(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Inequalities in provision of hip and knee replacement surgery for osteoarthritis by age, sex, and social deprivation in England between 2007–2017: A population-based cohort study of the National Joint [1]

['Erik Lenguerrand', 'Musculoskeletal Research Unit', 'Translational Health Sciences', 'Bristol Medical School', 'University Of Bristol', 'Southmead Hospital', 'Bristol', 'United Kingdom', 'Yoav Ben-Shlomo', 'Population Health Sciences']

Date: 2023-05

Abstract Background While the United Kingdom National Health Service aimed to reduce social inequalities in the provision of joint replacement, it is unclear whether these gaps have reduced. We describe secular trends in the provision of primary hip and knee replacement surgery between social deprivation groups. Methods and findings We used the National Joint Registry to identify all hip and knee replacements performed for osteoarthritis from 2007 to 2017 in England. The Index of Multiple Deprivation (IMD) 2015 was used to identify the relative level of deprivation of the patient living area. Multilevel negative binomial regression models were used to model the differences in rates of joint replacement. Choropleth maps of hip and knee replacement provision were produced to identify the geographical variation in provision by Clinical Commissioning Groups (CCGs). A total of 675,342 primary hip and 834,146 primary knee replacements were studied. The mean age was 70 years old (standard deviation: 9) with 60% and 56% of women undergoing hip and knee replacements, respectively. The overall rate of hip replacement increased from 27 to 36 per 10,000 person-years and knee replacement from 33 to 46. Inequalities of provision between the most (reference) and least affluent areas have remained constant for both joints (hip: rate ratio (RR) = 0.58, 95% confidence interval [0.56, 0.60] in 2007, RR = 0.59 [0.58, 0.61] in 2017; knee: RR = 0.82 [0.80, 0.85] in 2007, RR = 0.81 [0.80, 0.83] in 2017). For hip replacement, CCGs with the highest concentration of deprived areas had lower overall provision rates, and CCGs with very few deprived areas had higher provision rates. There was no clear pattern of provision inequalities between CCGs and deprivation concentration for knee replacement. Study limitations include the lack of publicly available information to explore these inequalities beyond age, sex, and geographical area. Information on clinical need for surgery or patient willingness to access care were unavailable. Conclusions In this study, we found that there were inequalities, which remained constant over time, especially in the provision of hip replacement, by degree of social deprivation. Providers of healthcare need to take action to reduce this unwarranted variation in provision of surgery.

Author summary Why was this study done? Joint replacements are among the more frequent elective surgeries performed in developed settings.

In England, inequalities in provision of joint replacement surgery were reported more than a decade ago, followed by a national effort to reduce these inequalities.

In a context of increasingly strained National Health Service funding and hospital budgets, alongside prolonged efforts to increase surgical capacity, it is unclear what impact these have had on inequalities in provision of joint replacement for osteoarthritis. What did the researchers do and find? We conducted an analysis of the National Joint Registry using all primary hip and knee replacements for osteoarthritis performed from 2007 to 2017 in England, merged with the Office for National Statistics official population statistics, to investigate inequalities in provision of surgery according to deprivation by patient area of residence, age, sex, and trends in these inequalities over time.

Using 675,342 primary hip and 834,146 primary knee replacements, we found that inequalities in provision of surgery between the most (reference) and least affluent areas have remained large (hip: rate ratio (RR) = 0.58, 95% confidence interval [0.56, 0.60] in 2007, RR = 0.59 [0.58, 0.61] in 2017; knee: RR = 0.82 [0.80, 0.85] in 2007, RR = 0.81 [0.80, 0.83] in 2017), despite the use of outsourcing of surgery to the private sector.

For hip replacement, Clinical Commissioning Groups (CCGs) with the highest concentration of deprived areas had lower overall provision of surgery, and CCGs with very few deprived areas had higher surgical provision.

