You are here

  • Home
  • Archive
  • Volume 15, Issue 2
  • Incidence rate and associated patient characteristics of liver disease in Wales 2004–2022: a retrospective population-scale observational study

Article Text

Article menu

Download PDFPDF

Epidemiology
Original research
Incidence rate and associated patient characteristics of liver disease in Wales 2004–2022: a retrospective population-scale observational study
  1. Jingwei Gao1,
  2. Ashley Akbari1,
  3. Haroon Ahmed2,
  4. Aled Davies3,
  5. Andrew Yeoman4,
  6. Thomas Peter Ignatius Pembroke5
  1. 1Population Data Science, Faculty of Medicine, Health & Life Science, Swansea University, Swansea, UK
  2. 2Division of Population Medicine, Cardiff University, Cardiff, UK
  3. 3PRIME Centre Wales, Cardiff University, Cardiff, UK
  4. 4Aneurin Bevan University Health Board, Newport, UK
  5. 5Department of Gastroenterology and Hepatology, University Hospital of Wales, Cardiff, UK
  1. Correspondence to Dr Jingwei Gao; jingwei.gao{at}swansea.ac.uk

Abstract

Objective To describe the incidence and key demographic, socioeconomic and clinical characteristics of individuals with liver disease in Wales.

Design and setting This study is designed as a retrospective observational study that linked data of anonymised identified individuals from primary, secondary care and mortality data from the Secure Anonymised Information Linkage (SAIL) Databank in Wales.

Participants All Welsh residents who registered with a SAIL-contributing general practitioner (GP) and diagnosed with liver disease from 2004 to 2022.

Primary and secondary outcome measures Our primary outcome is the annual age-standardised incidence rate of liver disease. Secondary outcome is the numbers and frequencies of underlying aetiology and the associated comorbidities.

Results Between 2004 and 2022, 111 098 individuals received a diagnosis of liver disease in Wales and were included in this study. The incidence of liver disease increased threefold during the study period (97.7 per 100 000 inhabitants in 2004 to 316.2 per 100 000 inhabitants in 2022). A total of 79 992 individuals (72%) entered the cohort with the underlying aetiology of liver disease, including alcohol-related liver disease, non-alcoholic fatty liver disease (NAFLD), viral hepatitis, metabolic, haemochromatosis and autoimmune liver diseases. NAFLD has contributed to most of the change in incidence.

Conclusions We observed increasing incidence rates of liver disease in Wales, with NAFLD showing a particularly sharp increase and frequently identified as an underlying condition. A better understanding of the incidence of liver disease is the first step towards effective prevention, early detection and targeted intervention to improve patient outcomes.

  • EPIDEMIOLOGY
  • Hepatology
  • PUBLIC HEALTH

Data availability statement

Data may be obtained from a third party and are not publicly available. The routine data used in this study are available in the SAIL Databank at Swansea University, Swansea, UK. All proposals to use SAIL data are subject to review by an Information Governance Review Panel (IGRP). Before any data can be accessed, approval must be given by the IGRP. The IGRP gives careful consideration to each project to ensure proper and appropriate use of SAIL data. When access has been approved, it is gained through a privacy-protecting trusted research environment (TRE). SAIL has established an application process to be followed by anyone who would like to access data via SAIL https://www.saildatabank.com/application-process. This study has been approved by the IGRP as project 1492. The research adhered to ethical guidelines and the Data Protection Act 2018 to ensure the privacy and confidentiality of all data subjects involved. The reproducible SQL and R code and code lists to identify study individuals are available on Github: https://github.com/SwanseaUniversityDataScience/1492-LDCP.

https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi-org.ezproxy.u-pec.fr/10.1136/bmjopen-2024-093335

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    STRENGTHS AND LIMITATIONS OF THIS STUDY

    • The national-scale population-based cohort and long follow-up maximised the generalisability of our finding.

    • By incorporating both inpatient and outpatient data, the study not only expanded the sample size but also enabled a thorough analysis of liver disease incidence, enhancing understanding of its clinical implications.

    • With the usage of the Secure Anonymised Information Linkage Databank as the research-ready data asset and the incorporation of reproducible research pipelines, this study strengthened its reliability and reproducibility.

