Data sources

Publicly available data sources were used, stored in open access domains in the NHS England website. National data was pooled and broken down by Integrated Care Board (ICB) and commissioning region, aiming to capture the variation in the RTT and respective population characteristics across distinct geographical areas.

We reviewed a period of five years, from August 2019 to August 2024 for commissioning regions, aiming to capture a snapshot of the impact of COVID-19 pandemic and the services’ recovery on regional level and four years, from August 2021 to August 2024 on ICB level. Although Integrated Care Systems existed since 2016, they predominantly operated as informal partnerships, becoming legal entities after the 2022 Health and Care Act [18] and the data of their respective ICBs was made publicly available via NHS England, from 2021.

The dataset that was used for the analysis later described in this manuscript can be found in the RTT Waiting Times data in NHS England website: https://www.england.nhs.uk/statistics/statistical-work-areas/rtt-waiting-times/. Data on deprivation levels was obtained from the 2019 UK Government and Ministry of Housing data on Indices of Multiple Deprivation (IMDs) and the Health Foundation’s data on the deprivation distribution across ICBs. The respective datasets can be accessed through the following links: https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019 and https://www.health.org.uk/reports-and-analysis/briefings/integrated-care-systems-what-do-they-look-like. Lastly, data was analysed on Ethnic Category Coverage per ICB and population per region, available through NHS England: https://digital.nhs.uk/data-and-information/publications/statistical/mi-ethnic-category-coverage/current. A table with their compiled data can be found in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Population, socioeconomic and waiting list characteristics of ICBs.

https://doi.org/10.1371/journal.pone.0346596.t001

Data analysis

Data was extracted and reviewed independently by two authors. Microsoft Excel and R software (version 4.4.2, R Foundation for Statistical Computing, Vienna, Austria) was used for data synthesis and analysis. Data on population was extracted by ICB and pooled into their respective regional distribution. Deprivation and ethnicity data was extracted by ICB, then analysed and reported at an ICB level. Due to missing data on the ethnic characteristics of the populations per ICB, the analysis for ethnicity was conducted on the percentage of white-British patients amidst the patients with known ethnicity.

Data for the waiting times was reported in median waiting in weeks, while data on the number of patients seen within the target of 18 weeks was reported as a percentage of the total of the referrals received by the specific ICB/region. IMD score and percentage of LSOAs in the most deprived quintile were handled as a continuous measure per ICB.

The data was not assumed to be normally distributed, as such Spearman test was used for the analysis. Correlation was assessed on ICB level between each of the variables of population, ethnic minority background and deprivation, and the waiting times in the first capture point, the percentage of patients seen within the 18-weeks, the change of waiting times over time, as well as the change in percentage of patients seen within 18-weeks over time. Bonferroni correction would be applied for statistically significant results.

Population

Spearman’s correlation drawn between the population of each ICB and percentage of patients seen in 2021, as well as the waiting time at that point, demonstrated no significant association (ρ = 0.05, p = 0.77 and ρ = 0.01, p = 0.96, respectively). Furthermore, there was no association between the size of the population of an ICB and the change in the percentage of patients seen over the period 2021–2024 (ρ = 0.07, p = 0.67) or the change in the waiting times during that period (ρ = −0.08, p = 0.6) (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Spearman correlation testing results of population size, deprivation and ethnic characteristics on waiting lists.

https://doi.org/10.1371/journal.pone.0346596.t002

Ethnicity

Ethnicity’s impact on waiting times and percentage of patients seen within 18 weeks was assessed on an ICB level (Table 2). Using Spearman’s rank correlation, the percentage of white-British patients was correlated to the percentage of patients seen within 18 weeks in 2021, demonstrating a weak positive but not statistically significant association (ρ = 0.13, p = 0.43), and the median waiting time in 2021, revealing a weak negative but not significant association (ρ = −0.11, p = 0.48). There was also not statistically significant evidence of monotonic relationship between the “white-British” ethnic group and the changes in proportion of patients seen within 18 weeks from 2021 to 2024 (ρ = 0.11, p = 0.47) or the changes in the median waiting times (ρ = 0.11, p = 0.48).

