Week 4 of 10
Integrated neighbourhood teams
The team is where the shift either happens or does not, and the evidence on multidisciplinary working is more conditional than the policy language suggests.
A team defined by cohort, not by employer
The Neighbourhood Health Framework does not describe an integrated neighbourhood team as a building, a contract or a single employer. It describes a group of professions pointed at a named population. INTs "will bring together different professions and partners to work side by side to support people", and they "will deliver assessment, care planning, co-ordination and follow-on support" (Neighbourhood Health Framework, 2026). The membership the framework has in mind is general practice, community health services including nursing, adult and children's social care, mental health, pharmacy and the allied health professions, with the voluntary, community and social enterprise sector as a partner rather than an afterthought. Care workers can belong too: "some INTs may benefit from the inclusion of care workers".
The framework then does something bolder than it first appears. It refuses to define the team nationally, stating that "the NHS will not define nationally what should constitute an INT". What it does define is the cohort. The initial focus is people with frailty and those needing end of life care, people with multiple long-term conditions, children and young people, and people living with cancer. The case for concentrating there is empirical and stark: people over 75 living with frailty, people at end of life and care home residents "account for 3 to 5% of the population yet represent over 25% of non-elective admissions and 50% of bed days" (Neighbourhood Health Framework, 2026). The same document ties the team to a care planning target, that 95% of people with complex needs will have an agreed care plan by 2027.
Take that cohort in a single neighbourhood and the team becomes small and nameable: a lead GP, a district or community nurse, a social worker or care co-ordinator, a mental health practitioner, a pharmacist for medicines review, an allied health professional for falls and reablement, and a named link into the voluntary sector. The framework's own contracting architecture anticipates a neighbourhood of around 50,000 people served by a single neighbourhood provider, nested under multi-neighbourhood providers of around 250,000 (Neighbourhood Health Framework, 2026). The team is where that architecture has to become a person with a caseload and a telephone number.
The evidence is about conditions, not about the team
The multidisciplinary literature does not support the claim that putting professions together improves outcomes by itself. Wei and colleagues reviewed 36 systematic reviews of interprofessional collaboration and found that the factors which make it work or fail cluster at three levels: organisation, team and individual, and that the outcomes reported for patients, staff and organisations are mutually dependent (Wei et al., 2022, Journal of Interprofessional Care). The organisational level is exactly where the framework is thinnest.
The English Pioneer evaluation is the nearest thing to a direct test. Durand and colleagues studied 11 community-based multidisciplinary teams across two contrasting Pioneer sites, covering four models of team working (Durand et al., 2025, Journal of Health Services Research and Policy). Staff reported real gains: better information sharing, less duplicated work, collective responsibility for decisions, and the erosion of professional hierarchies. The paper names the barriers in the same breath, and they are the conditions this topic turns on: the absence of shared patient records, inadequate infrastructure and resources, and uncertainty about how to demonstrate value. Patients and carers "often appeared unaware of an MDT's involvement in planning their care".
Four conditions recur across these sources, and all four are organisational rather than clinical:
- a shared record the whole team can write to, not a read-only summary;
- colocation, or at least protected shared time, because co-ordination happens in the gaps between visits;
- a clear route in and out of the team, with a named co-ordinator who holds the caseload; and
- time, measured in years rather than quarters, for trust to form before outcomes are judged.
Bouton and colleagues make the conditionality concrete. Across 61 interprofessional interventions in primary care, 23 of 28 studies of patients at cardiovascular risk reported a positive effect on patient-centred outcomes, but only 8 of 18 studies of older or polypathological patients did, and the authors judged that evidence heterogeneous and thin (Bouton et al., 2023, BMC Primary Care). Frailty and multiple long-term conditions are precisely where the evidence is weakest and the policy ambition is largest.
Why some case management trials failed
The community matron policy was the last serious English attempt to do this, and it was evaluated. Gravelle and colleagues studied the Evercare pilots across nine primary care trusts, comparing 62 Evercare practices with between 6,960 and 7,695 control practices (Gravelle et al., 2007, BMJ). Case management had no significant effect on emergency admissions, emergency bed days or mortality. The point estimates ran the wrong way: admissions up 16.5% in the high-risk group, bed days up 19.0% and mortality up 34.4%, none of them statistically significant. The authors' explanation is the one a CPD reader should carry. Case management "introduced an additional range of services into primary care without an associated reduction in hospital admissions", possibly because identifying more cases also identified more unmet need. Their conclusion was that "without more radical system redesign this policy is unlikely to reduce hospital admissions".
