academia.edu are not bibliographic databases.
We agree and have revised the wording accordingly.
ResearchGate and Academia.edu are now described as supplementary grey-literature sources rather than bibliographic databases.

" Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases.”

Page 7-8 (Search Strategy)
Full database search strategy is not provided.
We agree transparency and reproducibility are important. Full search strategies have been developed and executed and will be reported alongside the completed systematic review.
No change: Summary search concepts are reported in the protocol. Full search strategies will accompany the completed review publication.

Pages 8 (Eligibility & Population)
Eligibility implications are unclear. It is not clear whether studies must include a socioeconomic measure or whether studies reporting PROGRESS characteristics alone would be eligible.
We thank the reviewer for highlighting this issue. The review adopts an equity-focused approach informed by the PROGRESS-Plus framework. Relevant PROGRESS-Plus characteristics may be used as indicators of populations experiencing socioeconomic disadvantage and health inequalities.
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.

“All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications where examined in relation to inequalities in AMR BSI and related outcomes.”

Page 8–9 (Eligibility & outcomes)
Eligible AMR BSI episodes require clearer definition.
We agree greater clarity is beneficial.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance.

“The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance.

Page 9 (Eligibility & outcomes)
Additional detail requested regarding contaminants, recurrence, polymicrobial infection, episode definitions and deduplication.
We appreciate these recommendations. Given the anticipated heterogeneity of surveillance systems and study designs, the review will accept study-defined BSI and resistance definitions and extract these where reported.
Definitions used by study authors will be extracted and considered during interpretation. No additional surveillance-specific eligibility criteria introduced.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Page 9 (Eligibility & outcomes)
Clarify whether fungal bloodstream infections are eligible.
We agree clarification is important.
Eligibility criteria revised to clarify that the review focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

“This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.”

Page 9 (Eligibiliy & Exposures)
Exposure framework combines exposures, mediators, confounders and effect modifiers.
We acknowledge this issue. The review is intended to identify factors associated with inequalities rather than estimate causal effects.
Protocol revised to explicitly state that it does not seek to estimate causal effects. Study-specific analytical approaches and adjustment variables will be extracted and reported.

“The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.”

Page 9 (Exposure Section)
A conceptual framework or DAG should be developed.
We acknowledge the value of causal frameworks when estimating causal effects. However, this review is descriptive and exploratory rather than causal in nature.
No change. Clarification added regarding the purpose of the review (see above). The inability to estimate causal effects due to the heterogeneity of available studies will be included in the discussion on limitations.

Page 9 (Eligibility & Comparator Section)
Comparator should allow gradients and continuous measures.
We agree.
Comparator section revised to include deprivation gradients, ordered categories and continuous measures of socioeconomic disadvantage.

“Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.”

Page 9 (Eligibility & Comparator Section)
Socioeconomic measures across studies should be harmonised.
We agree that socioeconomic measures vary substantially across countries and settings. Given the expected heterogeneity, narrative synthesis is anticipated.
Socioeconomic measures will generally be reported as defined by study authors and interpreted narratively.

“Comparators will be reported as defined by study authors.”

Page 9 (Eligibility & Outcomes)
Prescribing-only studies should not be eligible unless linked to AMR BSI. Antimicrobial prescribing should be treated separately from AMR BSI burden.
We agree that the review should remain focused on AMR BSI and related outcomes.
Antimicrobial prescribing is treated as a healthcare utilisation outcome or explanatory factor relevant to AMR BSI inequalities rather than a standalone primary outcome.

“Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.”

Page 9 (Eligibility & Outcomes)
No publication-date restrictions may result in heterogeneous resistance definitions.

We agree these factors may contribute to heterogeneity. However, extensive harmonisation may not be feasible given the anticipated diversity of included studies.

Data extraction expanded to capture resistance definitions used by study authors. These factors will be considered during interpretation of findings. No commitment made to breakpoint-era sensitivity analyses.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Pages 10-11 (Screening)
LLM validation sample may be insufficient for robust recall estimation.
We acknowledge this limitation. The LLM component is exploratory and intended to support methodological evaluation rather than replace human review.
No change: Human reviewers remain responsible for study selection decisions. No substantive protocol amendment required.

Pages 10-11 (Screening)
Development and validation datasets should be separated.
We agree that development and evaluation are conceptually distinct.
Clarified that prompt refinement and performance evaluation occur as separate stages within the LLM evaluation process.

“Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.”

