Five principles for GP practices addressing inequalities in primary care

General practice sits at the centre of England's response to health inequalities. GP practices and primary care networks are often the first point of contact for the populations most affected by deprivation, ethnicity-related barriers to care, multimorbidity, digital exclusion, and social isolation. They are also under-resourced relative to the need they face.

The persistence of inequality in primary care is not, for the most part, a failure of intention. Most practices want to reach the patients who they are missing and are willing to invest time and resources in reaching seldom heard groups. The problem is that awareness alone does not change outcomes. A practice can know its cervical screening uptake is significantly lower among its most deprived patients, and still not know what to do about it on a Monday morning.

Universal access unfortunately does not produce equal outcomes. When the same appointment booking system, the same appointment invitation, and the same set of slots are offered to a population with very different levels of health literacy, digital confidence, and social standing, the result is not equity. Rather, it is a system that works well for the patients who need it least.

To help address this gap between intentions and outcomes, I've put together five practical principles for GP practices looking to make real progress. They are grounded in evidence of what works in the NHS. They do not require large new budgets. They do require a different approach to how practices identify need, design access, communicate with patients, and create capacity for proactive care.

1. Know who is not coming through your door

Most practices already hold the data they need: appointment records, QOF registers, screening recall lists, and coded diagnoses. The question is whether this data is being used systematically to identify which patients are not engaging, and whether that pattern reflects deprivation, ethnicity, age, or other structural factors.

The evidence suggests it usually does. People living in the most deprived areas in England develop multiple long-term conditions 10 to 15 years earlier than those in the most affluent [Lancet Healthy Longevity, 2021]. Despite this, a large observational study using 1.2 million GP consultations found that patients with multiple conditions in deprived areas receive consultation times equivalent to those without multimorbidity in affluent areas [BJGP, 2021]. This means that patients with higher needs had less time to talk about it. What Julian Tudor Hart described as the inverse care law over 50 years ago remains demonstrably present in general practice today [Health Foundation, 2022].

What this looks like in practice: Run regular searches to identify non-engagement patterns. Compare your QOF and LCS coverage against deprivation quintiles. Treat data as a trigger for action, not a reporting exercise.

What to avoid: Identifying the gap and taking no operational step in response.

2. Continuity of care is an inequality intervention

Seeing a familiar face, someone who knows something of your history and circumstances, is not a luxury. For patients with multimorbidity, complex social needs, or limited trust in health services, it is often the mechanism through which care actually works.

A systematic review in the BJGP found that higher continuity of care is associated with lower mortality, fewer unplanned hospital admissions, and higher patient satisfaction, with the greatest benefits for people with multiple long-term conditions [BJGP, 2020]. One study found mortality was 25% lower when GP-patient relationships exceeded 15 years compared with one year. This association is observational and should not be read as straightforwardly causal, but it is one of the more consistent findings in UK primary care research.

Access to continuity is also, unfortunately, unevenly distributed. A BJGP study of 381,474 patients in England found that five ethnic minority groups, including Black African, Black Caribbean, Bangladeshi, and Pakistani patients, had significantly lower continuity of care than white patients after adjustment [BJGP, 2023]. The gap was similar in size to the deprivation gradient.

This principle is not only about GP continuity. A named care coordinator, practice nurse, or social prescribing link worker who knows a patient's circumstances, follows up when appointments are missed, and bridges clinical and social support can provide the same relational function. For patients experiencing homelessness or complex mental health needs, having any consistent and trusted contact within the practice is often the difference between sustained engagement and dropout [Pathway, 2025]. In a recent conversation, a GP told me that he saw benefits for continuity even with Health Care Assistants: the trust created during a compassionate interaction where a HCA remembered that a patient was uncomfortable having a blood draw helped reduce the patient's anxiety.

What to avoid: Assuming that continuity is only a GP concern, or that it conflicts with efficient access models. For the most underserved patients, episodic and transactional care consistently produces poorer outcomes.

3. Access reform can deepen inequality if it is not designed carefully

The shift toward online booking, digital triage, and telephone-first access has improved convenience for patients who are digitally confident. For many others, it has made care harder to reach.