This is not an English peculiarity. Peikes and colleagues reported 15 randomised trials of nurse-led care co-ordination for chronically ill Medicare beneficiaries, and 13 of the 15 showed no significant difference in hospitalisations; none of the programmes generated net savings (Peikes et al., 2009, JAMA). The two programmes with favourable signals had substantial in-person contact and targeted moderate to severe patients. Mieda and colleagues pooled 15 trials covering 7,995 older adults and found little to no difference in institutionalisation, mortality or hospital admissions, rating that evidence very low in certainty, although quality of life improved slightly on the physical and mental domains (Mieda et al., 2026, International Journal of Older People Nursing).
The literature does contain a positive trial, and its design explains the disagreement. Counsell and colleagues tested home-based care management delivered by a nurse practitioner and a social worker, working with the primary care physician and a geriatrics team under 12 care protocols, for low-income older adults (Counsell et al., 2007, JAMA). Quality of life improved on four of eight SF-36 scales, and emergency department visits fell, but hospital admission rates did not differ overall; the reductions in admissions appeared only in a predefined high-risk subgroup, and only in the second year.
Read together, the trials point to three design rules. First, substitution is not addition: a team layered on top of existing services without releasing capacity elsewhere makes the system do more work at the same throughput. Second, targeting matters more than intensity: effects appear in defined high-risk groups and disappear when everyone with a long-term condition is included. Third, the promised outcomes are not the measured ones: the trials that improved experience and quality of life rarely moved admissions, so a team judged only on non-elective admissions will look like a failure even when it is doing something people value.
Woodhams and colleagues, reviewing interventions to reduce avoidable admissions across a regional network, found that "many of the individual projects claimed success, yet in aggregate they have failed to halt or slow the rise in ED admissions", and that there was "little independent evaluation" of each intervention (Woodhams et al., 2012, BMC Health Services Research). The 10 Year Health Plan's own case, that integrated teams in Derby produced 2,300 fewer category 3 ambulance callouts and 1,400 fewer short hospital stays among the over 65 population within a year, rests on a single site without an independent control; the plan also cites a systematic review reporting a 6% reduction in costs and a 6% improvement in patient outcomes from integrated care (10 Year Health Plan, 2025). That is a reason to build the team with an evaluation attached, not a reason to doubt the team.
The composition decision is the one that decides whether a profession is in or out, and the framework does not make it. It names cohorts and asks systems to build teams around them; the list of who sits in the room is left to each place. That silence is where speech and language therapy keeps losing. A therapist employed across children's services, adult community clinics and care homes is easy to leave out of an adult neighbourhood team, because their caseload is already counted somewhere else. The consequence is not a gap in a directory. Swallowing difficulty after a stroke, in dementia or in frailty carries a risk of aspiration that no other member of the team is trained to assess, and the resulting admission is a hospital admission that the neighbourhood service existed to prevent. Put the same point the other way for every profession: whatever the team does not employ, the team will still meet, in the form of a crisis it cannot manage.
The workforce arithmetic that constrains delivery
An INT is a redistribution of a workforce that is not growing. Parkinson and colleagues tracked community nursing and support staff in England between 2010 and 2021 and found a twofold variation between regions: 300 staff per 100,000 people aged over 64 in the South West against 697 in the North West in November 2021 (Parkinson et al., 2024, Journal of Nursing Management). Nationally, provision fell by 21.2% across the period, with the largest reduction, 39.3%, in the East of England; only London increased, by 2.1%.
The framework asks more from the same base. It wants capacity to meet demand growth of around 3% per year nationally, with at least 78% of community health service activity within 18 weeks by the 2026 to 2027 financial year and at least 80% by 2028 to 2029, and it wants non-elective admissions and bed days for the frailty cohort down by 10% by March 2029 (Neighbourhood Health Framework, 2026). Those commitments are not additive with an INT that introduces a new meeting, a new record and a new set of relationships unless somebody's existing work is removed.