    • Sensitive Read codes for conditions like hepatitis B virus and hepatitis C virus could not be included due to regulatory laws, impacting the completeness and precision of the data analysis.

    • The study only included individuals with a history of general practitioner registration and residency information in Wales, likely overlooking a subpopulation at a significant risk for liver disease.

    Background

    Liver disease is a significant global public health issue and a major contributor to morbidity and mortality.1 Globally, cirrhosis and hepatocellular carcinoma account for an estimated 2 million deaths every year.2 In the UK, liver disease has become the third most common cause of premature death,3 despite mortality rates for other major non-communicable diseases declining.4

    The management of chronic liver disease frequently involves lifestyle modification, including weight loss and reduced alcohol use, with the goal of reversing factors that can lead to disease progression.5 As a result, public health policies in the UK have focused on prevention and early detection, including the UK Government’s Prevention Green Paper to promote disease prevention,6 the National Institute for Health and Care Excellence guidance that focused on tackling obesity7 and a series of policies to reduce alcohol-related harm.8 In November 2022, the Welsh Government published a Quality Statement for Liver Disease,9 underlying the importance of the awareness of risk factors and early detection of liver disease and set out its plans to promote the delivery of better quality, higher value and more accessible services for individuals with liver disease.

    To effectively improve liver disease management, clinicians, researchers and policy makers must be aware of the epidemiology and clinical profile of individuals with liver disease. However, a significant gap remains in the integration of primary, secondary and mortality data, particularly across the different liver disease stages and aetiologies. To date, much of the epidemiological data is based on the International Classification of Diseases version 10 (ICD-10) coding, which is largely derived from secondary care data sources and therefore likely to underestimate the real-world incidence and prevalence of liver disease.10 11 This limitation currently precludes adequate prioritisation of research, targeting of interventions and recruitment of individuals to clinical trials. Understanding the clinical and sociodemographic features associated with liver disease in a large-scale population is essential for the improvement of disease prevention and treatment. Therefore, we aim to fill this gap by integrating primary, secondary and mortality data, providing a more comprehensive and accurate depiction of liver disease across different stages and aetiologies.

    The objective of our study is to describe the incidence of liver disease, as well as the key demographic and socioeconomic characteristics and the associated comorbidity of liver disease patients in Wales, as a first step towards improving capacity and capability for liver disease research.

    Materials and method

    Setting and data source

    This study is designed as a retrospective cohort study. We used data from the Secure Anonymised Information Linkage (SAIL) Databank, which contains anonymised, individual-level linked electronic health record data for the Welsh population.12–15 The SAIL Databank includes complete secondary care data and primary care data covering approximately 86% of the Welsh population. These data reflect the demographic diversity of Wales across age, sex and levels of deprivation16 and can be generalisable to the broader UK population due to demographic similarities.17 SAIL employs a split-file anonymisation process using National Health Service number, name, sex, date of birth and postcode, ensuring confidentiality while enabling the linkage of individual-level data sources.12 13 16

    To provide a comprehensive overview of liver disease epidemiology, we combined linked primary care, hospital admissions and mortality data. Primary care data were accessed from the Welsh Longitudinal General Practice (WLGP) data, which currently uses the Read Codes V.2 clinical coding system and collects event histories for people registered with a SAIL-supplying general practice (GP) in Wales. Hospital admission data, including inpatient admissions (emergency, elective and maternity) and daycare procedures, were collected from the Patient Episode Database for Wales (PEDW). Mortality data came from the Annual District Death Extract (ADDE) by the Office for National Statistics (ONS) death and contains the cause of death and contributory comorbidities. Both hospital admission and mortality data were coded using the ICD-10 system. We derived demographic and deprivation data from the Welsh Demographic Service Dataset and used the Welsh Index of Multiple Deprivation (WIMD) 2019 quintiles to measure relative area-level deprivation based on geographical residential location from the 2011 Lower-layer Super Output Areas (LSOAs).

