What health inequalities exist in access to, outcomes from and experience of treatment for lung cancer? A scoping review

Abstract

Objectives

Lung cancer (LC) continues to be the leading cause of cancer-related deaths and while there have been significant improvements in overall survival, this gain is not equally distributed. To address health inequalities (HIs), it is vital to identify whether and where they exist. This paper reviews existing literature on what HIs impact LC care and where these manifest on the care pathway.

Design

A systematic scoping review based on Arksey and O’Malley’s five-stage framework.

Data sources

Multiple databases (EMBASE, HMIC, Medline, PsycINFO, PubMed) were used to retrieve articles.

Eligibility criteria

Search limits were set to retrieve articles published between January 2012 and April 2022. Papers examining LC along with domains of HI were included. Two authors screened papers and independently assessed full texts.

Data extraction and synthesis

HIs were categorised according to: (a) HI domains: Protected Characteristics (PC); Socioeconomic and Deprivation Factors (SDF); Geographical Region (GR); Vulnerable or Socially Excluded Groups (VSG); and (b) where on the LC pathway (access to, outcomes from, experience of care) inequalities manifest. Data were extracted by two authors and collated in a spreadsheet for structured analysis and interpretation.

Results

41 papers were included. The most studied domain was PC (32/41), followed by SDF (19/41), GR (18/41) and VSG (13/41). Most studies investigated differences in access (31/41) or outcomes (27/41), with few (4/41) exploring experience inequalities. Evidence showed race, rural residence and being part of a VSG impacted the access to LC diagnosis, treatment and supportive care. Additionally, rural residence, older age or male sex negatively impacted survival and mortality. The relationship between outcomes and other factors (eg, race, deprivation) showed mixed results.

Conclusions

Findings offer an opportunity to reflect on the understanding of HIs in LC care and provide a platform to consider targeted efforts to improve equity of access, outcomes and experience for patients.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Provides first comprehensive summary of the literature published in the last decade pertaining to health inequalities (HIs) and where they may manifest on the lung cancer (LC) patient pathway.

• Two HI classification approaches were employed. One looking at four broad domains; Protected Characteristics; Socioeconomic and Deprivation Factors; Geographical Region; and Vulnerable or Socially Excluded Groups. The second categorising HI according to where on the care pathway they manifest: access to, outcomes from and experience of care.

• This comprehensive approach to studying HIs provides a holistic look at HIs and serves as a mechanism to begin consideration of how, and where, to target efforts to improve equity of LC care for patients.

• Due to the complex nature of the research question and study heterogeneity, assessment of comparable effect sizes, pooling of results or quantitative analysis were not possible.

• HIs in LC care are likely to be under-represented due to restrictions in recruitment and inclusion criteria for research studies investigating HIs (eg, exclusion of those who are homeless, disabled, minority ethnic groups).

Introduction

It has long been recognised that health policies and interventions do not benefit everyone equally, resulting in health inequalities (HIs). These may be described as, ‘unfair and avoidable differences in health across a population, and between different groups in society’.¹ Addressing these systematic differences is a question of social justice.

The global commitment to reduce HIs is reflected in the United Nations’ sustainable development goals.² In England, there are persistent HIs across the life-course, with disparities in healthy life expectancy rising in the last decade.³ This is despite explicit duties requiring the taxpayer-funded and universal National Health Service (NHS) to reduce unwarranted variation by having: ‘regard to the need to reduce inequalities between patients in access to health services and the outcomes achieved’.⁴ Building on this, the NHS Long Term Plan,⁵ outlines ambitions for the whole health system to close the gap on HIs and set specific targets such as significantly improving cancer survival.

The urgent need to reduce HIs has received particular focus due to the COVID-19⁶ pandemic, both globally and within England.⁷ It accentuated the inequitable access to hospital treatment, including cancer services⁸: for example, the shift to remote consultations⁹ disproportionately, negatively impact already-vulnerable groups and their ability to access healthcare.^{9 10} Following the first COVID-19 wave in 2020, the NHS announced it was accelerating the equitable and inclusive restoration of non-COVID-19 health and care services to enable all population groups to benefit equally.⁷ An explicit new goal was set for the NHS to deliver, ‘exceptional quality healthcare for all through equitable access, excellent experience, and optimal outcomes’.¹¹