There has been an overall improvement in the absolute numbers of hip and knee replacement operation being performed; there is now less geographical variation in rates of surgery. For publicly funded surgery, no socioeconomic inequalities were observed for knee replacements, with a smaller inequality observed for hip replacements, whereas for privately funded surgery, strong inequalities were observed for both joints. What do these findings mean? This study found that socioeconomic inequalities in provision of joint replacement surgery are still prevailing.

Care providers should take action to address unwarranted variation in the provision of joint replacement surgery.

Citation: Lenguerrand E, Ben-Shlomo Y, Rangan A, Beswick A, Whitehouse MR, Deere K, et al. (2023) Inequalities in provision of hip and knee replacement surgery for osteoarthritis by age, sex, and social deprivation in England between 2007–2017: A population-based cohort study of the National Joint Registry. PLoS Med 20(4): e1004210. https://doi.org/10.1371/journal.pmed.1004210 Academic Editor: Sanjay Basu, Harvard Medical School, UNITED STATES Received: April 30, 2022; Accepted: March 1, 2023; Published: April 27, 2023 Copyright: © 2023 Lenguerrand et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The data underlying the results presented in the study are available from The National Joint Registry (NJR) (https://www.njrcentre.org.uk). Restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data access applications can be made to the NJR Research Committee. With this in place and approval from the NJR scientific committee, the data and materials used for this research will be shared. With NJR permission in place, the data underlying the presented results will be available to access via the NJR data access network. The authors of this manuscript are not the data owner and do not have permission to share the data. Funding: This study was conducted as part of the contract between the University of Bristol and HQIP-NJR to provide statistical support for the activities of the NJR scientific committee (HQIP-FTS 010307-2022 to MRW). It was also supported by the NIHR Biomedical Research Centre at University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol (IS-BRC-1215-20011 to AWB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: We have read the journal’s policy and the authors of this manuscript have the following competing interests: EL, MRW and AWB have received grant funds to investigate the outcome of implant bearings surface for total hip replacement (CeramTec GmbH), outside the submitted work. AJ has received consultancy fees from Freshfields Bruckhaus Deringer and has held advisory board positions (which involved receipt of fees) from Anthera Pharmaceuticals, INC, outside the submitted work. AWB and MRW have received grant funds to investigate the outcome of a design of primary total knee replacement (Stryker, Triathlon), outside the submitted work. AR reported receiving grants from the National Institute of Health Research (NIHR), Orthopaedic Research UK, and DePuy Synthes Limited outside the submitted work. MRW conducts teaching on courses sponsored by Heraeus and DePuy for which his institution receives market rate payments. All other authors declare no conflicts of interest. AJ, MW, AWB, were supported by the NIHR Biomedical Research Centre at University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol. Abbreviations: CCG, Clinical Commissioning Group; COVID-19, Coronavirus Disease 2019; IMD, Index of Multiple Deprivation; LSOA, Lower Layer Super Output Area; NHS, National Health Service; ONS, Office for National Statistics; RR, rate ratio