Pages 10-11 (Screening)
More detail should be reported regarding prompts, model parameters and implementation settings.
We agree transparency is important. However, detailed evaluation of LLM performance forms part of a separate methodological programme of work and will be reported independently.
Protocol specifies the model used and overall screening procedure. Additional technical evaluation will be reported separately. (see above)

Pages 10-11 (Screening)
LLM may miss unusual study types, non-English studies or atypical terminology.
We agree this is an important consideration. However, the LLM functions as a supplementary rather than replacement screening method.
Human review remains the primary study-selection process. Further methodological evaluation will be reported separately.(see above)

Pages 10-11 (Screening)
Screening logic should explicitly state that AI-included studies progress to full-text screening.
We agree and thank the reviewer for this suggestion.
Protocol revised to clarify that all records identified for inclusion by the AI tool progress to full-text review.

“An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening”

Page 12 (Risk of Bias)
ROBINS-I may not be appropriate for exposure studies.
We thank the reviewer for this suggestion. We acknowledge the availability of ROBINS-E and will consider the most appropriate risk-of-bias approach based on included study designs.
No change: Any deviations from protocol will be transparently reported in the final review.

Page 12 (Risk of Bias)
Surveillance studies may require surveillance-specific risk-of-bias domains.
We agree surveillance studies have unique methodological considerations.
No change: No formal surveillance-specific framework introduced. These issues will be considered during interpretation of findings.

Page 12 (Plan for Data Synthesis)
Different pathogens and resistance phenotypes should not automatically be pooled. The protocol places too much emphasis on quantitative synthesis.
We agree substantial heterogeneity is anticipated across populations, socioeconomic measures, pathogens, resistance definitions and outcomes.
Studies will be grouped according to pathogen and resistance phenotype where appropriate. Meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists.

“Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity (including pathogens, resistance phenotypes) exist. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.”

Page 12 (Plan for Data Synthesis)
Detailed publication-bias, subgroup and heterogeneity procedures imply a complex meta-analysis.
We acknowledge this concern. The review is expected to rely primarily on narrative synthesis, although quantitative synthesis remains possible where appropriate.
Meta-analysis methods have been retained but reframed as contingent on sufficient methodological and clinical homogeneity rather than routine analyses (see above).
Competing Interests: None Close
Report a concern

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 13 Jul 2026

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1
Version 2 (revision) 28 Oct 26
Version 1 13 Jul 26	read

Razique Anwer, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia

Comments on this article

All Comments(0)

Add a comment

Back to all reports

Reviewer Report

22 Views

28 Jul 2026 | for Version 1

Razique Anwer, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia

22 Views Cite this report Responses(1)

Approved With Reservations

The distinction between deaths “associated with” and “attributable to” AMR should be retained throughout the manuscript.

Population-level question: Among people living in eligible countries, is socioeconomic or social disadvantage associated with the incidence rate of laboratory-confirmed AMR BSI?
Resistance-conditional question: Among patients with laboratory-confirmed BSI, is socioeconomic or social disadvantage associated with the probability that the causative organism is antimicrobial resistant?
Prognostic question: Among patients with established AMR BSI, is socioeconomic or social disadvantage associated with mortality, morbidity, length of stay or other clinical outcomes?

population;
infection type;
AMR outcome;
socioeconomic exposures;
geographical scope;
search dates;
and number of relevant BSI studies identified.

differential effectiveness across socioeconomic groups;
an interaction between intervention effect and a social stratifier;
or a change in an absolute or relative inequality measure.

Whether BSI must be confirmed by blood culture
Whether molecular detection without culture is eligible
How common blood-culture contaminants will be handled
Whether resistance must be phenotypically confirmed or whether resistance genes alone are sufficient
Whether “intermediate” susceptibility will be classified as resistant
Whether study-defined resistance is acceptable when CLSI or EUCAST criteria are not reported
How recurrent isolates from the same patient will be handled
Minimum interval defining a new BSI episode
How polymicrobial BSI will be classified, and
Whether community-onset and healthcare-onset BSI will be analysed separately.

Socioeconomic deprivation may be the principal exposure
Healthcare access and antimicrobial use may lie on the causal pathway
Age and comorbidity may confound or modify the association;
Ethnicity may represent a social stratifier rather than a biological exposure;
Disability may be an exposure, a mediator, or a consequence, depending on the analysis.

third-generation cephalosporin-resistant or ESBL-producing Enterobacterales;
carbapenem-resistant Enterobacterales;
methicillin-resistant Staphylococcus aureus;
vancomycin-resistant enterococci;
carbapenem-resistant P. aeruginosa;
and carbapenem-resistant A. baumannii.

susceptibility-testing method;
CLSI, EUCAST or other interpretive standard;
version or year of the breakpoint;
whether “resistant” or “non-susceptible” was reported;
and the study’s definition of multidrug resistance.

comparisons across ordered socioeconomic categories
continuous associations
lowest-versus-highest category comparisons, and
within-study social gradients.

extreme-group comparisons
effect per unit or standard-deviation increase
relative index of inequality, and
slope index of inequality.

the number of eligible records in each dataset;
recall with a confidence interval;
the number and characteristics of false negatives;
precision and specificity;
and performance for difficult exclusion categories.

a development dataset used for prompt refinement
an independent, locked validation dataset used only once after the prompt is finalised, and
the complete dataset to which the validated procedure will be applied.

