Themed Paper– Original ResearchSocial gradients in health and social care costs: Analysis of linked electronic health records in Kent, UK
Introduction
Health inequalities have been described as ‘the systematic differences in the health of people occupying unequal positions in society’, for example, due to differences in income, education, occupation, material resources and social status.1 Reducing these inequalities has become a key policy objective both in the UK2 and internationally,3 but despite this, health inequalities remain persistent and progress in reducing them has been a challenge.4, 5
Despite the vast literature demonstrating the existence of health inequalities, there has been less research into their impact on healthcare costs in England. A recent study of national hospital data by Asaria et al.6 found that inpatient costs in England in 2011/12 were 31% higher for patients in the most deprived quintile than for those in the most affluent quintile and estimated that the total annual cost associated with this inequality was £4.8bn. Another study of inpatient hospital costs by Kelly et al.7 found a 35% difference in costs between the most and least deprived quintiles, in patients older than 65 years. A study by Charlton et al.8 on primary care data in the UK found that deprivation was associated with greater morbidity and increased healthcare costs. No studies were found exploring this relationship on services outside of secondary care and primary care.
There has also been research on the relationship between deprivation and healthcare utilisation, from which the impacts on costs can be reasonably inferred. Reviews of the literature by Dixon et al.,9 Goddard and Smith10 and Cookson et al.11 conclude that deprived groups tend to consume more healthcare due to greater health needs. However, these inequities vary by service: in general, poorer populations tend to use more general practitioner (GP) services, relative to need, than affluent groups but are less likely to be referred on for specialist elective care. Uptake of health promotion and preventative services was also found to be lower in areas of high deprivation.10, 11
In multiple studies, deprivation has been found to be a strong predictor of accident and emergency (A&E) attendance and hospital admission.12, 13, 14, 15 The studies' authors suggest many possible reasons for this: increased need for healthcare, less capability for self-care, lack of awareness or understanding of the most appropriate health services and lower uptake of preventative services. This demonstrates the importance of looking at impacts between different services because they may be linked: lower use of preventative services may lead to higher use of emergency services. For example, one study showed that deprived populations had higher A&E attendance but lower use of the National Health Service (NHS) telephone line, ‘NHS Direct’.16 Goddard and Smith's10 review describes the difficulties in capturing the impacts of deprivation across the wide range of complementary and substitute services involved in long-term care, such as social care, due to the complexity of different providers involved and differing funding streams. At present, social care in the UK is funded from local authority budgets rather than via the NHS.
Given the policy goals of the NHS to better integrate care between these different sectors,17 it would be informative to assess the system-wide association between deprivation and costs. We found no literature on the socio-economic patterning of social care or community care costs. However, given that it is well established that there is a higher prevalence of multimorbidity and chronic long-term conditions in deprived populations,18, 19, 20 we would expect this to be reflected in higher community care and social care costs in deprived groups. Similarly, there was also no literature on the association between deprivation and the cost of mental health services. Again, we know that the prevalence of mental health conditions is associated with deprivation,18, 20, 21 and so we would also expect a social gradient in mental health expenditure with higher costs for those living in more deprived areas.
The difficulties in analysing system-wide impacts can be overcome through the analysis of linked electronic health records. The Kent Integrated Dataset (KID) is a ‘whole-population’ database, developed by Kent's local authority public health team since 2014, which links patient-level data across primary, secondary, community, mental health and social care while anonymising personal data.22 The database includes data for all residents of Kent and from most of the healthcare and social care providers in the area, linked by means of the patients' NHS number as a common identifier.
This study evaluates the association between socio-economic deprivation and annual per capita costs of health and social care in Kent. Previous studies at a patient level have tended to focus on a particular type of cost, such as hospital costs or primary care costs. The more comprehensive nature of this study and the disaggregated analysis by cost category is important because there may be differential impacts of deprivation across care settings, and impacts on one part of the system may be compensated for by impacts on other parts of the system.