Task shifting is the usual answer and it is only a partial one. Morris argues that district nursing investment and specialist practice education had already fallen, and that "redressing the shortages of specialist district nurse practitioners with increased numbers of health care support workers will not replace the skill, knowledge, experience required to meet the complex care needs" (Morris, 2017, Nurse Education in Practice). The plan concedes the leadership gap, noting that training curricula "do not adequately prepare staff for the realities of leading and managing modern multidisciplinary teams", and pointing to the Leng Review's finding of "a lack of clarity around roles, responsibilities, and accountability" in new team models (10 Year Health Plan, 2025). Delegation without a supervisor who has time to supervise is not task shifting; it is risk transfer.
The voluntary sector is not spare capacity
The framework expects voluntary, community and social enterprise organisations as partners, and the evidence on what happens when they are treated as free capacity is not encouraging. Bickerdike and colleagues systematically reviewed 15 evaluations of social prescribing, the main route by which primary care hands work to community organisations, and rated every one at high risk of bias, citing a lack of comparative controls, short follow-up, unvalidated measures and missing data, while observing that "despite clear methodological shortcomings, most evaluations presented positive conclusions" (Bickerdike et al., 2017, BMJ Open). Santa and colleagues mapped collaboration across 321 community-based organisations in Blackpool and the Fylde Coast and found network density of 2.2%, with organisations tending to work with others in the same service domain (Santa et al., 2025, Health Research Policy and Systems). A sector that already collaborates thinly, and is funded project by project, cannot absorb a new national expectation of partnership without either resource or a reduction in what it is already asked to do.
Leadership and accountability: entity or arrangement?
This is the live question. Should an INT be a new organisation with its own budget and leadership, or a co-ordination arrangement layered on existing providers? The framework leaves the choice local and keeps a red line: hospital standard contracts and general medical services contracts "will remain the primary vehicles of delivery for the 2 biggest groups of NHS providers", and "at least in the initial stages, neighbourhood health will be delivered through commissioning reform" (Neighbourhood Health Framework, 2026). It states plainly that "neighbourhoods are not currently single organisations. In many cases, they won't need to be", and that the focus will be on outcomes, not organisational form.
The evidence supports a qualified answer rather than that agnosticism. Layered arrangements leave accountability diffuse at exactly the point the Pioneer evaluation identified as the weakness: no shared record, no shared budget, no single person accountable for the cohort. The failures recorded above were not failures of goodwill or of professional skill; they were failures to change where money and responsibility sat. Gravelle's reading, that case management failed "without more radical system redesign", and Peikes's finding that only programmes with substantial in-person contact moved anything, point the same way. The minimum credible version of the new-entity argument is not a new trust. It is one named person, accountable for the cohort, holding a delegated budget that spans health and social care, inside a provider arrangement the framework already permits through the single neighbourhood provider contract. Where a system cannot yet do that, it should say so, and run the INT as a co-ordination arrangement with a published evaluation rather than as a delivery vehicle with a headline.
The contested question
Whether an integrated neighbourhood team should be a new organisational entity with its own budget and leadership, or a coordination arrangement layered on existing providers.
Working across professions
Referral or conversation
A team that only refers to itself is a referral hub with a new name. Take one high priority cohort and write the interaction that should replace a referral: who speaks to whom, about what, by when, and what happens to the patient if nobody does. Test it against a case where the two professionals disagree about the plan.
For your profession
Written for the groups City St George's trains. Read your own first, then the two nearest to you: the disagreements between them are the working material.
General practice and primary care (including physician associates)
Physician Associate Studies (Pre-Registration) MPAS
What changes for you. Inside the team a named lead GP works to a defined cohort of frail and multiply-ill patients rather than to demand as it arrives, and shares the plan with a co-ordinator who holds the caseload. The week's design rule is substitution rather than addition: the practice changes what it stops doing, not only what it adds.
The decision on your desk. Whether to release GP and physician associate sessions into the shared caseload and delegate the whole cohort to one accountable co-ordinator, or to keep the practice as the referral route into a team it does not staff.
What goes wrong if this is done badly. If the practice layers the team on top of a full appointment book, patients are listed on two systems, the lead GP has no session to attend the meeting, and, on Gravelle's Evercare evidence, admissions rise rather than fall.