Introduction Primary hip and knee replacements are common elective orthopaedic surgical procedures for the treatment of hip and knee pain due to end stage osteoarthritis [1]. These procedures are highly effective in reducing pain and functional limitations for the vast majority of patients [2–4] and are cost effective [5,6]. Over 200,000 hip and knee replacements are performed each year in the United Kingdom [7]. The lifetime risk in the UK is estimated to be 10.8% for women and 8.1% for men undergoing knee replacement and 11.6% women and 7.1% men for hip replacement [8]. These numbers are projected to increase [9] due to an ageing and increasingly obese population, thereby placing an increasing public health burden on the National Health Service (NHS) in respect of funding and capacity. Fairness in access to healthcare was one of the founding principles of the UK NHS at its inception in 1948 and remains so today. Commissioners of healthcare, who have responsibility for health services, need to be concerned about the quality of healthcare that they commission, with a focus on reducing unwarranted variations in quality and access [10,11]. There are well-known inequalities and variation in the provision of common surgical procedures including hip and knee replacement [12,13]. In recent years, and before the Coronavirus Disease 2019 (COVID-19) pandemic, treatment capacity in the NHS has not grown fast enough to keep up with patient need, including hip and knee replacements [14]. In February 2010, the report “Fair Society Healthy Lives” focused on evidence-based strategies for reducing health inequalities in England [15]. The past decade has coincided with austerity and increasingly strained NHS funding and hospital budgets, where the health gap has now grown between wealthy and deprived areas [16]. It has previously been shown that older patients, females, and those living in the most deprived areas have the greatest clinical need for both hip and knee replacement surgery [17]. In a study using hospital admission data from 2002, inequity in access to care was observed, with these patient groups least likely to receive access to surgery relative to their clinical need [18]. The current climate is one of increasingly strained NHS funding and hospital budgets. NHS funding growth is now slower than historical trends [19]. Independently of the COVID-19 pandemic, there was an average daily 10,015 occupied beds open overnight for trauma and orthopaedics in England between April and June 2010, dropping to 8,770 between October and December 2016 [20]. Despite there being fewer beds in the NHS, more publicly funded joint replacements are being done due to outsourcing of NHS patients to the independent sector [21,22]. From 2007 to 2017, the total numbers of hip replacements increased from 60,898 to 95,909 and knee replacements from 67,028 to 106,574 [7]. It is unclear in the context of such changes, what impact this has had on inequalities in provision and access to joint replacement in the public and private sectors. Using data on patients in England from the National Joint Registry, the aim of this study is to compare the change in provision of primary hip and knee replacement surgery for osteoarthritis by social deprivation, age, and sex across England between 2007 and 2017.