    Study population

    We linked data from all individuals in the WLGP, PEDW and ADDE data within the SAIL Databank uses a unique anonymised individual identifier known as an Anonymised Linkage Field (ALF). Individuals were extracted based on ALF and filtered for good data linkage status based on existing methodology.12 13 We excluded individuals not registered with a SAIL-contributing GP or lacking residency information and identified those who received a liver disease diagnosis from 1 January 2004 to 31 December 2022. The first liver disease diagnosis across all data sources is considered to be the index liver disease event. Individuals were required to be residents of Wales at the time of cohort entry. GP registration was required if the index event was from WLGP. All individuals were followed until the earliest of GP deregistration (applicable only to individuals whose index event was identified from WLGP), moving out of Wales, death or the end of the study period (31 December 2022). GP deregistration is defined as the termination of an individual’s registration with a SAIL-contributing GP, as indicated by the last end date in the registration records. We divided the full cohort into four distinct groups: individuals identified in WLGP were assigned to the primary care group; those identified in PEDW were assigned to the secondary care group; individuals identified in ADDE were assigned to the mortality group and individuals identified across multiple data sources (two or more) were assigned to a multisource group.

    Measurements

    Definition of stages of liver disease and time of cohort entry

    We applied a hierarchy of three tiers of potential aetiological diagnoses and five discrete stages of chronic liver disease based on perceived clinical importance and natural history of liver disease progression as was described in our previous study.18 The first stage is the underlying aetiological conditions without the presentation of cirrhosis and was divided into three tiers. Tier 1 aetiologies of liver disease include alcohol-related liver disease (ArLD), non-alcoholic fatty liver disease (NAFLD), metabolic liver disease, hepatitis B virus (HBV), hepatitis C virus (HCV), autoimmune liver disease, haemochromatosis; tier 2 aetiologies include unspecified hepatitis, congestive hepatopathy and toxic liver disease and tier 3 aetiologies were other miscellaneous diagnoses. As the disease progresses, various stages of liver disease are defined as follows: hepatic fibrosis and cirrhosis were categorised as stage 2, portal hypertension as stage 3, hepatic decompensation as stage 4 and hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC) and other primary liver cancers were classified under stage 5. Acute liver diseases were analysed separately from chronic liver diseases as they carry different challenges for primary and secondary care. Acute liver diseases were defined as conditions including acute viral hepatitis, Budd-Chiari syndrome, acute liver failure, infections and infarctions and other unspecified acute liver injuries. A list of ICD-10 and Read V.2 codes to identify individuals with liver disease can be found in the online supplemental table 1.

    We defined the time of the first chronic or acute liver disease diagnosis across all data sources as the index date. To assess the proportion of individuals with liver disease who presented late, the diagnoses were sequenced according to the natural history of the disease: aetiological diagnoses (stage 1), followed by cirrhosis, portal hypertension, decompensation and then HCC formation18 (online supplemental figure 1).

    Comorbidities and drug prescription history

    We collected data on demographic characteristics (age, sex and WIMD 2019 quintile), with age divided by birth year in 10-year intervals, except for the 0–17 age group and 18–29 age group. Comorbidities including cardiovascular disease (CVD)-related conditions (heart failure, transient ischaemic attack, other ischaemic disease, atrial fibrillation, peripheral vascular disease, angina, and stroke), diabetes, hypertension and antihypertensive usage were tracked up to 10 years prior to cohort entry. Lists of ICD-10 and Read V.2 codes to identify the associated conditions are available as supplemental tables (online supplemental tables 2-1 and 2-2).

    Incidence of liver disease

    The incidence rate was calculated as the number of incident cases per Welsh residents. An incident case was defined as an individual having a first liver-related diagnosis during the study period (1 January 2004–31 December 2022) with no prior liver disease history from 1 January 1994 to 31 December 2003. To determine the most clinically significant aetiology, we only considered the most advanced stage of diagnosis on the incident event date. The Welsh population data were obtained from the ONS population estimates.19 The incidence rate was directly standardised using the European Union standard population 2013 and were reported per 100 000 inhabitants in each calendar year from 2004 to 2022, with corresponding 95% CIs. Analysis was repeated for different data sources, liver disease aetiologies and stages.