HIs in lung cancer

Lung cancer (LC) originates in the lung due to uncontrolled growth of abnormal cells.¹² The most common types are small cell LC (SCLC) and non-small cell LC (NSCLC).¹³ As the leading cause of cancer-related deaths, LC is an important global public health issue.¹⁴ In the UK, LC is the third most common cancer accounting for 16%–18% of all new cancer cases and 21% of all cancer deaths.¹⁵ Annually, LC costs the UK economy £2.4 billion which is far higher than any other cancer.¹⁶ While recent years have seen significant overall improvements in LC survival, driven by improved awareness, earlier diagnosis and increasing rates of curative treatment, this trend of improvement is not equally distributed among all population groups.¹⁷ For example, people of lower socioeconomic status have lower LC survival^{18 19} and higher early LC mortality rates²⁰ and patients living in more socioeconomically deprived circumstances; from minority background; lower income or lower education are less likely to receive treatment including surgery, chemotherapy or radiotherapy.^{19 21}

Aims and objectives

To address HIs, it is vital to identify whether and where any exist. Clinical pathways are a common point of intervention for health system improvement initiatives and may, for example, be used to reduce unwarranted variation, enhance care quality or improve outcomes.²² In line with emerging national policy in England,^{23 24} the purpose of this review was to identify relevant existing literature to understand which HIs affect access to, outcomes from and experience of, a cancer pathway, using LC as an example.

Methods

A systematic scoping review was conducted based on Arksey and O’Malley’s five-stage framework²⁵ using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews.^{26 27}

Identifying the research question

The research questions were established through discussion between authors and agreed as:

1. What HIs impact LC care?

2. Where do HIs manifest on the LC care pathway (access, experience, outcomes)?

Identifying relevant studies

An online search was conducted in April 2022 (online supplemental file 1: Full search strategy). The following Cochrane Medical Subject Headings(MESH), derived terms were used: (“health inequalit*” OR “health inequit*” OR “health disparit*” OR equalit* OR equit* OR inequality* inequit*) AND (“lung cancer”). The following databases were searched: EMBASE, HMIC, Medline, PsycINFO and PubMed. To provide conclusions and recommendations using the most up-to-date literature,²⁸ search date limits were set to retrieve articles published in the last 10 years (January 2012 to April 2022). Snowballing of reference lists for included papers was also conducted (see figure 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram illustrating the process of identification, screening, eligibility and exclusion of papers (adapted from PRISMA 2020 statement²⁷). LC, lung cancer.

Study selection

Papers specifically looking at primary LC (SCLC and NSCLC) which examined domains of HI in relation to access to, outcomes from or experience of the LC pathway were included. The following types of papers were excluded: non-English language; study protocols; supplementary files; conference proceedings; editorials and opinion pieces. Investigations of other types of cancer or medical condition in conjunction to LC; those looking solely at factors such as risk and incidence relating to LC; LC screening (which is not currently endorsed as part of the LC pathway^{29 30}) were also excluded. Due to the complexity of reported changes in HIs restricting the ability to present a single finding, papers focused on trend data were excluded. Two authors screened papers based on title and abstract, and then assessed the full texts. Any discrepancies were resolved by discussion.

Charting the data

Data was organised using COVIDENCE,³¹ an online screening and extraction tool, and collated in a Microsoft Excel spreadsheet, allowing data to be sorted into themes, promoting structured analysis and interpretation.²⁵ Extracted variables included: author; year of publication; country of study; study design; population type; sample size; HI domain examined and point on care pathway (access, outcomes, experience). Data were independently retrieved by two authors and verified by a third author.

Collating, summarising and reporting the results

Examining HIs

The definition of HI factors varies across different contexts and settings.³² To apply findings to an NHS context, HIs were categorised in two separate ways. First, they were considered across four broad HI domains which have been adapted from national guidance.³³

1. Protected Characteristics (PC) as set out in the Equality Act,³⁴ for example, sex, race, religion, marital status or disability. For this review, ‘race’ encompasses nationality, skin colour and ethnic origin^{1 34} and ‘sex’ includes sex, and gender.¹

2. Socioeconomic and Deprivation Factors (SDF), for example, income, area deprivation.

3. Geographical Region (GR), for example, where people live or work for example, urban, rural, coastal.

4. Vulnerable or Socially Excluded Groups (VSG), which are not routinely well-provided for by healthcare services, for example, traveller communities, refugees, insecure housing tenure, etc.