Methods We conducted a retrospective analysis of prospectively collected anonymised data on patients in England from the National Joint Registry. We used NJR data from 1 January 2007 to 31 December 2017 to identify primary hip and knee procedures performed in England [7]. It is mandatory for surgeons and their hospital to register all hip and knee replacement activity in the NJR whether the procedures are funded by the NHS or independently. The NJR contains anonymised patient data on age, sex, and year of procedure. Information on the residential area of the patient, as defined by the 2011 census Lower Layer Super Output Areas (LSOAs), are also available. Data and related programming files are available from the authors with due to permission from the NJR scientific committee (https://www.njrcentre.org.uk/research/research-requests/). We used official statistics from the Office for National Statistics (ONS) to identify midyear estimates of the age–sex-specific usual resident population of each 2011 LSOA in England. Information on the level of deprivation was obtained using the Index of Multiple Deprivation (IMD). Each LSOA is assigned a deprivation score and ranked accordingly. We used the 2015 IMD version and derived IMD quintiles, with quintile 1 (IMD1) associated with the most deprived LSOAs and quintile 5 (IMD5) associated with the least deprived LSOAs. Participants eligible for inclusion in the study were all patients aged 50 years or more, operated on and living at the time of surgery in England, who had received a primary total hip, total or unicompartmental knee replacement between 2007 and 2017 for an indication of osteoarthritis only. This included patients treated in NHS hospitals, independent hospitals, and independent sector treatment centres. Patients living or operated on outside England were excluded, as were those operated on for an indication other than or in addition to osteoarthritis, and/or who had received a revision joint replacement. Finally, patients with no residential area information were excluded. The primary outcome was the rate of provision of primary hip or knee replacement. Yearly rates of provision were derived by aggregating the NJR patient-level data by year of procedure, age, sex, and LSOA group to obtain the count of hip and knee replacements in each of these strata (numerator) and using the aggregated ONS count of population living in each of these strata (denominator). The rates are reported per 10,000 persons. The IMD 2015 was used to identify the relative level of deprivation of the patient living area, modelled in quintiles (IMD = 1 (most deprived) to IMD = 5 (least deprived)). Other variables included age group (50 to 59, 60 to 69, 70 to 79, 80+), sex (male, female), and the year of the primary joint replacement (2007 to 2017). The planning and commissioning of healthcare services for each LSOA is defined at the level of Clinical Commissioning Groups (CCGs), with LSOAs nested within CCG. We used data provided by the ONS to identify the 2011 LSOAs nested in each CCG area [23]. Analyses were stratified according to whether or not patients had received public versus privately funded procedures for their joint replacement, to see if this explained any observed trends in inequalities by social deprivation group over time. Operations performed in the independent sector but funded by the NHS were considered as public funded procedures. Statistical analysis The rates of joint replacement in each IMD group and year of joint replacement, standardised by age and sex, were first produced. Multilevel negative binomial regression models were used to compare these rates, i.e., to produce rate ratios (RRs). These models were chosen to account for overdispersion. We included an offset term to account for the population at risk in each compared group, i.e., the denominators of the modelled rates. Areas of residence (LSOAs) were modelled as random effects. The models were adjusted for deprivation groups (IMD quintiles, measured at the LSOA level from patient postcode), procedure years, patient age, and sex. They were also adjusted for an interaction between year of joint replacement and IMD group to perform within and between comparisons, i.e., investigate the differences in joint provision between IMD groups in a particular year, or the difference in joint provision over time in a particular IMD group, or the difference in provision change over time between IMD groups. The rates of joint replacement were then produced for each year of joint replacement by age or sex groups, standardised by IMD and sex or age groups. Multilevel negative binomial regression models were also used to compare the rates by age or sex strata. Hip and knee replacement patients were analysed separately. The results of the univariable and multivariable models are presented in Tables A-C in S1 Appendix. Geographical variation in hip and knee provision were then considered by CCG area, using the directly standardised age–sex rate of each CCG for 2007, 2012, and 2017 (beginning, middle, and end of study period). The CCG boundaries, as at 1 April 2017, were used to derive choropleth maps of hip and knee replacement provision (obtained from the ONS) [24]. To investigate how deprivation impacted variation in provision rates across CCG areas, we classified the CCGs according to the concentration of LSOA(s) with the highest concentration of deprived areas (percentage of LSOAs with IMD rank in Q1): less than 5% (i.e., 0% to 5% of the LSOAs within a specific CCG have an IMD rank = Q1), 5% to 15%, 15% to 30%, and ≥30%. These thresholds correspond to the quartiles of LSOAs deprivation concentration by CCG. We reported the rates using caterpillar plots. Analyses were conducted using Stata version 15.1 statistical software (StataCorp, College Station, Texas). Geographical Information System maps were produced with ArcMap 10.6. We followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guideline to report our study S1 STROBE Checklist [25]. Ethics approval and consent to participate With support under Section 251 of the NHS Act 2006, the Ethics and Confidentiality Committee (ECC) (now the Health Research Authority Confidentiality Advisory Group) allows the NJR to collect patient data where consent is indicated as “Not Recorded.” Before Personal Data and Sensitive Personal Data are recorded, express written patient consent is provided. The NJR records patient consent as either “Yes,” “No,” or “Not Recorded.”