    Study findings were reported in accordance with the applicable reporting guidelines for observational studies using administrative data (Strengthening the Reporting of Observational Studies in Epidemiology20 and REporting of studies Conducted using Observational Routinely-collected Data.21 Data cleaning, cohort assembly and statistical analyses were performed using Structured Query Language (IBM Db2 V.11.1) and R (V.4.1.0–V.4.1.3) within the SAIL Databank privacy-protecting trusted research environment.

    Patient and public involvement

    The Liver Research Cymru Patient Advisory group was established to provide patient and public involvement (PPI) input into the development of research in Wales. The research questions and methodology addressed in this body of work were developed following the discussion of their priorities based on their lived experience of liver disease. The results were shared with this group, and their feedback was sought in the interpretation and presentation of these data.

    Results

    Demographic and socioeconomic characteristics

    We identified a total of 111 098 individuals with liver disease from all three data sources from 1 January 2004 to 31 December 2022, contributing 441 885 person-years of follow-up (figure 1A). Of the eligible individuals, 57 491 (51.7 %) were male. The median (IQR) age was 59 (47–72). Approximately 50% of the individuals (55 787) entered the cohort after 2016, and 27 178 (24.5 %) individuals came from the most deprived areas. Of the 111 098 individuals, 20 319 (18.3%) individuals could be tracked from WLGP (primary care data source), forming the primary care group, 56 710 (51.0 %) from PEDW (secondary data source) constituted the secondary care group and 3459 (3.1 %) from ADDE (mortality data source) formed the mortality group. Additionally, 30 610 (27.5%) appeared in multiple sources, forming the multi-source group (figure 1B). The demographic variation is notable with a higher percentage of male and elderly individuals, and a higher degree of socioeconomic deprivation in the mortality group (table 1).

    Figure 1
    Figure 1

    Flowchart of study population selection (A) and Venn diagram of study cohort composition (B) ADDE, Annual District Death Extract; PEDW, Patient Episode Database for Wales; WLGP, Welsh Longitudinal General Practice.

    View this table:
    Table 1

    Demographic and socioeconomic characteristics of liver disease patients

    Incidence of chronic liver disease

    We observed an increasing trend of liver disease from 2004 to 2022 in the Welsh population. The age-sex standardised incidence rate increased by three times during the 18 years of follow-up (97.7 per 100 000 inhabitants in 2004; 316.2 per 100 000 inhabitants in 2022, table 2). This was significantly contributed by a 6.5 times increase in the incidence rate of the primary care group, followed by 4.4 times increase in the secondary care group. The proportion of incident cases in primary care and secondary care groups increased by 14%–16% during the past two decades (figure 2A–C, online supplemental table 3-1).

    View this table:
    Table 2

    Crude and standardised incidence of chronic liver disease from 2004 to 2022

    Figure 2
    Figure 2

    Changes in standardised incidence per 100 000 inhabitants by data source (2004–2022). Bar chart presenting the changes in annual European age-sex standardised incidence rate and percentages by data sources. The x-axis represents time in years, and the y-axis represents the incidence rate in (A), percentages in (B). Line chart presenting the trends in annual European age-sex standardised incidence rate. The x-axis represents time in years, and the y-axis represents incidence rate in (C).

    The incidence of liver disease has risen across all stages, with stage 2 (cirrhosis) and stage 3 diagnoses (portal hypertension) experiencing the most notable increase from 2004 to 2022 (stage 2 by four times; stage 3 by 3.9 times). However, the increases in the higher-stage liver disease cases were less pronounced, particularly for stage 4 (1.5 times). Consequently, the proportion of more advanced liver disease diagnoses decreased (stage 4 and stage 5: 8.6% in 2004, 5.3% in 2022, decreased by 1.62 times since 2004, figure 3A–C, online supplemental table 3-2).

    Figure 3
    Figure 3

    Changes in standardised incidence per 100 000 inhabitants by stages (2004–2022). Bar chart presenting the changes in annual European age-sex standardised incidence rate and percentages by stages. The x-axis represents time in years, and the y-axis represents the incidence rate in (A), percentages in (B). Line chart presenting the trends in annual European age-sex standardised incidence rate. The x-axis represents time in years, and the y-axis represents the incidence rate in (C).