Exact model identifier and version
Complete final prompt
System instructions (if any)
Temperature and other sampling parameters
How non-binary, missing or malformed responses will be handled
Date on which the model was run
Whether the internal UKHSA tool applies any preprocessing or post-processing.

cohort studies of exposure effects;
case-control studies;
analytical cross-sectional studies;
prevalence studies;
ecological or spatial studies;
surveillance reports;
randomised intervention studies;
and mixed-method studies.

A further difficulty is that national surveillance reports may not be adequately assessed by NOS, ROBINS-E or the JBI prevalence checklist alone. Important surveillance-specific domains include:

completeness of laboratory participation;
consistency of blood-culture sampling;
episode deduplication;
changes in resistance definitions;
completeness of population denominators;
linkage success for socioeconomic variables;
and differential missingness.

retain the socioeconomic focus and require every included study to report an individual-level or area-level socioeconomic indicator; or
broaden the review to social inequalities and revise the title, objectives and eligibility criteria accordingly.

Is the rationale for, and objectives of, the study clearly described?

Partly
Is the study design appropriate for the research question?

Partly
Are sufficient details of the methods provided to allow replication by others?

Partly
Are the datasets clearly presented in a useable and accessible format?

Not applicable

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Clinical Microbiology; Antimicrobial Resistance; Antimicrobial Stewardship; Healthcare-Associated Infections; Bloodstream Infections; Infectious Disease Epidemiology; Bacterial Pathogenesis; Systematic Reviews and Meta-analysis.

Respond to this report

Responses (1)

Author Response

28 Oct 2026

Diane Ashiru-Oredope, UK Health Security Agency, London, UK

We thank the reviewer for the comprehensive review. We have made significant clarification edits to the protocol.

Format of responses:
Reviewer Comment
Author Response and Action Taken

Title and objective use terms such as “determinants” and “drivers” which imply causality that observational studies cannot establish.
We agree that the review is expected to synthesise predominantly observational evidence and therefore cannot establish causal relationships. The review aims to identify factors associated with observed inequalities rather than determine causal effects.
Title revised to focus on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.
“Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol”

Page 2 (Abstract)
Wording such as “drivers and risk factors contributing to disparities” implies causal effects.
We agree and have revised wording throughout the manuscript to reflect associations rather than causation.
Abstract revised to refer to factors associated with inequalities rather than drivers or determinants.
“The aim of this review is to identify and synthesise factors associated with inequalities in AMR BSI and related outcomes among socioeconomically disadvantaged populations in high-income countries (HICs).”

Page 2 (Abstract)
Reviewer numbers are inconsistently reported.
We thank the reviewer for identifying this inconsistency. Reviewer availability continues to evolve during protocol implementation.
Reviewer numbers removed and replaced with descriptions of review-team procedures and quality assurance processes.

“A team of reviewers will complete screening of allocated abstracts and titles for eligibility...”

Page 4 (Introduction)
Distinguish deaths associated with AMR from deaths attributable to AMR.
We thank the reviewer for this clarification and agree.
Introduction revised to distinguish AMR-associated and AMR-attributable mortality.
“In 2019, an estimated 4.95 million deaths were associated with bacterial antimicrobial resistance, including approximately 1.27 million deaths attributable to resistance.² Annual treatment costs for antimicrobial resistant bacterial infections are estimated at US$412 billion; however, the effects of these infections are not felt evenly across all regions, populations and communities.^2,3”

Page 5 (Rationale)
Existing reviews overlap with the review topic and the rationale requires clarification.
We thank the reviewer for highlighting this issue. Existing reviews have examined AMR broadly, resistant infections more generally and socioeconomic influences on AMR, but none focus specifically on AMR BSI or synthesise factors associated with inequalities in AMR BSI and related outcomes.
Additional rationale text added to clarify the unique focus of the current review.

“While these reviews have examined social determinants of AMR, resistant infections more broadly, and socioeconomic influences on AMR, none have focused specifically on AMR BSI. In addition, no previous review has synthesised evidence on factors associated with socioeconomic inequalities in AMR BSI and related outcomes.”

