Strengths and limitations
The study used a large sample of general practice data. With over 95% of the England population registered with a GP,28 the sample is nationally representative. Two indices were used in this study to capture different aspects of continuity of care and considered continuity over 1–4 years of follow-up. It was possible to disaggregate ethnicity, although the main analysis used all White ethnic groups as the comparator to minimise loss because of missing detailed ethnicity data. The main analysis may underestimate ethnic inequalities because the White Irish ethnic group had lower continuity than the White British ethnic group.
As with most previous studies, only consultations with GPs were considered. There are increasing numbers of consultations with practice nurses and allied health professionals so it would be of interest to additionally examine ethnic and other differences in continuity in those relationships.
The study controlled for key patient factors, including the presence or absence of >30 long-term conditions and combinations of physical and mental health conditions. However, data on practice factors were not available to the authors. Several practice-level factors, including those relating to the supply of and demand for GP appointments, are modifiable features that can promote or inhibit opportunities to provide continuity of care.
Comparison with existing literature
Previous studies of patients in England have used cross-sectional survey data to examine preferences for and experience of relational continuity of care across ethnic groups. A study of >2 million responders to the 2009/2010 GPPS described continuity across detailed ethnic groups.17 All ethnic minority groups were less likely to see their preferred doctor most of the time compared with those of White British ethnicity. The present findings align with that study for Black, Bangladeshi, and Pakistani groups but not for patients of Indian or Chinese ethnicity. That study adjusted for a similar set of covariates including sex, age group, deprivation quintile, and presence of long-term medical and psychological conditions. They additionally adjusted for number of GPs in the practice and the urgency of appointments the patient had recently sought, but these adjustments did not alter the direction of differences in continuity across ethnic groups.
Several other analyses of surveys of general practice have shown poorer experience of primary health care, including experience of making appointments, communication, interpersonal care, and continuity of care in most ethnic minority groups.18,19,38 The present study and one other20 using routine health data indicate that differences in expectations and other aspects of self-reporting bias16,39 are not feasible explanations for these ethnic inequalities.
Implications for research and practice
Continuity of care is valued by patients and GPs40 but this research shows that it is less available to some ethnic minority groups, including for people with multiple long-term conditions. It is plausible that aspects of healthcare delivery, including lower continuity of care, could contribute to poorer outcomes for people with multiple long-term conditions from ethnic minority groups, although this was not directly tested in the present study. Insufficient local service support to help manage long-term conditions has been described in the lived experience of people from ethnic minority groups.41 Given the established associations between continuity of care and adverse outcomes, including higher mortality, unplanned admissions, and complications,1–3 further analysis is required to test and quantify continuity as a possible link between ethnicity and these outcomes. Continuity of care has been identified as one way of tackling inequalities in other medical specialties. For example, the NHS Long Term Plan committed to improving continuity of care during pregnancy as one approach to tackling ethnic inequalities in maternity outcomes.42 The findings in the present study suggest that there could be a need to tackle inequalities in continuity of care in primary care, with possible implications for improving outcomes for people from ethnic minority groups with long-term conditions.
The present study did not explore barriers and enablers to continuity of care. There are several practice-level factors that may affect continuity and may result in differences across ethnic groups. These include aspects of demand for general practice, such as the size of the practice (with practices with a larger list size having lower continuity14,43), the number of new registrations, the health of the practice catchment area population, and local socioeconomic deprivation.16
Aspects of supply may also be related to continuity. Practices differ in how they balance rapid access to GPs with promoting higher continuity of care. GP staffing patterns, booking systems, and the total number of appointments available to book are also likely to affect patients’ ability to see a preferred GP. Continuity of care is lower where there are more part-time GPs or more GPs in the practice, and is higher in single-handed practices.17,44,45 There are also national drivers that likely affect access and the ability of practices to offer higher continuity. General practice in deprived areas is underfunded and underdoctored relative to need.46 Making the distribution of funding between general practices more equitable, and developing workforce initiatives to attract and retain general practice staff in underdoctored areas47 may reduce inequalities in continuity of care by increasing supply.
Several of the factors discussed above — the health of the practice catchment area population, underfunding and underdoctoring relative to need, and area-level socioeconomic deprivation — are unequally distributed across ethnic groups. These are manifestations of structural racism. Structural racism refers to the way in which societies foster racial discrimination through mutually reinforcing systems of health care, housing, education, employment, criminal justice, earnings, and benefits, among others. These patterns and practices in turn reinforce discriminatory beliefs, values, and the distribution of resources.48 The structural factors operating across the life course49 that lead to overrepresentation of ethnic minority groups in more socioeconomically deprived areas will therefore make a contribution to the inequalities in continuity of care seen in this and other studies. The persistence of the association between ethnicity and continuity even after statistically controlling for area deprivation indicates that additional pathways may be operating. Racism may have additional effects on continuity of care through differences in the way that people from ethnic minority groups are treated in GP practices or experience barriers in living near and accessing the highest level of GP services. Sociocultural norms and language barriers may also contribute to ethnic differences in what patients expect or feel they can influence, and this could affect the extent to which people from some ethnic minority groups seek continuity of care.50,51
Further research is needed to understand why some ethnic minority groups have poorer continuity of care and to identify initiatives that could be made to improve services to meet their needs. The patient–practitioner relationship that is developed through higher continuity of care could help bridge cultural differences and reduce the lack of trust, experiences of insensitive behaviour, and lack of listening that is more commonly experienced in healthcare settings by people from some ethnic minority groups.51
Studies show that continuity of care has been declining in England in recent years.14,15 It is not yet clear how the introduction of primary care networks, the chronic shortage of GPs, the increasing use of remote consultations, and other changes in general practice will affect continuity52,53 and inequalities in continuity as the UK recovers from the long-term effects of the pandemic. Ongoing monitoring will be needed.
In conclusion, this analysis of routine health record data from a large sample of patients followed over 4 years shows that relational continuity of care is lower for people from Black African, Black Caribbean, any other Black background, Bangladeshi, and Pakistani ethnic groups. These ethnic inequalities are not accounted for by socioeconomic deprivation and are seen for people with and without multiple long-term conditions.