These domains are not exhaustive; therefore, classification is open to interpretation. Even where domains are clearly defined, they may still overlap or transcend one another.³⁵

The second approach to categorising HIs follows NHS England’s (NHSE) approach,^{24 36} whereby HIs are categorised according to where on the care pathway they manifest:

• Access to health services: uptake of diagnostics; treatment (including surgery, chemotherapy, radiotherapy); palliative care; or supportive care (eg, pain management, nutritional support, counselling).

• Experience: encompassing views of patients, their families and carers, but also the staff providing care.^37–39

• Health outcomes: formal diagnosis and tumour staging; mortality and survival rates.⁴⁰

Results reporting

Results were summarised narratively to present study characteristics, HI domains identified in LC care, and point of HIs on the care pathway. Study heterogeneity did not allow for any pooling of results or quantitative analysis.

Patient and public involvement

None.

Results

Following the database searches, screening and snowballing, 41 papers were included in this review (figure 1), with summary characteristics presented in table 1. Two-thirds of studies were conducted in the USA (68.3%), followed by the UK (7.3%).

Table 1.

Summary characteristics of included papers (n=41)

Author	Year	Location	Study population	Sample size (n)	Health inequality domain(s) explored	Access	Outcomes	Experience
Andrykowski et al,70	2014	USA	LC survivors (aged 18+), 10–15 months post-diagnosis	193	GR	N	N	Y
Annesi et al,41	2022	USA	Black and white patients diagnosed with NSCLC during 2004–2016	229 018	GR; PC	Y	Y	N
Ascha et al,42	2020	USA	Medicare patients with LC with brain metastases	74 142	PC	Y	Y	N
Atkins et al,52	2017	USA	Patients diagnosed with LC between 2000 and 2006	348 002	GR; VSG	Y	Y	N
Backhus et al,66	2013	USA	County level data. No defined population	US population 296 314 208	GR; PC	N	Y	N
Bergamo et al,59	2014	USA	Individuals ≥66 years of age with NSCLC with schizophrenia	96 702	PC; VSG	Y	Y	N
Berglund et al,60	2012	UK	Patients with LC	15 582	PC; SDF	Y	Y	N
Check et al,43	2018	USA	Patients with NSCLC	5786	PC	Y	N	N
Cheyne et al,78	2013	UK	Patients with LC	1432	SDF	Y	Y	N
Concannon et al,72	2020	USA	Patients with NSCLC	162	GR; SDF; VSG	Y	Y	N
Dalwadi et al,44	2019	USA	Patients with LC age 60+ (stage I NSCLC)	62 312	PC	Y	Y	N
Elkbuli et al,64	2020	USA	Patients with LC	179 630	PC	N	Y	N
Erhunmwunsee et al,73	2012	USA	Patients with complete tumour staging and complete address in census	4820	SDF	N	Y	N
Evans et al,53	2017	Canada	Patients with LC	32 502	GR; PC; SDF; VSG	Y	N	N
Ganti et al,45	2014	USA	Patients with NSCLC	82 414	PC	Y	Y	N
Gibberd et al,46	2016	Australia	Patients with NSCLC	20 154	PC; VSG	Y	Y	N
Holmes et al,119	2018	USA	Patients with NSCLC	White=1 23 388; African American=11 824	PC	Y	N	N
John et al,67	2014	USA	Patients aged ≥21 years within 4 months of diagnosis	378	PC; SDF; VSG	N	N	Y
Johnson et al,47	2016	USA	Patients with LC	8322	GR; PC; SDF	Y	Y	N
Johnson et al,48	2020	USA	Patients with NSCLC	22 750	GR; PC; SDF	Y	Y	N
Koshy et al,120	2015	USA	Patients with primary NSCLC and received all or part of their first course of treatment	39, 822	GR; PC; SDF	Y	N	N
Lee et al,121	2020	USA	Patients with LC from 50 US states	1 087 810	GR	N	Y	N
Mazor et al,68	2022	USA	Patients with NSCLC or SCLC	99	PC	N	N	Y
Mehta et al,122	2012	USA	Patients with NSCLC	62 514	PC; SDF; VSG	Y	N	N
Morere et al,69	2015	France	‘Representative sample’ (aged 40–75)	1603	SDG; VSG	Y	N	N
Morgan et al,106	2020	USA	Patients with LC	36 469	PC	Y	N	N
Nadpara et al,58	2016	USA	Patients with LC	1689 (cohort A) and 1924 (cohort B) patients	GR; PC; VSG	Y	Y	N
Neroda et al,49	2021	USA	Patients with NSCLC	3616	PC; SDF	Y	N	N
Nur et al,61	2015	UK	Patients with NSCLC	192 658	GR; PC; SDF	Y	Y	N
Osuoha et al,54	2018	USA	Patients with LC	12 964	GR; PC	Y	Y	N
Pilleron et al,62	2021	New Zealand	Patients with LC	22 487	PC; SDF; VSG	N	Y	N
Rapp et al,51	2020	USA	Patients with NSCLC	56 534	GR; PC; SDF	Y	Y	N
Ray et al,55	2020	USA	Patients with NSCLC	6259	GR; SDF	Y	Y	N
Schroder et al,107	2020	Germany	Individuals with LC on statutory health insurance	3 163 211	SDF	N	Y	N
Shin et al,123	2019	South Korea	Patients with LC	57 400	PC	Y	Y	N
Starr et al,124	2012	Denmark	Patients with NSCLC	5, 538	PC; SDF; VSG	Y	N	N
Thomas et al,56	2017	Ireland	Patients with NSCLC	15 031	GR; PC; SDF; VSG	Y	Y	N
Walter et al,57	2018	Germany	Patients with LC	12 929	GR; PC; VSG	Y	N	N
Williams et al,50	2016	USA	Patients with NSCLC	18 466	PC	Y	Y	N
Zhang et al,65	2021	China	Patients with LC	3687	GR; PC	N	Y	N
Zullig et al,63	2013	USA	Patients with LC	2200	PC	Y	Y	N