Page 5-6 (Rationale)
Previous reviews should be used for citation chasing rather than as primary evidence.
We agree. Existing reviews are used to establish rationale and identify potentially relevant primary studies.
Reviews will not be treated as primary evidence during synthesis.

Page 6-7 (Objectives)
Intervention objective does not necessarily address inequalities.
We agree and have refined the objective accordingly.
Intervention objective revised to focus specifically on interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.

“...interventions evaluated for their effect on inequalities in AMR BSI incidence or outcomes.”

Page 6 (Objectives)
Incidence, resistance among BSI cases and outcomes following AMR BSI represent different epidemiological questions.
We agree these are distinct outcome domains and should not automatically be combined.
Outcomes revised to distinguish incidence of AMR BSI, probability of resistance among BSI patients and outcomes following AMR BSI. Narrative synthesis will consider these separately where appropriate.

“To identify factors reported to be associated with socioeconomic inequalities in the incidence of AMR BSI, the probability of antimicrobial resistance among patients with BSI, and outcomes among patients with AMR BSI in HICs.”

Page 7 (Information Sources)
ResearchGate and Academia.edu are not bibliographic databases.
We agree and have revised the wording accordingly.
ResearchGate and Academia.edu are now described as supplementary grey-literature sources rather than bibliographic databases.

" Grey literature will be searched using Policy Commons, Core, Researchgate and Academia databases.”

Page 7-8 (Search Strategy)
Full database search strategy is not provided.
We agree transparency and reproducibility are important. Full search strategies have been developed and executed and will be reported alongside the completed systematic review.
No change: Summary search concepts are reported in the protocol. Full search strategies will accompany the completed review publication.

Pages 8 (Eligibility & Population)
Eligibility implications are unclear. It is not clear whether studies must include a socioeconomic measure or whether studies reporting PROGRESS characteristics alone would be eligible.
We thank the reviewer for highlighting this issue. The review adopts an equity-focused approach informed by the PROGRESS-Plus framework. Relevant PROGRESS-Plus characteristics may be used as indicators of populations experiencing socioeconomic disadvantage and health inequalities.
Eligibility criteria, objectives and population definitions were revised to clarify that studies may identify populations experiencing inequalities using direct socioeconomic indicators and/or relevant PROGRESS-Plus characteristics.

“All patient groups will be included where studies report data on socioeconomic disadvantage or health inequalities using direct socioeconomic measures (e.g. income, education, occupation, deprivation indices) and/or relevant PROGRESS-Plus characteristics associated with social and economic disadvantage and inequities in health outcomes. These may include place of residence, ethnicity, migration status, disability, social exclusion, or other relevant social stratifications where examined in relation to inequalities in AMR BSI and related outcomes.”

Page 8–9 (Eligibility & outcomes)
Eligible AMR BSI episodes require clearer definition.
We agree greater clarity is beneficial.
Eligibility criteria revised to require microbiologically confirmed bloodstream infection (or clearly defined surveillance-defined BSI) with phenotypic or genotypic antimicrobial resistance.

“The primary outcome of interest for included studies is population-level incidence of AMR BSI or probability of resistance among patients with BSI with AMR BSI defined as microbiologically confirmed bacterial bloodstream infection, or a clearly defined national surveillance BSI episode, with phenotypic or genotypic antimicrobial resistance.

Page 9 (Eligibility & outcomes)
Additional detail requested regarding contaminants, recurrence, polymicrobial infection, episode definitions and deduplication.
We appreciate these recommendations. Given the anticipated heterogeneity of surveillance systems and study designs, the review will accept study-defined BSI and resistance definitions and extract these where reported.
Definitions used by study authors will be extracted and considered during interpretation. No additional surveillance-specific eligibility criteria introduced.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Page 9 (Eligibility & outcomes)
Clarify whether fungal bloodstream infections are eligible.
We agree clarification is important.
Eligibility criteria revised to clarify that the review focuses on antibiotic-resistant bacterial bloodstream infections and excludes studies focused exclusively on antifungal-resistant fungal bloodstream infections.

“This review focuses on antibiotic-resistant bacterial bloodstream infections. Studies exclusively examining antifungal-resistant fungal bloodstream infections will be excluded.”

Page 9 (Eligibiliy & Exposures)
Exposure framework combines exposures, mediators, confounders and effect modifiers.
We acknowledge this issue. The review is intended to identify factors associated with inequalities rather than estimate causal effects.
Protocol revised to explicitly state that it does not seek to estimate causal effects. Study-specific analytical approaches and adjustment variables will be extracted and reported.