Community nursing and district nursing
Public Health and Specialist Community Nursing MSc/PGDip
What changes for you. Community and district nursing becomes the team's reach into the home, carrying the frailty caseload, catheter and wound care, palliative and end of life support, and supervising delegated care. The week's arithmetic is that this is a redistribution of a workforce that fell by 21.2 per cent between 2010 and 2021, with twofold regional variation, so the change is in supervision rather than in hours.
The decision on your desk. Whether to accept the neighbourhood team's housebound caseload with protected supervision time and a safe caseload figure, or to let the team inherit whatever the hospital cannot discharge, as district nursing has before.
What goes wrong if this is done badly. If the team is stood up without matching community nursing capacity, district nursing becomes the discharge destination of last resort and, as Morris warns, support workers are deployed without the specialist district nurse skill the complex caseload needs.
Adult social care and social work
Partner profession: local authority adult social care
What changes for you. Inside an integrated team adult social care supplies the care co-ordinator role, the social work judgement on risk and safeguarding, and the route to care homes, home care and reablement. Counsell's trial paired a nurse practitioner with a social worker under shared care protocols, which is the closest thing the week has to a tested division of labour between health and social care in one team.
The decision on your desk. Whether to pool a delegated budget and one accountable co-ordinator across health and social care, which is the week's minimum credible version of the team, or to run the team as a co-ordination arrangement with separate budgets.
What goes wrong if this is done badly. If budgets stay separate, accountability is diffuse at the point the Pioneer evaluation identified as the weakness, no single person can release a care package, and the co-ordinator holding the caseload carries risk without the money to act.
Mental health nursing and mental health practice
Mental Health Nursing BSc (Hons)
What changes for you. Inside the team a mental health practitioner keeps people with severe mental illness visible in a service built around frailty and long term conditions, running physical health checks and medicines management for a group whose physical outcomes are markedly worse. The framework keeps their core care in community mental health services, so the team role is deliberate joined-up planning rather than absorption.
The decision on your desk. Whether to place a mental health practitioner in the team as the named link for the cohort, or to rely on referral into core community mental health services and accept that this group falls outside the frailty and long term condition cohorts.
What goes wrong if this is done badly. If mental health is only a referral destination, people with co-occurring severe mental illness are excluded from the cohorts the team is built around, and the professions that reach them are absent when the care plan and its risk are agreed.
Physiotherapy
Physiotherapy BSc (Hons)
What changes for you. Physiotherapy joins the team to treat falls risk, movement and function directly, which is what keeps the frailty cohort out of hospital, in a service that names allied health professions for falls and reablement. The trials sharpen the target: effects appeared only in predefined high risk groups and only in the second year, so physiotherapy works the frailty list rather than everyone with a long term condition.
The decision on your desk. Whether to place physiotherapy in the team for the named high risk frailty cohort, or to give it a broader long term condition caseload that the evidence says dilutes the effect it produces.
What goes wrong if this is done badly. If physiotherapy is deployed at discharge and after crisis rather than at the point of functional decline, reablement arrives after independence is lost, and the admission the team was built to prevent has already happened.
Pharmacy (pharmacists and pharmacy technicians)
Pharmacy MPharm (Hons) (awarded by Kingston University, delivered jointly with St George's, University of London)
What changes for you. A clinical pharmacist takes the whole medicines list of the frailty cohort as a team function, running structured medication reviews for the people the team targets: those with complex polypharmacy, care home residents and people with multiple conditions. Medication review is named as a function of the neighbourhood team, so the review happens before a crisis rather than at one.
The decision on your desk. Whether structured medication review is planned team work with protected sessions, or an activity the pharmacist does when a crisis admission or a prescription query happens to bring a patient across the desk.
What goes wrong if this is done badly. If review is not protected in the team's caseload, care home residents accumulate prescribing without review, repeat prescribing absorbs the pharmacist, and the deprescribing work that no single condition clinician can see is never done.
Nursing associates
Foundation Degree in Healthcare - Nursing Associate
What changes for you. Nursing associates extend the registered nurse's reach into routine monitoring, wound care, health promotion and the care home caseload, and the framework allows care workers into some teams. The week's rule is that delegation without a supervisor who has time to supervise is risk transfer, so the role is a registered pair of hands inside a supervised team, not a substitute for the district nurse.
The decision on your desk. Whether to build the team's delegated tasks around a defined nursing associate scope with protected supervision, or to fill the rota with support staff whose registration and scope are not recognised in job planning.
