Community Health and Wellbeing Workers (CHWWs) for Preventive Healthcare Services Uptake in the UK: a cluster-randomised controlled trial protocol
File(s)
Accepted version
Author(s)
Type
Journal Article
Abstract
Introduction
Recognised within the NHS Ten Year Plan as an innovative workforce solution to enable the left shift care towards integrated, neighbourhood models, Community Health and Wellbeing Workers (CHWWs) are rapidly expanding across England. Inspired by Brazil’s Family Health Strategy, the model has scaled from an initial pilot in Westminster to over 28 localities. However, despite this national expansion, robust empirical evidence of its impact remains limited. This protocol describes a pragmatic cluster randomised controlled trial in Westminster designed to provide definitive evidence on whether embedding CHWWs within local communities effectively improves the uptake of preventive healthcare services.
Methods and analysis
A pragmatic, parallel-group, superiority cluster-randomised controlled trial is being conducted using routinely collected electronic health records. Sixty geographically defined "village" clusters (approximately 120 households in each cluster, 7,200 households in total) located in socioeconomically deprived areas of Westminster, London, were randomised (1:2 allocation ratio) to either the intervention or control group. The intervention involves locally recruited CHWWs conducting comprehensive, proactive monthly household visits to promote preventive care and address social determinants of health between January 2024 and December 2025. The control group receives usual care. The primary outcome is the Composite Referral Completion Indicator (CRCI), defined as the household-level rate of completed eligible preventive healthcare services per person-year (including immunisations, cancer screenings, and NHS Health Checks). Secondary outcomes include individual preventive service uptake and healthcare utilisation metrics (e.g. GP and A&E attendances). Outcomes will be extracted from the Whole Systems Integrated Care (WSIC) database. Primary analysis will follow the intention-to-treat principle using Generalised Linear Mixed Models (GLMM), supplemented by Hierarchical Targeted Maximum Likelihood Estimation (TMLE) for absolute effects.
Ethics and dissemination
The study relies on de-identified, routinely collected data to evaluate a new routine primary care service. Individual informed consent was waived because the use of retrospective routinely collected data for analysis fell under the existing global ethics agreement for the WSIC database (REC reference: REC18/WM/0323/IRAS252449) Nevertheless, the study was reviewed and approved by the Health Research Authority (HRA 23/HRA/4972;20th December 2023). Results will be disseminated through peer-reviewed academic publications, conference presentations, and plain-language summaries shared with the local community.
Trial registration number
NCT06144580 (ClinicalTrials.gov)
Strength and limitation of this study:
- The pragmatic, cluster-randomised design conducted within routine primary care settings strongly improves the external validity of the findings.
- Utilising the comprehensive Whole Systems Integrated Care (WSIC) database to establish a fixed point-in-time cohort allows the study to capture data on all eligible residents without requiring individual consent, thereby eliminating selection, and survivorship biases.
- The use of the Composite Referral Completion Indicator (CRCI) is a major strength as it effectively captures the holistic, multi-domain impact of the CHWWs on preventive care, though this relies heavily on the administrative accuracy and completeness of the routine electronic health records.
- While randomising at the village/cluster level was specifically chosen to minimise contamination between study arms, there remains an inherent risk of social contamination; the community-based nature of the intervention makes it easy for engaged residents to share health advice and influence the preventive health behaviours of relatives, friends, or neighbours residing in nearby control clusters.
- The study is conducted within the London Borough of Westminster, targeting the most socioeconomically deprived LSOAs; while this focus strengthens internal validity and relevance to deprived urban communities, findings may not be directly generalisable to CHWW programmes operating in different geographic, demographic, or healthcare system contexts.
Recognised within the NHS Ten Year Plan as an innovative workforce solution to enable the left shift care towards integrated, neighbourhood models, Community Health and Wellbeing Workers (CHWWs) are rapidly expanding across England. Inspired by Brazil’s Family Health Strategy, the model has scaled from an initial pilot in Westminster to over 28 localities. However, despite this national expansion, robust empirical evidence of its impact remains limited. This protocol describes a pragmatic cluster randomised controlled trial in Westminster designed to provide definitive evidence on whether embedding CHWWs within local communities effectively improves the uptake of preventive healthcare services.
Methods and analysis
A pragmatic, parallel-group, superiority cluster-randomised controlled trial is being conducted using routinely collected electronic health records. Sixty geographically defined "village" clusters (approximately 120 households in each cluster, 7,200 households in total) located in socioeconomically deprived areas of Westminster, London, were randomised (1:2 allocation ratio) to either the intervention or control group. The intervention involves locally recruited CHWWs conducting comprehensive, proactive monthly household visits to promote preventive care and address social determinants of health between January 2024 and December 2025. The control group receives usual care. The primary outcome is the Composite Referral Completion Indicator (CRCI), defined as the household-level rate of completed eligible preventive healthcare services per person-year (including immunisations, cancer screenings, and NHS Health Checks). Secondary outcomes include individual preventive service uptake and healthcare utilisation metrics (e.g. GP and A&E attendances). Outcomes will be extracted from the Whole Systems Integrated Care (WSIC) database. Primary analysis will follow the intention-to-treat principle using Generalised Linear Mixed Models (GLMM), supplemented by Hierarchical Targeted Maximum Likelihood Estimation (TMLE) for absolute effects.
Ethics and dissemination
The study relies on de-identified, routinely collected data to evaluate a new routine primary care service. Individual informed consent was waived because the use of retrospective routinely collected data for analysis fell under the existing global ethics agreement for the WSIC database (REC reference: REC18/WM/0323/IRAS252449) Nevertheless, the study was reviewed and approved by the Health Research Authority (HRA 23/HRA/4972;20th December 2023). Results will be disseminated through peer-reviewed academic publications, conference presentations, and plain-language summaries shared with the local community.
Trial registration number
NCT06144580 (ClinicalTrials.gov)
Strength and limitation of this study:
- The pragmatic, cluster-randomised design conducted within routine primary care settings strongly improves the external validity of the findings.
- Utilising the comprehensive Whole Systems Integrated Care (WSIC) database to establish a fixed point-in-time cohort allows the study to capture data on all eligible residents without requiring individual consent, thereby eliminating selection, and survivorship biases.
- The use of the Composite Referral Completion Indicator (CRCI) is a major strength as it effectively captures the holistic, multi-domain impact of the CHWWs on preventive care, though this relies heavily on the administrative accuracy and completeness of the routine electronic health records.
- While randomising at the village/cluster level was specifically chosen to minimise contamination between study arms, there remains an inherent risk of social contamination; the community-based nature of the intervention makes it easy for engaged residents to share health advice and influence the preventive health behaviours of relatives, friends, or neighbours residing in nearby control clusters.
- The study is conducted within the London Borough of Westminster, targeting the most socioeconomically deprived LSOAs; while this focus strengthens internal validity and relevance to deprived urban communities, findings may not be directly generalisable to CHWW programmes operating in different geographic, demographic, or healthcare system contexts.
Date Acceptance
2026-10-12
Citation
BMJ Open
ISSN
2044-6055
Publisher
BMJ Publishing Group
Journal / Book Title
BMJ Open
Copyright Statement
Copyright This paper is embargoed until publication. Once published the Version of Record (VoR) will be available on immediate open access.
License URL
Publication Status
Accepted