Discussion Using data from a large national linked dataset of patients receiving primary hip and knee replacement surgery, we observed evidence of inequalities by social deprivation. The relative inequality gap by social deprivation group between the most versus least affluent areas have remained constant over time for both joints. When stratifying the effect of social deprivation by sex, the patterns seen for IMD deprivation were the same for hip and knee replacement. For knee replacement, the social deprivation inequalities were only observed for the privately funded procedures, whereas for hip replacement, evidence of disparities was observed for both publicly and privately funded procedures. The pattern of inequality by IMD deprivation group for hip replacement was consistent when stratified by age groups, i.e., was larger among older patients (70 years and older) than younger patients (<70 years old). For knee replacement, no pattern of inequalities was observed among those aged 60 years old or younger. Strengths of this study include the use of a large mandatory national dataset, capturing approximately 95% of all such procedures [7]. Data completion and accuracy are excellent for procedures recorded within the NJR [26]. The NJR captures operations for patients treated in NHS and independent hospitals along with independent sector treatment centres, and for this study, the IMD 2015 deprivation index was linked to all patients in the NJR based on the LSOA a patient lived in. Previous studies using data from Hospital Episode Statistics could only explore inequalities for NHS patients [18]. Denominator data used to calculate rates was obtained from the ONS, with these population counts obtained for each year. A strength of the study was having a decade of data to monitor trends in inequalities over a period of time that included publication of the Marmot review [15] and national focus on health inequalities, alongside austerity and increasingly strained NHS funding and hospital budgets. Limitations of the study are that we were only able to look at inequalities according to age, sex, area deprivation at the LSOA level, and geographical area at the CCG level. The reason being that we were limited by data on population counts only being available by age, sex, year, and geographical area. Hence, we could not describe inequalities by other important domains at the patient level such as ethnicity, body mass index, social class, income, and education. Further limitations were that we could not adjust for measures of clinical need for surgery, nor patient willingness to access care. This study is focussed on inequalities in the provision of joint replacement rather than about the inequalities in clinical need. Finally, this research has focused on the most frequent indication for joint replacement, around 90% of the primary operations are performed for osteoarthritis, and in patients aged 50 years or older in the UK [7,27]; the presented results do not apply to those younger or operated for another indication. Our findings are consistent with previous research where rates of joint replacement increase with age before falling in the oldest age groups [12,28,29], women receive more operations than men [12,28,29], and more affluent groups receive greater provision [12,28,30,31]. Our findings also highlight the expected social differences in the private sector but also reveal the need for tailored public health intervention to reduce social inequality in the provision of hip replacement in the public sector in the lowest social class operated. Previous studies have shown that clinical need for both hip and knee replacements increase with age up to age 84, before decreasing in those aged over 85 [17,32–34]. The rates of provision in 2016 and 2017 have remained constant for hip replacement or decreased for knee replacement, especially compared to those aged <60 years old. This is despite an increase in the oldest age groups in the general population [35]. These figures are therefore suggestive of inequity for the oldest age groups noted in previous studies [18,36]. Women have a greater clinical need for hip and knee replacement [17], and this reflects patterns of provision we observe. People living in the most deprived areas have greatest clinical need for surgery [17]. The NHS has opted to deliver part of its orthopaedic activities, funded by public resources, within the private sector. This approach has not modified the inequalities previously reported. We have identified striking, but expected, socioeconomic differences in the procedures privately funded and also in the public sector for hip procedures. Differences between the findings for hip and knee joints may be partly explained by the outcomes of hip replacement surgery being more successful, where 10% of patients are not satisfied after hip replacement compared to up to 20% for knee [37,38], differences in patient willingness of those in more deprived areas to want hip versus knee surgery, and the impact of the symptoms and disabling nature of the disease in the different joints meaning patients may be more or less likely to seek help. Lessons could be learnt from the provision of knee replacements funded by the public sector where no socioeconomic inequalities were identified. Long-term underprovision of hip replacement could explain these inequalities, in which clinical need is known to be higher in the most deprived groups, which, in turn, were least able to privately fund these procedures. On the other hand, this could demonstrate evidence of the Inverse Care Law [39], where the availability of medical care varies inversely with the need of the population served (which would apply just as well to both hip and knee joints). This is consistent with the recent report on health equity 10 years on from the Marmot review of growing inequalities in health according to deprivation and region [16]. Previous studies have suggested the observed inequities by social deprivation may be partly explained by people from more deprived areas being less willing to seek help and access for joint replacement surgery [40], as they are less positive about the benefits and outcomes of surgery, with the decision influenced by friends and family and experiences of those who have had surgery rather than opinions of health professionals. A Canadian study showed that after adjusting for patient willingness, inequity according to patient education was no longer observed [36]. This is unlikely to explain all the reported social inequalities and the NHS needs to organise the delivery of joint replacement to guarantee equality between patients. Lower social deprivation groups have poorer health status than those in higher social classes [16], but it is assumed that those undergoing joint replacement are more likely to be fitter than their peers in the same social class. This health selection effect may explain why no inequality was observed between patients aged <60 years old. Beyond this age, and despite better health status than their peers from the same social class making them eligible for surgery, lower social class patients had systematically the lowest joint replacement provision. These were already evident in the previous 2 decades [17,18], and these inequalities are still strong today, especially for the lowest social classes in the oldest age groups. Further work will be required to understand the impact of COVID-19 on these inequalities, and it is likely that they have either persisted or increased.