“The review aims to identify factors reported to be associated with socioeconomic inequalities in AMR BSI and related outcomes rather than estimate causal effects. Included studies may conceptualise variables differently as exposures, mediators, confounders or effect modifiers. Study-specific definitions and analytical approaches will therefore be extracted and reported.”

Page 9 (Exposure Section)
A conceptual framework or DAG should be developed.
We acknowledge the value of causal frameworks when estimating causal effects. However, this review is descriptive and exploratory rather than causal in nature.
No change. Clarification added regarding the purpose of the review (see above). The inability to estimate causal effects due to the heterogeneity of available studies will be included in the discussion on limitations.

Page 9 (Eligibility & Comparator Section)
Comparator should allow gradients and continuous measures.
We agree.
Comparator section revised to include deprivation gradients, ordered categories and continuous measures of socioeconomic disadvantage.

“Comparators may include populations within HICs in less deprived socioeconomic groups, comparisons across deprivation gradients or ordered socioeconomic categories, or studies reporting continuous measures of socioeconomic disadvantage.”

Page 9 (Eligibility & Comparator Section)
Socioeconomic measures across studies should be harmonised.
We agree that socioeconomic measures vary substantially across countries and settings. Given the expected heterogeneity, narrative synthesis is anticipated.
Socioeconomic measures will generally be reported as defined by study authors and interpreted narratively.

“Comparators will be reported as defined by study authors.”

Page 9 (Eligibility & Outcomes)
Prescribing-only studies should not be eligible unless linked to AMR BSI. Antimicrobial prescribing should be treated separately from AMR BSI burden.
We agree that the review should remain focused on AMR BSI and related outcomes.
Antimicrobial prescribing is treated as a healthcare utilisation outcome or explanatory factor relevant to AMR BSI inequalities rather than a standalone primary outcome.

“Secondary outcomes of interest for included studies are health outcomes of AMR BSI (e.g. morbidity and mortality), healthcare utilisation and direct medical costs associated with AMR BSI treatment.”

Page 9 (Eligibility & Outcomes)
No publication-date restrictions may result in heterogeneous resistance definitions.

We agree these factors may contribute to heterogeneity. However, extensive harmonisation may not be feasible given the anticipated diversity of included studies.

Data extraction expanded to capture resistance definitions used by study authors. These factors will be considered during interpretation of findings. No commitment made to breakpoint-era sensitivity analyses.

“Resistance definitions used by study authors will be extracted and considered when interpreting findings.”

Pages 10-11 (Screening)
LLM validation sample may be insufficient for robust recall estimation.
We acknowledge this limitation. The LLM component is exploratory and intended to support methodological evaluation rather than replace human review.
No change: Human reviewers remain responsible for study selection decisions. No substantive protocol amendment required.

Pages 10-11 (Screening)
Development and validation datasets should be separated.
We agree that development and evaluation are conceptually distinct.
Clarified that prompt refinement and performance evaluation occur as separate stages within the LLM evaluation process.

“Once performance validation is complete, the finalised prompt will be applied to the full dataset of abstracts. An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening. Further methodological evaluation of LLM performance will be reported separately from the systematic review findings.”

Pages 10-11 (Screening)
More detail should be reported regarding prompts, model parameters and implementation settings.
We agree transparency is important. However, detailed evaluation of LLM performance forms part of a separate methodological programme of work and will be reported independently.
Protocol specifies the model used and overall screening procedure. Additional technical evaluation will be reported separately. (see above)

Pages 10-11 (Screening)
LLM may miss unusual study types, non-English studies or atypical terminology.
We agree this is an important consideration. However, the LLM functions as a supplementary rather than replacement screening method.
Human review remains the primary study-selection process. Further methodological evaluation will be reported separately.(see above)

Pages 10-11 (Screening)
Screening logic should explicitly state that AI-included studies progress to full-text screening.
We agree and thank the reviewer for this suggestion.
Protocol revised to clarify that all records identified for inclusion by the AI tool progress to full-text review.

“An inclusive approach will be taken and all abstracts that are tagged for inclusion by the AI tool will also progress to full text screening”

Page 12 (Risk of Bias)
ROBINS-I may not be appropriate for exposure studies.
We thank the reviewer for this suggestion. We acknowledge the availability of ROBINS-E and will consider the most appropriate risk-of-bias approach based on included study designs.
No change: Any deviations from protocol will be transparently reported in the final review.