    We further analysed the stage 1 aetiologies by each diagnosis. We observed the most notable increase in the incidence of NAFLD (10.8 per 100 000 inhabitants in 2004, 124.5 per 100 000 inhabitants in 2022, increased 11 times since 2004), resulting in a 3.4 time increase in the proportion and ultimately accounting for over half of the cases by the end of our follow-up. Conversely, the proportion of ArLD cases decreased by 71%, given that the incidence rate remained unchanged during the past two decades (figure 4A,B, online supplemental table 3-3). A breakdown of NAFLD incidence by data sources demonstrated 14 and 18 times increases in the primary care and the secondary care group, respectively (online supplemental table 4-1). In contrast, the increase in the incidence of ArLD in the primary care group was more than twice that of the secondary care group (online supplemental table 4-2).

    Figure 4
    Figure 4

    Changes in standardised incidence per 100 000 inhabitants by liver disease aetiologies (2004–2022). Bar chart presenting the changes in annual European age-sex standardised incidence rate and percentages by aetiologies. The x-axis represents time in years, and the y-axis represents the incidence rate in (A), percentages in (B). Line chart presenting the trends in annual European age-sex standardised incidence rate. The x-axis represents time in years, and the y-axis represents the incidence rate in (C). ArLD, alcohol-related liver disease; HBV, hepatitis B virus; HCV, hepatitis C virus; NAFLD, non-alcoholic fatty liver disease. * ArLD overlap: if an individual had two or more diagnoses from liver disease aetiologies (ArLD, NAFLD, HBV, HCV, metabolic, haemochromatosis and autoimmune liver diseases) and one of them was ArLD; non-ArLD overlap: if an individual had two or more diagnoses from liver disease aetiologies but none of them was ArLD.

    We did not observe an increasing trend for HBV and HCV. Both conditions fluctuated during the past two decades, with HBV cases increasing by 4.3 times and HCV cases 2.3 times from 2004 to 2017, followed by a decrease from 2018 to 2022 (HBV by 1.7 times; HCV by 2.6 times). Consequently, the proportion of HBV and HCV cases decreased from 2004 to 2022 (HBV by 2.3 times; HCV by 8.6 times, figure 4A–C, online supplemental table 3-3).

    Underlying liver disease conditions

    Of the 111 098 individuals with liver disease, 79 992 (72.0 %) entered the cohort due to a tier 1 aetiology. This was mainly contributed by the high prevalence of NAFLD (33 655 (30.3%)). Most NAFLD cases were identified in the primary care group (13 565 (66.8%)) and the least NAFLD cases in the mortality group (727 (21.0%)). This aligns with the distribution of individuals with tier 1 aetiology, whose proportion was the highest in the primary care group (18 453 (90.8%)) and the lowest in the mortality group (1728 (50.0%)). In contrast, tier 2 and tier 3 diagnoses were higher in the secondary care group. The proportion of advanced liver disease stages was lower in the primary care group (stage 3: 166 (0.8%), stage 4: 53 (0.3%), stage 5: 121 (0.6%)), but higher in the mortality group (stage 4: 276 (8.0%), stage 5: 473 (13.7%)) (table 3).

    View this table:
    Table 3

    The underlying liver disease conditions at cohort entry

    Comorbidities associated with liver disease

    Among all eligible individuals with liver disease, 40 427 (36.4%) individuals had onset hypertension or initiated an antihypertensive medication, followed by 8923 (8.0%) individuals who had CVD-related conditions and 7658 (6.9%) had diabetes. We observed the higher proportions of comorbidities in the mortality group (CVD: 11.2%; diabetes: 8.1%; hypertension/antihypertensive: 39.7%) and the lowest in the primary care group (CVD: 4.2%; diabetes: 6.3%; hypertension/antihypertensive: 35.7%) (table 4). The burden of comorbidities was the lowest for individuals with stage 1 liver disease conditions but higher for late-stage (CVD highest for stage 4: 14.2%; hypertension/antihypertensive for stage 5: 45.1%) and stage 2 individuals (diabetes: 13.4%) (online supplemental table 5-1). Among the individuals who entered the cohort with a stage 1 aetiology, individuals with congestive hepatopathy had the highest rates of hypertension/antihypertensive (47.6%) and CVD conditions (23.6%), while those with NAFLD had the highest rates of diabetes (8.4%) (online supplemental table 5-2).