Conclusions As common elective procedures, large inequalities by social deprivation group exist in the delivery of hip and knee replacement surgery [13,41–45]. Although the absolute number of these procedures has increased over time, and geographical variation in rates of surgery across commissioning groups areas has declined over time, there is still evidence of socioeconomic inequalities. For publicly funded surgery, no socioeconomic inequalities were observed for knee replacements, with a smaller inequality observed for hip replacements, whereas for privately funded surgery, strong evidence of inequalities were observed for both hip and knee replacements. The relative gap between those in the most affluent and poorest areas has been constant over time. This is inverse to health need where we know those living in the poorest areas have the greatest clinical need for hip and knee replacement surgery. For hip replacement, variation in rates of provision by CCG area was ecologically related to the concentration of deprived areas within a CCG, where those with the highest concentration of deprived areas had the lowest rates of provision. This was not observed for knee replacement, suggesting that inequalities in care provision can be improved for hip replacement. The findings will be informative to commissioners of healthcare to identify how joint replacement provision should be provided in the NHS to best address these disparities.

Supporting information S1 STROBE Checklist. STROBE statement. https://doi.org/10.1371/journal.pmed.1004210.s001 (DOCX) S1 Appendix. Figure A1. Patient flow diagram for hip replacement. Figure A2. Patient flow diagram for knee replacement. Figure B. Rates and 95% confidence intervals of joint replacement provision by sex and year of procedure. Figure C. Rates ratio and 95% confidence intervals of joint replacement for males (reference: females) adjusted for age, deprivation, and area of residence (Lower Layer Super Output Area). Figure D. Rates and 95% confidence intervals of joint replacement provision by age and year of procedure. Figure E. Rates ratio and 95% confidence intervals of joint replacement between age groups (reference: <60 years old) adjusted for sex, deprivation, and area of residence (Lower Layer Super Output Area). Figure F. Rates with 95% confidence intervals of joint replacement provision by area of residence deprivation level and year of procedure by sex. Figure G. Rates ratio and 95% confidence intervals of joint replacement between level of area of residence deprivation level (reference: IMD = 5) stratified by sex and adjusted for age and area of residence (Lower Layer Super Output Area). Figure H. Rates ratio and 95% confidence intervals of joint replacement between level of area of residence deprivation level (reference: IMD = 5) stratified by type of healthcare provider and adjusted for age, sex, and area of residence (Lower Layer Super Output Area). Figure I1. Rates with 95% confidence intervals of hip replacement provision by area of residence deprivation level and year of procedure stratified by age groups. Figure I2. Rates with 95% confidence intervals of knee replacement provision by area of residence deprivation level and year of procedure stratified by age groups. Figure J1. Rates ratio and 95% confidence intervals of hip replacement between level of area of residence deprivation level (reference: IMD = 5) stratified by age groups and adjusted for sex and area of residence (Lower Layer Super Output Area). Figure J2. Rates ratio and 95% confidence intervals of knee replacement between level of area of residence deprivation level (reference: IMD = 5) stratified by age groups and adjusted for sex and area of residence (Lower Layer Super Output Area). https://doi.org/10.1371/journal.pmed.1004210.s002 (DOCX)

Acknowledgments We thank the patients and staff of all the hospitals in England, Wales, and Northern Ireland, who have contributed data to the National Joint Registry. We are grateful to the Healthcare Quality Improvement Partnership (HQIP), the NJR Research Committee, and staff at the NJR Centre for facilitating this work. The authors have conformed to the NJR’s standard protocol for data access and publication. The views expressed represent those of the authors and do not necessarily reflect those of the National Joint Registry Steering Committee or the Healthcare Quality Improvement Partnership (HQIP) who do not vouch for how the information is presented. The views expressed in this article are those of the author(s) and not necessarily those of the NHS, the NIHR, and the Department of Health and Social Care.

[END]
---
[1] Url: https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1004210

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.