Deprivation

Deprivation data was assessed by IMD score (where higher score means higher deprivation) and percentage of most deprived quintile of LSOAs, per ICB (Table 2). Spearman’s rank correlation was undertaken to assess any relationship between the IMD score and the percentage of patients seen within 18 weeks in 2021 (ρ = −0.11, p = 0.48), as well as the median waiting time (ρ = 0.07, p = 0.65), revealing an absence of statistically significant monotonic association. There was no evidence of a statistically significant association between the deprivation and the changes in proportion of patients seen within 18 weeks from 2021 to 2024 (ρ = 0.1, p = 0.51) or the changes in the median waiting times (ρ = −0.12, p = 0.45).

Spearman’s test was also undertaken for the percentage of most deprived LSOAs in regards to the same variables as IMD score, demonstrating no evidence of statistically significant association. Specifically, the respective test values are the following: ρ = −0.11, p = 0.51 for the percentage of patients seen within 18 weeks, ρ = 0.07, p = 0.65 for the median waiting time, ρ = 0.10, p = 0.51 for the changes in proportion of patients seen within 18 weeks and ρ = −0.12, p = 0.45 for the changes in the median waiting times.

Deprivation

These results do not support the absence of association between deprivation and healthcare disparities in general, but suggest that on an ICB level this relationship is not apparent. A potential interpretation of this may be attributed to the structure of the ICBs, consisting of multiple LSOAs with different deprivation characteristics, resulting in more homogenous areas compared to socioeconomic variability across NHS Trusts. This can be postulated after reviewing the deprivation data of the ICBs: comparing Surrey Heartlands ICB, which had 0.7% of the more deprived quintile of LSOAs, to Birmingham and Solihull ICB (49.4% of the more deprived quintile), their respective IMD score difference was −23.9, which does not capture the significant difference between the least and the most deprived ICBs. A different interpretation may arise from the nature of the scoring itself- IMD scores and quintiles of deprivation are relative measures, predominantly facilitating the ranking of different areas, without reflecting the absolute deprivation of an area. Different authors have argued about the validity of composite deprivation markers capturing the true effect of a specific socioeconomic phenomenon- such as education, employment or rurality- to healthcare access [7,26]. From a population health perspective, it is a well acknowledged problem that aggregating up over larger, heterogenous geographical areas can mask significant variation in health status, with pockets of deprivation existing within larger, more affluent areas. This understanding is one of the driving factors behind the move to Neighbourhood Health Services; a new way of more integrated, place-based working for the NHS (of which the elective care reform plan is an example). The complexities of analysis of deprivation over large geographies as discussed in this paper add further weight to the argument for place-based care- the idea that devolving budgets and decision-making to a more local level allows an integrated, partnership approach which is organised around local geogrpahies and better able to respond to local need [27,28].

Although a correlation between the socioeconomic status and the effect of the COVID-19 pandemic was not established on the level of ICB areas, there was variability in the waiting times and the proportion of patients seen, as illustrated in Fig 3 and 4. Organisations, such as the King’s Fund, having studied the early impact of COVID-19 on healthcare, reported that individuals in the most deprived areas were 60% more likely to experience prolonged waits (>52 weeks) than those in affluent regions, with Northern England exhibiting the longest delays [29], between 2020 and 2021. A detailed breakdown and investigation by the King’s Fund, during the peak of the pandemic, revealed that waiting times increased by 55.2% in areas with the highest IMD, compared to 36% in the least deprived areas [30]. Similarly, The Nuffield Trust and Health Foundation reported 21% of patients’ waiting time exceeding one year across the most deprived areas, compared to 12% of patients in the least deprived areas [31].

While the IMD score and lowest quintile of LSOAs, did not reflect such a relationship, we recognise a potential limitation of this study, where such an analysis on the more homogeneous level of NHS Trusts, may have yielded a stronger association. Although such an analysis exceeds the scope of this manuscript, correlation and regression model have been recognised to differ according to the scale and size of the spatial reporting units [32–34]. Another challenge recognised relating to the scalability of spatial models, concerns the uncertainty caused by the different delineation of analysis units, as described by Kwan (2012) [35].

Ethnicity

Ethnicity was recognised as one of the variables affecting access to healthcare across the literature; being from a non-“white British” ethnic background is recognised as one of the factors predisposing to health inequalities, including worsening waiting times in both elective and emergency care [ 31,36,37]. Prolonged waiting times have been shown to have a detrimental impact on patients [38] and particularly their mental health and quality of life [39]. Delayed appointments also increase healthcare costs. Waiting times have been described as an indicator for a health service under strain [40], underlining the importance of tackling the complex social circumstances that burden such services, such as poverty [41]. The ethnographic study of the Taskforce of Multiple Conditions [42], highlighted the intersectionality of such disadvantages. Patients of non-white ethnic backgrounds, who were living in an area of socioeconomic deprivation were more likely to suffer from the summative effects of adverse outcomes, including limited access to healthcare or poor compliance to treatment regimens [42].