Around 10 million adults in the UK lack basic digital skills [Good Things Foundation, 2024]. A BJGP qualitative study involving 100 minoritised ethnic adults across four UK sites found that digitalised primary care is replicating and potentially exacerbating existing barriers, including digital precarity, limited language support, and staff shortcomings in responding to ethnically diverse patients [BJGP, 2024]. A separate survey found that 44% of people with complex disabilities found it difficult to book a GP appointment online [Sense, 2024].

The evidence on what specifically mitigates these risks is still developing. The Health Equity Evidence Centre's review of telephone and digital triage found no definitive evidence yet of interventions that prevent triage-related inequalities [HEEC, 2023]. Primary care is implementing access reform at scale ahead of a tested equity safeguard model.

What this looks like in practice: Maintain a real, actively signposted non-digital booking route. Train reception and care navigation staff to identify patients who cannot navigate digital systems. Audit consultation patterns periodically by age, deprivation, and ethnicity.

What to avoid: Treating access reform as equity-neutral. Offering multiple channels is not sufficient without checking whether all channels are genuinely being used.

4. Change the pathway, not just the invitation

Sending a booking link is not the same as reaching a patient. For people facing structural barriers, the standard invitation pathway is insufficient. Closing the gap requires changing what happens at every stage after the initial contact.

The evidence from cancer screening is among the clearest available. A randomised controlled trial in minority ethnic women found that combining community support worker education, appointment-scheduling assistance, and follow-up calls produced cervical screening uptake of 71%, compared with 22% in the control group at six months [HEEC synthesis of RCT evidence, 2023]. Telephone invitations to NHS Health Checks achieved approximately three times higher uptake than postal invitations in some settings, particularly when callers could culturally identify with the patient [BMC Public Health, 2019]. A community health worker initiative in a deprived Westminster ward achieved 82% higher uptake of cancer screening and NHS Health Checks compared with a control group [BMC Health Services Research, 2023].

Also consider Appt Health’s service, which can help practices embed best-practice approaches to patient engagement, ones that don’t expect patients to fit into a one-size-fits-all invitation.  

What to avoid: Equating activity with impact. Sending an invite and running a clinic does not automatically reach the patient. The pathway needs to be designed for the person who faces the most barriers, not the one for whom the process is already straightforward.

5. The clinical problem and the social context cannot be separated

Patients do not leave their circumstances behind when they attend a GP appointment. Poor housing, financial stress, caring responsibilities, social isolation, and insecure employment all shape health outcomes in ways that a clinical encounter alone cannot address.

Social prescribing link workers were introduced through PCNs to bridge exactly this gap. The evidence on their impact is consistently positive on patient experience and confidence, but more limited on measurable clinical outcomes. A BJGP study found that an additional full-time equivalent link worker per 50,000 patients was associated with higher confidence among people with long-term conditions in managing their health, and with better reported experience at their general practice [BJGP, 2025]. A systematic review found consistent improvement in mental wellbeing, but concluded that robust evidence for clinical outcome improvement remains limited [PMC, 2022].

Deployment equity is also a real problem. Link workers are most likely to be under-deployed in the most deprived and urban areas, with only 46% of social prescribing referrals between 2020 and 2022 reaching the most deprived three deprivation deciles [Lancet Public Health, 2025; BJGP, 2024] – the patients we should most want to reach with link workers. The lesson from the Deep End programme in Scotland is clear: integration into clinical workflows is the critical success factor. Link workers operating in parallel to the clinical team, rather than embedded within it, produce limited results [BJGP, 2021].

What to avoid: Assuming that creating a link worker role is sufficient without directing it toward the patients who most need it, and without a referral culture that treats social need as a legitimate clinical concern.

What progress actually requires

Health inequalities in primary care do not have to be fixed features of the system. They are produced, in part, by how care is designed and delivered: who the system reaches by default, who falls out of the pathway, and what happens when standard approaches do not work.

None of these five principles requires a large new budget. Most require a different use of what practices already have: data, staff, relationships, and access to community networks. What they do require is intention. A deliberate decision to design care around the patients who need it most, rather than those for whom the current model is already working.

General practice has always carried both individual and population health responsibilities. Reducing inequalities is not an addition to that role. The practices making the most progress are not doing something fundamentally different. They are applying good general practice principles with greater awareness of who is missing, and a more deliberate effort to reach them.