Section snippets
Methods
Patient data were extracted from the KID using Microsoft SQL Server. Age is known to be a key determinant of healthcare expenditure, with older people more likely to utilise healthcare and social care services. Because of this, the study was restricted to people older than 55 years, as a group with high care costs overall. Therefore, the inclusion criteria were people older than 55 years and currently alive, with a registered address in Kent, as of 1st May 2017. From this population list
Results
The study sample was 323,401 (Table 1). This compares to a whole population of 512,120 people in Kent older than 55 years and follows exclusion of patients registered to GP practices not flowing data to the KID. The sample was highly representative of the overall population of Kent, with very similar mean age, gender split and distribution among the deprivation quintiles.
Costs increased with each deprivation quintile, with mean annual cost of £1623 for people living in the most deprived
Discussion
The annual mean per capita cost was £1629 in the most deprived quintile compared with £1211 in the least deprived quintile. There was a clear social gradient in mean costs across all deprivation quintiles and all cost subcategories. The results for secondary care, an increase in costs of 27% between the most and least deprived quintiles, are similar to those in the literature (secondary care being the only cost category for which comparable literature exists). One study of all-age national
Acknowledgements
Thanks to Gerard Abi-Aad for hosting this analysis at the Kent Public Health Observatory in Kent County Council.
Ethical approval
This study was approved by London School of Hygiene and Tropical Medicine MSc Research Ethics Committee, and also Kent County Council.
Funding
None declared.
Competing interests
None declared.
References (30)
- et al.
What (or who) causes health inequalities: theories, evidence and implications?
Health Policy
(2013 Dec) The persistence of health inequalities in modern welfare states: the explanation of a paradox
Soc Sci Med
(2012 Aug)- et al.
Changes in health in England, with analysis by English regions and areas of deprivation, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013
Lancet
(2015 Dec 5) - et al.
Equity of access to health care services
Soc Sci Med
(2001 Nov 1) - et al.
Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study
Lancet
(2012 Jul 7) Fair Society, Healthy Lives: The marmot review. [Internet]
(2010)World health organization
- et al.
The costs of inequality: whole-population modelling study of lifetime inpatient hospital costs in the English National Health Service by level of neighbourhood deprivation
J Epidemiol Community Health
(2016 Oct) - et al.
Public hospital spending in England: evidence from National Health Service administrative records
Fisc Stud
(2016) - et al.
Impact of deprivation on occurrence, outcomes and health care costs of people with multiple morbidity
J Health Serv Res Pol
(2013 Oct)
Is the British National Health Service equitable? The evidence on socioeconomic differences in utilization
J Health Serv Res Pol
Socio-economic inequalities in health care in England
Fisc Stud
Socioeconomic deprivation and accident and emergency attendances: cross-sectional analysis of general practices in England
Br J Gen Pract
Cross sectional study of primary care groups in London: association of measures of socioeconomic and health status with hospital admission rates
BMJ
Characteristics of general practices associated with emergency-department attendance rates: a cross-sectional study
BMJ Qual Saf
Cited by (9)
Electronic health record adopters: a typology based on patients’ privacy concerns and perceived benefits
2022, Public HealthCitation Excerpt :It stands as a major technology for health data exchange among the triad: health policy-makers, healthcare providers and patients. Previous studies report that EHRs increase patients’ engagement in their care follow-up and the relationship they have with their physicians,4,5 reduce healthcare-related organisational and innovation costs,6–9 support accurate diagnoses and treatment decision-making and improve research as well as preventive and chronic illness care.10–15 Although healthcare providers and governments agree on the usefulness and benefits arising from EHR implementation, the understanding of patients’ adoption and extent of use remains an important challenge in many countries.
The economics of prevention
2019, Public HealthMultimorbidity healthcare expenditure in Belgium: a 4-year analysis (COMORB study)
2024, Health Research Policy and SystemsCost of health inequality to the NHS in Wales
2022, Frontiers in Public Health