What goes wrong if this is done badly. If the team is staffed by rostering anyone available, nursing associates are used as unqualified gap fillers, their registration and progression route go unrecognised, and the supervision the complex caseload needs is the first thing dropped.
Voluntary, community, faith and social enterprise (VCFSE) partners
Voluntary Sector Leadership and Management MSc/PGDip
What changes for you. The voluntary sector is named as a partner in the team, taking the social prescribing referrals and the community support that sits under a care plan. The week's evidence treats it as neither spare capacity nor proven: every social prescribing evaluation Bickerdike reviewed was at high risk of bias, and the Blackpool network Santa mapped had a density of 2.2 per cent.
The decision on your desk. Whether to fund the sector as a defined partner in the team with protected capacity, or to route social prescribing to it as unpaid spare capacity on top of what it already delivers.
What goes wrong if this is done badly. If the sector is treated as free capacity, project-by-project funding meets a national expectation of partnership, thinly connected organisations are asked to absorb the referral flow, and the community route in the care plan quietly closes.
Speech and language therapy
Speech and Language Therapy BSc (Hons) and MSc
What changes for you. Build the team from caseloads that already exist and the swallowing and communication work has no owner. A therapist counted under children's services is not reassigned to an adult neighbourhood team, so dysphagia after stroke, in dementia and in frailty stays with whoever refers it, and the team meets it as a crisis.
The decision on your desk. Whether the neighbourhood team employs speech and language therapy for the adult frailty and stroke cohort, or continues to reach swallowing and communication need through a referral route that no team member is trained to make.
What goes wrong if this is done badly. If the team redistributes the caseloads that exist but cannot reassign a therapist from children's services, nobody assesses aspiration risk before it becomes an admission, and the prevention target falls on a team that structurally cannot reach the work.
Critical commentary
Where this week is argued with
Full-text commentaries and critical analyses from peer-reviewed journals. The policy column tests the claim against what government actually commissioned. The practice column tests it against what staff can deliver. Read one from each before the contested question.
Policy-facing
BMJ Analysis reviewing 25 years of English partnership policies and the evidence on their effects. It argues that collaboration between local agencies is no replacement for national policy and investment, since partnership working rarely delivers the benefits policymakers imagine. It is the sharpest policy-facing test of this week's thesis and lands firmly against uncritical optimism about integrated teams.
IJIC editorial arguing that integrated care has mainly been used to cut acute demand and contain costs, so it fails on its promise to improve experience, outcomes and equity. It presses this week's thesis further than the week does, claiming clinician-led multidisciplinary teams preserve a power imbalance, and lands against professionally driven integration.
Practice-facing
Editorial assessing the evidence on multidisciplinary teams in primary care. It argues that outcomes are mixed, that teams need training and organisational support to realise their promise, and that without purposeful coordination they create siloes that harm continuity. Practice-facing, and it agrees with the week that a team is only as effective as the conditions around it.
Editorial in BMJ Quality and Safety arguing that distributing medicines work across multidisciplinary teams risks fragmenting responsibility, weakening continuity and leaving staff unsure of each other's roles. It warns that team care can undermine, not strengthen, the relational continuity complex patients need, and so challenges this week's assumption that team working is straightforwardly beneficial.
Editorial arguing the evidence supports continuity with a GP but not yet continuity with other multidisciplinary team professionals, so general practice is becoming multidisciplinary without robust evidence that this is the right direction, and role uncertainty complicates patient experience. It disagrees with this week's assumption that team working is straightforwardly good for patients and continuity alike.
Key sources
Evaluated the Evercare case management pilots in nine English primary care trusts, comparing 62 intervention practices with up to 7,695 controls. Emergency admissions, emergency bed days and mortality all showed no significant effect, with point estimates running towards more admissions. The authors' explanation, that services were added without system redesign, is the central cautionary finding for this week's argument about integrated teams.
Reports 15 randomised trials of nurse-led care co-ordination for 18,309 chronically ill Medicare beneficiaries. Thirteen of the 15 programmes showed no significant difference in hospitalisations and none generated net savings; only programmes with substantial in-person contact targeting moderate to severe patients looked cost-neutral. It shows that the null result is not a single English anomaly but a recurring pattern across trial evidence.