    View this table:
    Table 4

    Comorbidities associated with liver disease

    Discussion

    In this large-scale population-based study, we observed that the incidence of liver disease had increased dramatically in Wales during the past two decades. Notably, NAFLD played a significant role in this rise, with its incidence and proportion increasing sharply.

    The rapid increase in the incidence rate of liver disease observed in our study is similar to our previous findings in Wales,18 where a 3.6-fold increase was observed in the inpatient chronic liver disease cases between 2001 and 2019, aligning with the trends presented in our secondary care data. This was mainly driven by the 11-fold increase in NAFLD incidence. Other studies, such as P Nasr et al’s work in Sweden,22 reported a twofold increase in the incidence rate of NAFLD during 2005–2019. Similarly, H Tian et al23 used data from the Global Burden of Disease study 2019 and showed an increase of 95.4% in NAFLD globally from 1990 to 2019. A systematic review that included 578 studies demonstrated a 13% higher prevalence of NAFLD during the year 2011–2021 compared with year 2000–2010.24 A community study in the USA reported a 5-fold increase in the NAFLD incidence from 1997 to 2014.25 However, none of these findings compare to our observation in Wales. This substantial difference may be partly attributed to our inclusion of primary care data, which experienced more rapid increase than inpatient data and thus boosted the NAFLD incidence across the study population. The high prevalence of obesity26 and rapidly increasing incidence of diabetes27 in Wales in the recent decades may also have attributed to the sharp increase of NAFLD, as these metabolic factors including obesity28 29 and insulin resistance were considered to be strongly linked to NAFLD due to metabolic dysfunctions.30 31 This observation is further supported by our observation of autoimmune liver disease, which is not linked to metabolic factors and remained unchanged during our study period, highlighting the lifestyle-related nature of the increase in NAFLD. Another interesting observation related to NAFLD is the high proportion of those with comorbidities. NAFLD evolves in those with metabolic syndrome and these comorbidities are used to define the more recently adopted term metabolic-associated fatty liver disease.32 In keeping with this definition of NAFLD, we observed a high proportion of individuals having onset comorbidities such as hypertension, diabetes, CVD-related conditions up to 10 years prior to their first liver disease diagnosis. The multimorbidity among people living with liver disease is anticipated considering the increasing incidence of NAFLD and the high prevalence of obesity in Wales, and this will require engagement with primary care practitioners to address their complex multidisciplinary healthcare needs and identify those at risk of significant liver disease early in the disease trajectory.33

    Contrary to previous European studies showing a decline in ArLD incidence,18 34 our research observed a resurgence after 2015 following a slight decrease from 2004 to 2014. We believe that changes in drinking habits35 36 and government actions on alcohol pricing and taxation37 likely influenced the declining trend before 2015, while the resurgence after 2015 was contributed by the inclusion of outpatient data from the primary care group. The addition of primary care data could explain the discrepancy in our results compared with the prior studies that focused solely on inpatients, filling the gap by providing a broader perspective by including outpatient data.18 34 Similar to ArLD, the incidence of HBV increased until 2017, then sharply dropped after the year of 2018. The observed decrease in the incidence of HBV and HCV post-2020 could be attributed to the initiatives outlined in the Welsh Health Circular, which seeks to meet the goal of eradicating HBV and HCV as a major public health concern.38 This trend may represent the effort in better case finding and eradication, which led to reduced transmission for HCV and enhanced vaccination and viral suppression for HBV.39 Additionally, disruptions caused by the COVID-19 pandemic may also have played a role in this decline.40