Page 12 (Risk of Bias)
Surveillance studies may require surveillance-specific risk-of-bias domains.
We agree surveillance studies have unique methodological considerations.
No change: No formal surveillance-specific framework introduced. These issues will be considered during interpretation of findings.

Page 12 (Plan for Data Synthesis)
Different pathogens and resistance phenotypes should not automatically be pooled. The protocol places too much emphasis on quantitative synthesis.
We agree substantial heterogeneity is anticipated across populations, socioeconomic measures, pathogens, resistance definitions and outcomes.
Studies will be grouped according to pathogen and resistance phenotype where appropriate. Meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity exists.

“Quantitative meta-analysis will only be undertaken where sufficient methodological and clinical homogeneity (including pathogens, resistance phenotypes) exist. If this is feasible, appropriate summary effect measures will be selected based on the outcome and study design.”

Page 12 (Plan for Data Synthesis)
Detailed publication-bias, subgroup and heterogeneity procedures imply a complex meta-analysis.
We acknowledge this concern. The review is expected to rely primarily on narrative synthesis, although quantitative synthesis remains possible where appropriate.
Meta-analysis methods have been retained but reframed as contingent on sufficient methodological and clinical homogeneity rather than routine analyses (see above).

View more View less

Competing Interests

None

Alongside their report, reviewers assign a status to the article:

Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions

[1] 1. Okeke IN, de Kraker MEA , Van Boeckel TP, et al.: The scope of the antimicrobial resistance challenge. Lancet. 2024; 403(10442): 2426–2438. Publisher Full Text

[2] 2. Antimicrobial Resistance C: Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022; 399(10325): 629–655. Publisher Full Text

[3] 3. The Lancet: Antimicrobial resistance: an agenda for all. Lancet. Jun 1 2024; 403(10442): 2349. Publisher Full Text

[4] 4. Ayorinde A, Ghosh I, Ali I, et al.: Health inequalities in infectious diseases: a systematic overview of reviews. BMJ Open. 2023; 13(4): e067429. Publisher Full Text

[5] 5. CSDH: Closing the gap in a generation: health equity through action on the social determinants of health. Final Report of the Commission on Social Determinants of Health. Geneva: World Health Organization; 2008.

[6] 6. Marmot M: Fair Society, Healthy Lives: Strategic review of health inequalities in England post-2010.2010.

[7] 7. Alividza V, Mariano V, Ahmad R, et al.: Investigating the impact of poverty on colonization and infection with drug-resistant organisms in humans: a systematic review. Infect. Dis. Poverty. 2018; 7: 76. PubMed Abstract | Publisher Full Text | Free Full Text

[8] 8. Tony Atkinson BC, Marlier E, Nolan B: Social Indicators: The EU and Social Inclusion.2002.

[9] 9. Jourdain S, Smeesters PR, Denis O, et al.: Differences in nasopharyngeal bacterial carriage in preschool children from different socio-economic origins. Clin. Microbiol. Infect. 2010; 17.

[10] 10. Bagger JP, Taylor KM: Postoperative infection with meticillin-resistant Staphylococcus aureus and socioeconomic background. Lancet. 2004; 363.

[11] 11. Young DM, Charlebois ED, Chambers H, et al.: An Epidemic of Methicillin-Resistant Staphylococcus aureus Soft Tissue Infections Among Medically Underserved Patients. Arch. Surg. 2004; 139.

[12] 12. Bratu S, Landman D, Gupta J, et al.: A population-based study examining the emergence of community-associated methicillin-resistant Staphylococcus aureus USA300 in New York City. Ann. Clin. Microbiol. Antimicrob. 2006; 5: 29. Publisher Full Text

[13] 13. Nomamiukor BO, Horner C, Kirby A, et al.: Living conditions are associated with increased antibiotic resistance in community isolates of Escherichia coli. J. Antimicrob. Chemother. 2015; 70(11): 3154–3158. PubMed Abstract | Publisher Full Text

[14] 14. McMullen KM, Warren DK, Woeltje KF: The changing susceptibilities of methicillin-resistant Staphylococcus aureus at a midwestern hospital: the emergence of “community-associated” MRSA. Am. J. Infect. Control. 2009; 37(6): 454–457. Publisher Full Text

[15] 15. Chen FM, Farley M, Plikaytis B, et al.: Geocoding and Linking Data from Population-based Surveillance and the US Census to Evaluate the Impact of Median Household Income on the Epidemiology of Invasive Streptococcus pneumoniae Infections. Am. J. Epidemiol. 1998; 148.

