Extended Roles in Healthcare Delivery: What Is the Role of the Laboratory in Addressing Ethnicity-Related Healthcare Disparities?

Abstract

Full text

1. Introduction

The drivers of healthcare inequalities, often interlinked, include socioeconomic status, geographic barriers, gender, age, and ethnicity [1]. These factors limit access to healthcare and bias medical care, contributing to delayed diagnosis and poorer outcomes [2,3,4,5]. Such inequalities are well-recognised; in England, for example, the National Health Service (NHS) launched its Healthcare Inequalities Improvement Programme in 2021 to address disparities in healthcare access, experience, and outcomes across the population [6].

An estimated 70–80% of patient diagnoses and 95% of all clinical pathways rely on pathology results [7]. Clinical laboratories may therefore play an important, yet underexplored, role in mitigating these disparities. The recent removal of the ethnicity correction from estimated glomerular filtration rate equations [8,9] has underscored the need to identify and eliminate ethnic disparities in laboratory healthcare delivery. This editorial focuses on the role of clinical laboratories in addressing ethnicity-related healthcare disparities within cosmopolitan populations. Ethnicity, race, and ancestry are often used interchangeably but represent distinct, though sometimes overlapping concepts. Ethnicity refers to groups sharing a common cultural and social identity; race is a social construct used to group people, largely but not exclusively, on physical characteristics; and ancestry denotes genetic lineage and geographical origins [10,11]. Ethnicity is routinely collected in NHS electronic health records (EHR), though not in all cosmopolitan healthcare systems, and may be used to identify disparities in access, outcomes, and service use. Broad ethnicity or race categories are often used as convenient proxies for underlying genetic diversity and environmental factors affecting laboratory results. For example, a gene affecting neutrophil levels is more prevalent in individuals of African descent, contributing to benign neutropenia [12]. However, ethnicity is not a perfect proxy for genetic diversity. Substantial genetic heterogeneity exists within ethnic groups, and descriptors like race, ethnicity, or geographic origin do not capture the continuum of individual variation. In the USA, The National Academies of Sciences, Engineering, and Medicine therefore recommended shifting away from using race, ethnicity, and geographic origin as proxies for genetic ancestry groups [13]. Although not ideal, until genomic databases include broader population coverage, ethnicity remains the most relevant construct for current routine laboratory practice.

This editorial focuses on reducing ethnicity-related healthcare inequalities. This includes addressing potential ethnicity bias in test development and interpretation, such as removing inappropriate ethnicity adjustments [8,14], and improving laboratory access for populations that are traditionally difficult to engage [15,16], with the aim of delivering equitable healthcare.

2. Inclusive Reference Intervals for Diverse Populations: Ethnicity-Specific or Personalised?

Reference intervals (RIs) for laboratory tests are typically defined as the central 95% ranges in a healthy (reference) population. Traditionally, laboratories apply “one-size-fits-all” or universal Ris, with adjustments for factors like sex and age. Laboratory results from minority ethnic groups are typically interpreted using RIs derived from majority ethnic populations. Yet ethnicity-related biological differences in biomarkers exist (Table 1) [8,12,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. When applied to minority ethnic groups, generic RIs may misclassify results from healthy individuals as abnormal (outside RI), resulting in unnecessary further investigation and patient anxiety, or classifiy results from individuals with disease as normal (within RI), delaying diagnosis and treatment. Ethnicity-specific RIs may therefore reduce ethnicity-related misclassification bias and improve equity. However, the clinical significance of ethnic differences in biomarkers remains incompletely established. The impact of ethnicity-related differences, for example, in HbA1c irrespective of glycaemia [33], lipid profiles [34], cardiac enzymes and serum magnesium [35,36], on clinical management remains to be further studied and clarified.

Although ethnicity is a useful indicator of disparities, ethnicity-specific RIs must be applied with caution. They risk overlooking variation within ethnic groups and may inadvertently reinforce racial bias. Additionally, not all laboratories have sufficient data to derive robust ethnicity-specific RIs, particularly for minority and paediatric subpopulations.