GR, Geographical Region; LC, lung cancer; NSCLC, non-small cell lung cancer; PC, Protected Characteristics; SCLC, small cell lung cancer; SDF, Socioeconomic Deprivation Factors; VSG, Vulnerable or Socially Excluded Groups.

Identifying HIs impacting LC care

All four HI domain categories were represented within the reviewed articles, with 24 HI factors investigated (figure 2). The most studied HI domain was PC in 78.0% (32/41) of papers. Within the PC domain, race was the most frequently studied factor (19/31), followed by age (15/31), sex (13/31), marital status (6/31) and disability (1/41). SDF was the second most studied domain with 48.8% (19/41), with deprivation (9/20) most commonly investigated, followed by income and insurance status (6/20). GR was studied within 43.9% (18/41) of papers, most commonly rural residence (10/18) was investigated, followed by studies on proximity to or density of, specific services or infrastructure (4/18). VSG was the least studied domain (13/41) with comorbidities (7/14) and smoking status (4/14) as the most frequently investigated factors.

Figure 2.

Health inequality domains and factors investigated within the reviewed articles.

Where HIs manifest across the LC care pathway

HIs for LC patients were investigated based on where they manifest on the care pathway in terms of: (a) access to, (b) outcomes from or (c) experience of services. Most studies investigated inequalities in access (31/41) or outcomes from LC care (28/41). Few papers investigated inequalities of patient or staff experience (4/41). Summary findings for each pathway point can be found in online supplemental files 2 and 3 and table 2.

Table 2.