[16] 16. Huang SS, Finkelstein JA, Rifas-Shiman SL, et al.: Community-level predictors of pneumococcal carriage and resistance in young children. Am. J. Epidemiol. Apr 1 2004; 159(7): 645–654. Publisher Full Text

[17] 17. Garcia-Rey C, Fenoll A, Aguilar L, et al.: Effect of social and climatological factors on antimicrobial use and Streptococcus pneumoniae resistance in different provinces in Spain. J. Antimicrob. Chemother. Aug 2004; 54(2): 465–471. PubMed Abstract | Publisher Full Text

[18] 18. Parsons HK, Carter MJ, Sanders DS, et al.: Helicobacter pylori antimicrobial resistance in the United Kingdom: the effect of age, sex and socio-economic status. Aliment. Pharmacol. Ther. Sep 2001; 15(9): 1473–1478. Publisher Full Text

[19] 19. Harvey EJ, De Brun C, Casale E, et al.: Influence of factors commonly known to be associated with health inequalities on antibiotic use in high-income countries: a systematic scoping review. J. Antimicrob. Chemother. Apr 3 2023; 78(4): 861–870. Publisher Full Text

[20] 20. Aldridge RW, Story A, Hwang SW, et al.: Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis. Lancet. Jan 20 2018; 391(10117): 241–250. PubMed Abstract | Publisher Full Text | Free Full Text

[21] 21. Crichton J, Hickman M, Campbell R, et al.: Prevalence of chlamydia in young adulthood and association with life course socioeconomic position: birth cohort study. PLoS One. 2014; 9(8): e104943. PubMed Abstract | Publisher Full Text | Free Full Text

[22] 22. Oren E, Koepsell T, Leroux BG, et al.: Area-based socio-economic disadvantage and tuberculosis incidence. Int. J. Tuberc. Lung Dis. Jul 2012; 16(7): 880–885. PubMed Abstract | Publisher Full Text

[23] 23. Pini A, Stenbeck M, Galanis I, et al.: Socioeconomic disparities associated with 29 common infectious diseases in Sweden, 2005-14: an individually matched case-control study. Lancet Infect. Dis. Feb 2019; 19(2): 165–176. PubMed Abstract | Publisher Full Text

[24] 24. Mancuso JD, Diffenderfer JM, Ghassemieh BJ, et al.: The Prevalence of Latent Tuberculosis Infection in the United States. Am. J. Respir. Crit. Care Med. Aug 15 2016; 194(4): 501–509. Publisher Full Text

[25] 25. UK Health Security Agency: English surveillance programme for antimicrobial utilisation and resistance (ESPAUR) Report 2022 to 2023.2023.

[26] 26. Gilham EL, Higgins H, Mason E, et al.: National reporting on factors associated with health inequalities and infection, antimicrobial resistance and consumption rates in England: 2018–2022. Lancet. 2024; 404. Publisher Full Text

[27] 27. Ray GT, Suaya JA, Baxter R: Trends and characteristics of culture-confirmed Staphylococcus aureus infections in a large U.S. integrated health care organization. J. Clin. Microbiol. Jun 2012; 50(6): 1950–1957. PubMed Abstract | Publisher Full Text | Free Full Text

[28] 28. Turnidge SYCTSJvHLEBJCJD: Impact of ethnicity and socio-economic status on Staphylococcus aureus bacteremia incidence and mortality: a heavy burden in Indigenous Australians. BMC Infect. Dis. 2012; 12.

[29] 29. Gualandi N, Mu Y, Bamberg WM, et al.: Racial Disparities in Invasive Methicillin-resistant Staphylococcus aureus Infections, 2005-2014. Clin. Infect. Dis. Sep 28 2018; 67(8): 1175–1181. Publisher Full Text

[30] 30. See I, Wesson P, Gualandi N, et al.: Socioeconomic Factors Explain Racial Disparities in Invasive Community-Associated Methicillin-Resistant Staphylococcus aureus Disease Rates. Clin. Infect. Dis. Mar 1 2017; 64(5): 597–604. Publisher Full Text

[31] 31. Tosas Auguet O, Betley JR, Stabler RA, et al.: Evidence for Community Transmission of Community-Associated but Not Health-Care-Associated Methicillin-Resistant Staphylococcus Aureus Strains Linked to Social and Material Deprivation: Spatial Analysis of Cross-sectional Data. PLoS Med. Jan 2016; 13(1): e1001944. PubMed Abstract | Publisher Full Text | Free Full Text

[32] 32. Matthew R, Blakiston JTF: Population-level exposures associated with MRSA and ESBL-E. coli infection across district health boards in Aotearoa New Zealand: an ecological study. NZMJ. 2020; 133.