A promising alternative is personalised RIs—precision ranges tailored to an individual’s characteristics and incorporating multiple covariates, including age, sex, comorbidities, pregnancy status, ethnicity, and relevant genetic data. Individualised RIs derived from big datasets could reflect a more comprehensive picture of physiology and pathology, thereby reducing both false positives and false negatives across all populations [60,61]. Evidence of improved outcomes, however, remains limited [60]. Personalised RIs could be enabled by artificial intelligence (AI) using historical data stored in EHRs. Yet significant challenges remain. Large datasets required to derive personalised RIs are sparse, and personalised RIs are dependent on patients’ prior results [60]. Existing information technology infrastructure in most laboratories and hospitals is not equipped to implement such RIs [62]. AI may also perpetuate inequality [63] by reproducing biases from training data or design processes [64,65,66,67,68,69,70,71,72,73]. Avoiding bias in machine learning and AI is complex and currently without definitive solutions [74,75,76,77,78,79,80].

Personalised RIs are therefore unlikely to become a reality in the near future. Meanwhile, laboratories, especially those serving ethnically diverse populations, should remain alert to the limitations of traditional RIs and aware of ethnicity-related differences when assessing the need for ethnicity-specific intervals. Where indicated, this could include adopting suitable ethnicity-specific RIs from literature or deriving RIs for the local population using indirect data-mining approaches from laboratory and hospital electronic records [8].

3. Improving Access to Laboratory Healthcare Services

In addition to addressing ethnicity bias in test development and interpretation, all individuals should have equitable access to laboratory services. Ethnic minority populations are more likely to encounter barriers to accessing healthcare such as limited transportation, mistrust of healthcare systems, a lack of culturally appropriate services, and language differences [81,82,83]. Improving laboratory access for underserved groups through point-of-care testing (POCT) and extension of phlebotomy services offers potential mechanisms to reduce ethnic disparities in laboratory healthcare delivery.

POCT involves performing tests near the patient in a variety of clinical settings using standalone devices, mobile laboratories, or home-testing approaches. POCT has traditionally been used to provide test results in real time so that patients may quickly be triaged or given appropriate treatment, yielding improved clinical and economic outcomes compared to laboratory testing [84,85]. POCT, however, has also been used to reach individuals who may not engage conventional healthcare settings, including in places of worship, community centres, community pharmacies and primary care practices, for both diagnosis and optimisation of care [16,72,86,87,88].

The convenience and improved accessibility of POCT must be balanced with assurances of analytical quality and operator competence, within a framework of robust clinical governance that meets quality standards [89]. Clinical laboratories are uniquely positioned to ensure that POCT is safe, accurate, precise, and clinically meaningful.

Outreach phlebotomy directly addresses issues of transport, mobility, and geographical isolation that may disproportionately affect individuals from ethnic minorities. It would enable early detection and management of diseases, ultimately reducing ethnic laboratory health disparities, although its effectiveness and cost-effectiveness have not been systematically studied. Home-based sampling offers a discreet and accessible alternative that may overcome modesty-related barriers that deter individuals in certain minority populations from seeking healthcare services [90,91].

Strategies to achieve equitable access to laboratory healthcare services are more likely to succeed when developed in partnership with community and religious organisations, which can help bridge cultural barriers. For example, a successful outreach programme for hepatitis C testing among Pakistani immigrants in Scotland used mosques and women’s centres for education and testing [72]. Similarly, association with religious organisations improved access to and uptake rates of COVID-19 vaccinations in minority communities [92]. Linguistic barriers also reduce participation in preventative health programmes [93]; providing multilingual education and translated resources may enhance engagement and counter misinformation known to deter healthcare utilisation [72,94,95].

4. Conclusions

Ethnicity-related disparities in laboratory healthcare provision arise from multiple sources, including bias in test development and interpretation and reduced engagement with healthcare services due to mistrust and cultural barriers. In the short term, reducing these inequalities requires removal of inappropriate ethnicity-based equations and adoption of clinically relevant ethnicity-specific RIs. In the medium term, engaging ethnic minority community leaders and organisations in the development of health services, including outreach phlebotomy and laboratory testing, can help ensure that services are accessible, culturally appropriate, and continually evaluated for their impact on disparities. In the long term, awareness that AI models used in diagnostics may amplify existing biases must guide the development and application of AI methods that actively mitigate inequity. Laboratory medicine professionals, in collaboration with clinical and community stakeholders, are ideally positioned to implement several key strategies that address ethnicity-related disparities in laboratory healthcare delivery.