Randomised controlled trial of two years of home-based care management by a nurse practitioner and social worker for 951 low-income older adults, guided by 12 care protocols and a geriatrics team. Quality of life improved on four of eight SF-36 scales and emergency department visits fell, but hospital admissions did not differ overall and fell only in a predefined high-risk subgroup. It is the clearest positive counterexample and it shows how much depends on targeting.
Systematic review of 15 UK evaluations of social prescribing, the main route by which primary care refers patients to voluntary sector support. Every evaluation was rated at high risk of bias, with short follow-up, no comparative controls and unvalidated measures, yet most reported positive conclusions. It is the evidence base for treating community sector referral as promising but unproven rather than as established spare capacity.
Service evaluation of admission reduction projects across one regional network, covering risk prediction tools, community matrons, virtual wards and emergency department interventions. The authors found little independent evaluation of any intervention and concluded that in aggregate the projects had failed to slow the rise in emergency admissions. It is the paper that explains why local claims of success do not add up to a system effect.
Meta-review of 36 systematic reviews of interprofessional collaboration published between 2010 and 2020. Facilitators and barriers cluster at organisational, team and individual levels, and the authors argue that improvement requires combined effort at all three rather than professional goodwill alone. It supplies the framework for the conditions this week treats as necessary for a team to function.
Synthesis of the evaluation of 11 community-based multidisciplinary teams across two contrasting English Pioneer sites, drawing on interviews with leaders, staff, patients and carers plus observation of team meetings. Staff valued better information sharing, reduced duplication and eroded hierarchies, but the absence of shared patient records, inadequate infrastructure and difficulty demonstrating value were named as barriers. Patients and carers often did not know a team was involved in their care.
Systematic review of 65 articles covering 61 interprofessional interventions in primary care, of which 43 were prospective and randomised. Twenty-three of 28 studies in patients at cardiovascular risk reported positive patient-centred effects, but only 8 of 18 studies of older or polypathological patients did, and that evidence was judged heterogeneous and thin. It demonstrates that effectiveness is condition specific, with the weakest evidence where the neighbourhood health cohorts sit.
Longitudinal analysis of NHS workforce statistics on community nurses and nursing support staff between 2010 and 2021. Provision varied twofold between regions, from 300 staff per 100,000 people aged over 64 in the South West to 697 in the North West, and fell by 21.2% nationally, with the largest reduction in the East of England. It establishes the supply constraint on any plan that expands community-based teams.
Systematic review and meta-analysis of 15 randomised trials involving 7,995 community-dwelling older adults with frailty or functional impairment risk. Pooled estimates showed little to no difference in institutionalisation, mortality or hospital admissions, with certainty rated very low, although quality of life improved slightly on physical and mental domains. It is the most recent synthesis and it confirms the conditional reading of the case management literature.
Discussion paper on the decline of investment in district nursing and specialist district nursing education in the United Kingdom, alongside an ageing workforce, recruitment problems and growing caseloads. The author argues that increasing the number of health care support workers will not replace the specialist skill and knowledge required for complex care at home. It grounds this week's point that task shifting is not a substitute for a supervised specialist workforce.
Social network analysis of collaboration among community-based health and social care organisations in a deprived English locality, mapping 453 providers and surveying 44 organisations reporting links to 321 others. Network density was 2.2%, and organisations mostly collaborated within their own service domain. It shows how thin the community sector's existing collaborative ties are, which matters when partnership is assumed rather than resourced.
Department of Health and Social Care and NHS England (2026). Neighbourhood Health Framework. GOV.UK.
FULL TEXTThe framework that sets out what integrated neighbourhood teams are asked to be, including the refusal to define INT membership nationally, the priority cohorts, the 95% care plan target for people with complex needs by 2027, and the community waiting and admission reduction metrics. It also sets the contracting options, from single neighbourhood providers to integrated health organisations, and states that neighbourhoods are not currently single organisations. It is the primary policy object of this week's argument.
The plan that commits to the hospital to community shift and cites local integrated team results, including 2,300 fewer category 3 ambulance callouts and 1,400 fewer short hospital stays in Derby within a year. It also concedes that training curricula do not adequately prepare staff for leading multidisciplinary teams and that the Leng Review found unclear roles, responsibilities and accountability. These passages set up the contested question of evidence and accountability.