[33] 33. Piper Jenks N, Pardos de la Gandara M, D’Orazio BM, et al.: Differences in prevalence of community-associated MRSA and MSSA among U.S. and non-U.S. born populations in six New York Community Health Centers. Travel Med. Infect. Dis. Nov–Dec 2016; 14(6): 551–560. Publisher Full Text

[34] 34. Shutt AE, Ashiru-Oredope D, Price J, et al.: The intersection of the social determinants of health and antimicrobial resistance in human populations: a systematic review. BMJ Glob. Health. May 30 2025; 10(5). Publisher Full Text

[35] 35. Ghataure A, Gilham EL, Casale E, et al.: Association of factors linked to health inequalities and the risk of antibiotic-resistant infection in high-income countries: a systematic scoping review. JAC Antimicrob. Resist. 2025; 7. PubMed Abstract | Publisher Full Text | Free Full Text

[36] 36. Levitch E, Matthews L, Choi E, et al.: The impact of socioeconomic status on the prevalence of antimicrobial resistance in high-income nations: a systematic review. Antimicrob. Steward. Healthc. Epidemiol. 2025; 5(1): e264. PubMed Abstract | Publisher Full Text | Free Full Text

[37] 37. Moher D, Shamseer L, Clarke M, et al.: Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst. Rev. 2015; 4.

[38] 38. Welch VA, Jull J, Hartling L, et al.: Cochrane Handbook for Systematic Reviews of Interventions - Chapter 16: Equity and specific populations. Cochrane Handbook for Systematic Reviews of Interventions. 2024.

[39] 39. Welch V, Petticrew M, Tugwell P, et al.: PRISMA-Equity 2012 extension: reporting guidelines for systematic reviews with a focus on health equity. PLoS Med. 2012; 9(10): e1001333. PubMed Abstract | Publisher Full Text | Free Full Text

[40] 40. Ashiru-Oredope D: Data from: Risk factors associated with disparities in antibiotic-resistant bloodstream infections among socioeconomically disadvantaged groups in High-Income Countries. PROSPERO; 2024. Reference Source

[41] 41. Bank TW: Population, total - OECD members. Accessed 17/11/2025. Reference Source

[42] 42. Welch V, Brand K, Kristjansson E, et al.: Systematic reviews need to consider applicability to disadvantaged populations: inter-rater agreement for a health equity plausibility algorithm. BMC Med. Res. Methodol. 2012; 12. PubMed Abstract | Publisher Full Text | Free Full Text

[43] 43. Wu G, Eastwood C, Sapiro N, et al.: Achieving high inter-rater reliability in establishing data labels: a retrospective chart review study. BMJ Open Qual. 2024; 13(2). Publisher Full Text

[44] 44. Randolph JJ: Free-Marginal Multirater Kappa (multirater κfree): An Alternative to Fleiss’ FixedMarginal Multirater Kappa. Joensuu Learning and Instruction Symposium 2005. 2005.

[45] 45. Ouzzani M, Hammady H, Fedorowicz Z, et al.: Rayyan-a web and mobile app for systematic reviews. Syst. Rev. Dec 5 2016; 5. PubMed Abstract | Publisher Full Text | Free Full Text

[46] 46. O’Neill J, Tabish H, Welch V, et al.: Applying an equity lens to interventions: using PROGRESS ensures consideration of socially stratifying factors to illuminate inequities in health. J. Clin. Epidemiol. Jan 2014; 67(1): 56–64. PubMed Abstract | Publisher Full Text

[47] 47. Ashiru-Oredope D: Data from: PRISMA-P reporting checklist for protocols with principles of PRISMA-Equity integrated. Open Science Framework (OSF). 2026. Publisher Full Text

Factors associated with inequalities in antimicrobial-resistant bloodstream infections among socioeconomically disadvantaged populations in high-income countries: a systematic review protocol

Abstract

Introduction

Methods and analysis

Ethics and dissemination

Plain Language Summary

Background

Methods

Sharing the findings

Keywords

Revised Amendments from Version 1

Strengths and limitations of this study

Introduction

Rationale

Review objectives

Methods

Design

Electronic searches

Table 1. Summary of search strategy for electronic database search.

Eligibility

Figure 1. Map* of the OECD’s members by estimated population size in 2024.

Screening

Figure 2. Screening methodology of manual human reviewers and AI Abstract Screening Tool.

Data management and extraction

Risk of bias (quality) assessment

Plan for data synthesis

Patient and public involvement

Ethics and dissemination

Data availability

Underlying data

Extended data

Reporting guidelines

Acknowledgements

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Competing Interests Policy

Stay Updated

Are you an NIHR-funded researcher?

Thank you!