Several authors have worked on identifying disparities of access to emergency healthcare by ethnic minority groups [43–46], providing compelling evidence on inequalities varying from waiting times in the emergency department to simple pain relief. There is, however, limited evidence on the impact of ethnicity on the referral pathways for elective care. Fluck et al (2025) recently studied elective admissions to the hospital in the UK [47], reviewing data of a single NHS Trust in Surrey from 2019 to 2023, demonstrating a strong correlation between prolonged waiting, and ethnic minority background and deprivation. An older study by Morgan et al (2004), assessing the waiting times in a single hospital in London, suggested that there was a reverse correlation with the white-British experiencing longer waiting times [48]. This study focuses on the RTT waiting times to the first appointment, and throughout our analysis there was not evidence of correlation between ethnicity and waiting time. While this may be attributed to a masking effect by the diverse LSOAs of ICBs, the discrepancies have been predominantly reported on the admission pathways of the hospital rather than the waiting times, offering an alternative interpretation that the RTT waiting times are not as impacted by ethnicity as the emergency and routine admissions.

ENT practice

Currently in England, ENT is one of the most heavily affected specialties in regards to the RTT waiting times [49]. According to the Getting It Right First Time (GIRFT) update in 2023, a threefold increase was noted over the last 10 years, with current numbers exceeding 700,000 patients [50]. This growth may indicate systemic inefficiencies and capacity constraints, exacerbated by the COVID-19 pandemic, which redirected healthcare resources and reduced elective service capacity [16]. Given the high incidence of ENT-related symptoms in the general population [51], the challenges of RTT waiting lists are further aggravated for the ENT services.

The increased demand was partially mitigated by an increase in the capacity, NHS staff delivering services at a 5% increased capacity compared to 2019, treating 18 million patients in 2024 [52]. Unfortunately, as we highlighted earlier, through the plateauing of the recovery curves in Fig 1 and 2, such measures did not resolve the continued increase in the demand and consequently in the waiting times. The previous and current governments have sought a partnership with the independent sector, where patients can be outsourced, increasing further the capacity of the NHS [53]. Such attempts to mitigate some of the RTT waiting times burden, started before the COVID-19 pandemic [54], but now have become widely accepted as a mainstay strategy by the stakeholders. There is, however, scepticism on whether the involvement of private sector would reduce the gap of accessibility to healthcare for the underserved populations, which is now transforming the landscape of healthcare into a more complex environment to navigate, demanding more information from the service users, in order to reach any decision [55]. The concerns on private sector being a hidden driving force behind health inequalities were explored by Kirkwood et al (2025) [56], who demonstrated an increase in health disparities and in the RTT waiting times, upon the introduction of the private sector in healthcare in 2003.

In January 2025, NHS England released their new strategy on reforming elective care for patients, highlighting the importance of a paradigm shift away from hospital care [17]. Identifying that over 80% of elective pathways conclude without patients being admitted, the focus is now the digital transformation of the service and patient’s choice and accessibility. ENT was identified as one of the five specialties that need to set clear targets and transform the activity delivery model in the community [17]. Early attempts of implementing community-based services have been reported by different teams. Cottrell et al (2020) [ 57] sought to implement a primary-care based triaging tool via endoscopic video capture, along with clinical examination by a general practitioner and hearing test, reporting success in their pilot. A slightly different approach with a community-based audiology service, has been launched by the ENT team of the University College London Hospitals [58], but there are not any preliminary results available on the performance of their pathway. Remote otology and rhinology services both within England, and beyond, have been reported to delivery equitable healthcare service and tackling accessibility restrictions for the hardly reached, underserved communities, while at the same time increasing consultant capacity to review a higher volume of patients [59]. Further investigation is needed to assess whether such digitally-enabled services may actually play a role in improving the RTT waiting times for ENT, as concerns have been voiced on the technological access of the more deprived communities, potentially increasing the socioeconomic disparity in healthcare access [60,61].