Summary of papers investigating health inequalities in experience of LC care

Pathway Point	Study population	Sample size (n)	HI domain explored	HI category	Indicator variable	Summary of findings	Reference
Experience	Survivors of LC (aged 18+), 10–15 months post-diagnosis	193	GR	Rural residence	Mental health (MH) outcomes	Rural LC survivors reported poorer MH status than urban LC survivors in all MH outcomes with three outcomes reaching significance (MOS-Mental Health, HADS-Depression and HADS-Total; all p<0.05). Rural LC survivors reported poorer MH relative to urban LC survivors with a mean effect size (ES) of 0.29 SD Model adjusted for education and number of physical comorbidities.	Andrykowski et al,70
Experience	Patients with LC aged ≥21	378	PC	Age; marital status; race; sex	Unmet needs	Patients with any perceived unmet need included 7% of white-US born (USB), 9% of white-foreign-born (FB), 13% of black-USB, 8% of Latino-USB, 24% of Latino-FB, 4% of Asian/Pacific Islander (API)-USB, 14% of API-FB and 11% of ‘other’ patients (p<0.01). Black-USB, Latino-FB and Asian-FB patients were more likely to perceive an unmet need than white-USB patients (p<0.05). Being younger, female, never married, uninsured, a current smoker or having comorbidities, anxiety/depression were significantly associated with unmet need (p<0.05). Models adjusted for race, age, sex, marital status, income, English proficiency, income, insurance status, smoking status, healthcare access and health system, need and perceived discrimination in care.	John et al,67
			SDF	Insurance Status
			VSG	Patients with comorbidities; smoking status
Experience	Patients with NSCLC or SCLC	99	PC	Race	Self-reported supportive care needs	At baseline, minorities reported higher needs across each domain (mean differences in need domains: psychological/emotional 4.9; daily living 1.9; financial 2.0; symptom 2.7; spiritual/existential 1.1; social 0.8; p<0.01 for all) except medical communication/information (mean difference 3.0; p=0.09). Over time, reported differences remained consistent except for medical communication. Model adjusted for age, sex, marital status, functional status and comorbidity index.	Mazor et al,68
Experience	Sample population of 1603 individuals aged 40–75 years	1603	SDF	Deprivation	Confidence in health system	Vulnerable population shows a lower level of confidence in the national healthcare system (p≤0.05). Vulnerable individuals are less likely to cite their general practitioner as their source of information (p≤0.05).	Morere et al,69
			VSG	Patients with comorbidities; smoking status

GR, Geographical Region; HI, health inequality; LC, lung cancer; NSCLC, non-small cell lung cancer; PC, Protected Characteristics; SCLC, small cell lung cancer; SDF, Socioeconomic and Deprivation Factors; VSG, Vulnerable or Socially Excluded Groups.

HIs in access to LC care (n=31)

Access to LC treatment (surgery, radiation, chemotherapy) was covered in the majority (29/31) of access studies (online supplemental file 2). Differences in receipt of treatment by race was investigated within 10 studies.^41–50 Black and non-white patients had decreased odds of receiving surgical treatment for LC.^41–50 For example, Ascha et al demonstrated that black, American Indian (AI) and ‘white Hispanic’ patients had a 0.70 (95% CI 0.65, 0.75) and 0.86 (95% CI 0.79, 0.93) times the odds of treatment compared with ‘white non-Hispanic’ patients.⁴² Similarly, Gibberd et al found that Aboriginal people were 46% less likely to have surgery than non-Aboriginal people (OR: 0.54; 95% CI 0.36, 0.80).⁴⁶ One study found patients who were not referred for surgery were more likely to be non-white (p≤0.01).⁵¹ Five papers^52–56 found rural patients were less likely to undergo surgery^{52 54–56} or have chemotherapy.⁵³

Access to appropriate or supportive care showed mixed results in relation to sex and race. Walter et al found men were given supportive care less often than females⁵⁷ which was also corroborated by Nadpara et al, who found that male patients were 27% (p≤0.05) less likely to receive appropriate care.⁵⁸

HIs in outcomes from LC care (n=28)

Outcomes from LC treatment were covered in 28/41 studies (online supplemental file 3). PC was the most studied domain in 19/28 papers, followed SDF in 11/28, GR in 10/28 and VSG in 6/28. Formal diagnosis and staging were discussed in 6/2841 46 52 58–60 studies and found to be influenced by age, sex, comorbidities,^{58 60} race^{41 46} and being part of a VSG.^{58 59} One study found that black patients who lived in more segregated areas were more likely to be diagnosed at stage IV (p≤0.01),⁴¹ while another found patients with schizophrenia were more likely to be diagnosed with early-stage LC compared with the general population (34.9% vs 30.6%, respectively; p<0.01).⁵⁹

Most studies on outcomes investigated differences in survival (18/28) and/or mortality (13/28). Sex and age were both predictors of mortality and survival, with several studies finding being older,51 54 56 61–63 and male42 54 56 61 64 65 both negatively impact survival and mortality for LC patients. Studies investigating the relationship between survival (n=5)41 44 50 62 63 and mortality (n=8)42 45–48 50 54 66 and race reported mixed results. For example, Dalwadi et al found that African American (AA) and AI patients had worse overall survival from early-stage NSCLC (AA 65%, AI 60% vs 70% for Caucasian individuals p≤0.01).⁴⁴ Annesi et al found that black patients in the highest quartile of segregation had 5% increased risk of death compared with white patients (HR 1.05, 95% CI 1.03, 1.08).⁴¹ Conversely, Zullig et al found black patients had longer survival rates than Caucasian patients (133 days vs 117 days, HR: 0.31; p≤0.01),⁶³ while Williams et al found no association between overall survival and race (HR: 0.97; 95% CI 0.93, 1.02).⁵⁰

Survival and its links to geographic region was outlined in four studies.51 52 55 65 Rural residence was a predictor of worse survival⁵¹ with rural patients having significantly reduced median survival (40 vs 52 months; p=0.06) compared with urban patients.⁵²

HIs in experience of LC care (n=4)

Four papers investigated inequalities in patient experience of LC care, with none considering staff experience (table 2). Two explored patient needs,^{67 68} with one finding that USA-born black and Latino patients, and overseas-born Asian patients, were more likely to report unmet needs for supportive services compared with white-USA born patients (p≤0.05).⁶⁷ Minority ethnic groups were also reported as having higher supportive care needs (p≤0.05).⁶⁸ The impact of being part of a vulnerable population was also shown to impact experience of LC care with those in VSGs showing lower confidence levels in national healthcare systems.⁶⁹ Finally, rural residence was also shown to impact patient experience with LC survivors living in rural areas reporting poorer mental status than those living in urban areas (p≤0.05).⁷⁰

Discussion

This scoping review provides a comprehensive summary of the literature published in the last decade pertaining to HIs and where they may manifest along the LC patient pathway in terms of access to, outcomes from or experience of care, and classified by one of four domains: PC, SDF, GR or VSG.

Access

We identified numerous studies that demonstrate that race impacts access to LC diagnosis and treatment.^41–50 This finding is reinforced within recent findings that black and Asian patients wait up to a month longer than white patients for some cancer diagnoses.⁷¹ Rural residence and being part of a VSG also appear linked to limited treatment access^51–56 including access to timely and appropriate care.^{58 59 72} Multiple studies demonstrated the negative impact of deprivation on access to surgery,48 60 61 73 a finding consistent with previous work which found that low socioeconomic position reduced the likelihood of receipt of any type of LC treatment, surgery or chemotherapy.²¹

Outcomes

We found numerous studies that that demonstrated that being older,51 54 56 61–63 and male42 54 56 61 64 65 both negatively impact survival and mortality for LC patients. This finding reflects current LC mortality rates in the UK which are significantly lower in females than in males.⁷⁴ Decreasing survival with age also reflects UK trends with the 5-year net survival in men ranging from 42% for 15–39 years old to just 6% for 80–99 years old.⁷⁵ Within reviewed articles, the relationship between race and survival or mortality was mixed, with studies reporting both better, worse and similar outcomes for specific groups.42 45–48 50 54 66 National LC mortality rates for England and Wales demonstrate that people of non-white ethnicity had lower mortality rates compared with the white ethnic group between 2017 and 2019,⁷⁶ similar to a study, which found that Bangladeshi, Indian, Caribbean and Black African men had higher LC survival estimates compared with white men.⁷⁷

While several reviewed studies showed deprivation impacted survival or mortality,47 48 61 73 others found no such association.^{62 78} However, an analysis of Cancer Registry data for England found LC patients from the most deprived areas lost more life years than those from the least deprived.⁷⁹

Experience

Studies assessing experience-related HIs were limited, though this review highlights the potential for factors such as race^{67 68} and rural residence⁷⁰ to impact patient care needs and mental health outcomes. These findings support the 2021 National Cancer Patient Survey which found that respondents from mixed ethnic backgrounds were least likely to say they were always treated with dignity and respect while receiving hospital treatment.⁸⁰

Implications for HI-reduction

The COVID-19 pandemic increased the spotlight on the differences experienced by patients receiving NHS care.^{6 23} As a result, the NHS’s Board announced strategic changes intended to ensure providers and commissioners of NHS services proactively deliver equitable services^81–83: the ‘Core20PLUS5’ initiative aims to reduce HIs in the 20% most deprived geographic areas, along with targeting five clinical areas with recognised inherent HIs, including early cancer diagnosis.²⁴ Accompanying this strategic shift, are several structural changes, including: a requirement for NHS organisations to name an accountable officer for reducing HIs and the Care Quality Commission, announcing a focus on HI-reduction as part of its inspection regime.^81–83

Despite some variability, the findings from this review offer a timely opportunity to not only reflect on the current understanding of HIs in LC care, but also provide a platform to begin consideration of targeted efforts to improve equity of access, outcomes and experience for patients. Based on our findings two key recommendations are suggested:

1) Collect, interrogate and act on the data

Understanding existing data is an important starting point to first recognise, and then mitigate HIs. To do this, services must be supported to collect, analyse, act on and share relevant HI data. Service evaluations should employ mixed method approaches to not only identify unwarranted variation within care but also understand the experiences of those using services.

There are some emerging practical examples of how inequalities in access, outcomes or experience are being addressed or mitigated. For example, an NHS Trust in London uses annual equity audits to identify and proactively target underrepresented groups in accessing clinical services,⁸⁴ including a review of sexual health screening coverage by PC which highlighted low screening offer rates for men. Using a combination of community events and in-reach and outreach clinics, the screening test offer was successfully increased to 98% of patients in this group.⁸⁴ Clinical guidelines have also been successfully developed to improve their cultural relevance and sensitivity to specific populations, thereby improving health outcomes (eg, for patients with diabetes who wish to fast safely during Ramadan^{85 86}).

Interventions such as these have valuable lessons for translation in LC care. For example, newly recommended targeted LC screening programmes in the UK will be designed to screen specific high-risk groups, who could be engaged through targeted events or outreach services.⁸⁷ Equally, tailored resources such as guidelines or factsheets could be developed for specific LC populations to support practitioners in addressing the HIs identified in this review.

2) Embrace the complexity of studying HIs-intersectionality and cumulative impact

Many studies included in this review focused on sole HI indicators (eg, race). This is an important limitation of existing research, as it is increasingly recognised that, ‘people are shaped by their simultaneous membership of multiple interconnected social categories’.⁸⁸ Without consideration of the combined effect of HI domains, studies are unable to accurately or adequately describe their collective impact.⁸⁹ Using an intersectional approach, defined as, ‘a way of identifying, understanding, and tackling structural inequality in a given context that accounts for the lived experience of people with intersecting identities’⁸⁸, to explore HIs can give a deeper, more nuanced understanding.⁸⁹

Three reviewed papers discussed aspects of intersectionality between HI variables.^{47 48 73} One study found that black patients were not affected by neighbourhood economic deprivation alone but were significantly impacted by the combined negative effects of segregation and poverty.⁴⁷ Another found LC outcomes are impacted by neighbourhood environments that are shaped by distribution of race, ethnicity and class.⁴⁸ Finally, Erhunmwunsee et al explored the relationship between poverty/median income and higher educational attainment and concluded these indicators were highly correlated: those living in areas with higher percentages of residents achieving higher education having improved LC outcomes.⁷³

As well as considering the intersectionality of HIs, there is also credible evidence of a cumulative effect of HIs. Experiencing inequalities in access to care will ultimately impact patient outcomes with several studies acknowledging that differences in survival may be attributed to disparities in receipt of treatment.^{52 90–94} To improve outcomes, healthcare planners should prioritise addressing issues in access to and uptake of LC treatment.⁹⁵ This has the potential to promote more equitable care by avoiding a cumulative effect of disadvantage across care pathways.

Limitations and future research

While this review provides a comprehensive summary of HIs along the LC patient pathway, some limitations should be considered.

Due to the novel and complex nature of the research question a scoping review was conducted to enable the range and type of HIs in LC to be investigated.²⁵ While this approach was considered particularly appropriate given: HIs are not universally defined; there are many potential HI domains, and study designs vary considerably, it did however, preclude quality assessment of the included studies^{25 96 97} as well as assessment of comparable effect sizes. However, this review provides a valuable precursor to a full systematic review with relevant keywords, inclusion criteria and research questions defined.²⁵ Another limitation is related to the time-bound nature of the results. The literature search was conducted in April 2022 and further evidence may have accumulated in the intervening period. However, the method presented here provides a template for updating the search and/or expanding it to a systematic review in future work.

This review excluded studies investigating screening for LC. While LC screening programmes have already been introduced in several countries, for example, Australia,⁹⁸ it was only in June 2022 when the UK’s National Screening Committee recommended a targeted programme be introduced to address HIs.⁹⁹ The programme invites people aged 55–74 years who are current or previous smokers, and therefore are at the highest risk of LC.^{99 100} An initial 10-region roll-out began in summer 2023 with national coverage expected by 2024.^{101 102} An evaluation is expected to explore impacts on HIs, including health outcomes and experiences, though it will require some time for sufficient data to accumulate.¹⁰⁰ This review also excluded papers which outlined trend data on HI indicators, due to the complexity of reporting a single finding for each study. To understand changes to HIs overtime, future work may seek to explore and monitor how HIs are impacted by changes to access and treatment options.

It should also be noted that most included studies were conducted in the USA. Our findings may therefore be significantly influenced by the characteristics of the local healthcare system. As a mixed-system without universal coverage, availability and accessibility of care is often fragmented and based on individual and geographical factors.^{103 104} Therefore, caution should be applied when generalising these findings to other countries and settings.

There are also several limitations of existing published research in this area which may have impacted our findings. First, recruitment and inclusion criteria for research studies often exclude those groups (eg, homeless, disabled, minority ethnic groups) most at risk of HIs.¹⁰⁵ Our review identified examples of exclusion of those without: spoken or written English^{68 70}; a postcode^{53 73}; complete housing records47 48 53 57 62 73; medical insurance,42 51 58 59 106 107 suggesting findings of HIs may be significantly under-represented.

The lack of common or agreed definitions for HIs factors¹⁰⁸ also poses a limitation and complicates the topic in terms of inclusion criteria and scope. For example, our review found definitions of deprivation ranging from census variables (eg, poverty level; education level; income; employment status; telephone access, etc47–49 62 73) to index of multiple deprivations^{61 78 109} making meaningful comparisons even in single HI domains difficult. Similarly, some variables interact, for example, the UK’s Equality Act 2010 defines a cancer diagnosis in itself as a disability^{110 111} thereby potentially allowing for ‘double counting’ of PC characteristics within studies. Additionally, ‘gender’ is related but distinct from ‘sex’: while the former is a social construct (eg, societal roles or norms), the latter a physiological characteristic.¹¹² So, for example, gender-based variations in smoking patterns may explain LC-incidence variations but are less plausible drivers of diagnosis disparities in the never-smoker population, or differing treatment access rates.^{113 114}

Another complication is that while ‘Ethnicity’ is a self-determined identity reflecting culture, traditions, history, language, religion, it is often conflated with ‘race’ which is based on externally observed characteristics such as skin colour.^{34 115 116} Categories of race and ethnicity varied from country of birth (eg, ‘foreign-born Asian’; ‘Aboriginal people’) to race (eg, ‘Hispanic’) to both skin colour and race (eg, ‘white Hispanic’). While localised characterisation of race aids in responding to specific research questions, it impedes consolidation of findings across studies. Equally, multiple studies allocated race or ethnicity to binary categories (eg, ‘white and ‘non-white’⁵¹). This limited categorisation prevents more nuanced understanding of HIs experienced by patients from other/additional ethnic backgrounds.

Finally, we cannot exclude possible misclassification bias or missing data, a problem that is increasingly identified in health datasets.¹⁰⁵ Despite a gold standard for how to capture census and ethnicity data existing in the UK,¹¹⁷ this is not universally applied within the health services or research: indeed, a recent Race Health Observatory report found systematic inaccuracy of NHS ethnicity data,²⁸ highlighting the ongoing challenge of conducting meaningful, unbiased HI research.¹¹⁸

Conclusion

This review provides a comprehensive overview of the current evidence for how HIs impact LC care and identifies where these HIs manifest in terms of access to, outcomes from or experience of care. There are numerous studies that provide evidence detailing that overall, HIs impact patient access to LC diagnosis, treatment and supportive care. While there is more evidence of the impact of specific HI factors (eg, age, sex) on outcomes such as mortality and survival, the relationship with other factors like race, show mixed evidence. This review provides a mechanism to begin consideration of how, and where, to target efforts to improve equity of LC care for patients. Specifically, both research and service improvement efforts to address HIs should consider the need for common definitions to align HI research, the cumulative impact of disadvantage and the role that intersectionality plays in exacerbating disparities in care for LC patients.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Not applicable.