    It is worth to mention our observation on the increasing number of incident cases identified from primary care data, particularly after 2015. This observation, together with the rapid increase in NAFLD and ArLD incidence, indicated an increasing detection rate of liver disease in primary care at an early stage. This change might be attributed to the implementation of the Wales Liver Disease Delivery Plan, which emphasised the importance of early detection and ensured that ‘excellent care’ is accessible when necessary.41 Another notable trend observed during our study period is the decreasing proportion of severe late-stage presentations, such as hepatic decompensation. Additionally, we observed a declining trend in the proportion of individuals who died from liver disease without receiving a prior diagnosis. This is to say, despite the increasing incidence of liver disease, there is a promising indication that more individuals were identified as the result of early detection. This provides the healthcare system with more opportunities for early-stage intervention. Early detection of liver disease has been linked to improved long-term outcomes, and a series of reviews and guidelines42 43 emphasise the importance of thorough evaluation and timely referral for patients with abnormal liver function tests. Despite these recommendations, adherence remains inconsistent, and the factors influencing clinicians’ referral practices—including potential barriers and motivating factors—are not well understood. This issue is especially pressing in the context of rising obesity rates and the emergence of new weightloss therapies, both of which carry significant implications for liver health. Consequently, further research is essential to assess how integrated care models and early detection strategies might reduce liver disease progression and mortality.

    Our study has several strengths. First, it used the national-scale setting and created a population-based cohort with long-term follow-up, which allowed us to maximise the generalisability of our findings to the wider population. Furthermore, by combining primary care, secondary care and mortality data, our study uniquely expands on previous research and addresses a significant gap in understanding liver disease.10 11 This novel approach increased our sample size by at least 20%, allowing us to provide a more comprehensive view of liver disease incidence and characteristics. Specifically, this integrated data approach enabled us to (1) capture a broader range of liver disease cases, reducing the likelihood of underestimating case numbers and ensuring a more accurate representation of disease incidence and (2) include patients from diverse healthcare settings, each with unique clinical characteristics, which allowed for a detailed comparison across these settings and a deeper understanding of liver disease in varied clinical contexts. Additionally, a significant strength of the study is the consistency of coding practices in Wales, which have not changed in response to funding incentives.44 This stability helps mitigate potential biases and misdiagnoses, providing a more reliable and objective reflection of disease incidence and management trends. Another strength lies in the reproducibility and the use of the SAIL Databank as a research-ready data asset. By incorporating reproducible research pipelines, we standardised and documented the cohort curation process. This enhances the study’s reliability and reproducibility, enabling other researchers to replicate our findings and explore further research questions within the same framework.

    Our study is also vulnerable to several limitations. Regulatory laws associated with the SAIL Databank prohibited the inclusion of conditions classified as sensitive in our analysis. For instance, the sensitive Read codes for HBV and HCV could not be extracted along with the rest of the cohort, resulting in potentially incomplete data for these conditions, affecting the completeness and precision of our data analysis. Our study exclusively included individuals who had a history of GP registration and residency information in Wales. Consequently, we likely overlooked individuals who were unable to register with a GP, a subpopulation that represents a group at significant risk for liver disease. The exclusion may also introduce a potential underestimation of the true incidence rate of liver disease in the primary care data. Moreover, there may be changes in diagnostic practices and healthcare-seeking behaviour over the study period that could influence the observed trends and introduce variations, potentially affecting the accuracy of our findings regarding liver disease incidence and management. Furthermore, due to the observational nature and design of our study, we were unable to establish causal relationships between the onset of liver disease and other comorbidities.

    Conclusion

    Our study observed a significant rise in the incidence of liver disease in Wales over the past two decades, primarily driven by the increase in NAFLD. The high prevalence of comorbid conditions among liver disease patients and the increased role of primary care in disease identification highlight the need for integrated healthcare approaches to address this growing public health concern. A better understanding of the incidence of liver disease, driven by a more comprehensive analysis of the integrated primary and secondary care data, is essential as a foundation for effective prevention, early detection and targeted interventions to improve patient outcomes.

    Data availability statement

    Data may be obtained from a third party and are not publicly available. The routine data used in this study are available in the SAIL Databank at Swansea University, Swansea, UK. All proposals to use SAIL data are subject to review by an Information Governance Review Panel (IGRP). Before any data can be accessed, approval must be given by the IGRP. The IGRP gives careful consideration to each project to ensure proper and appropriate use of SAIL data. When access has been approved, it is gained through a privacy-protecting trusted research environment (TRE). SAIL has established an application process to be followed by anyone who would like to access data via SAIL https://www.saildatabank.com/application-process. This study has been approved by the IGRP as project 1492. The research adhered to ethical guidelines and the Data Protection Act 2018 to ensure the privacy and confidentiality of all data subjects involved. The reproducible SQL and R code and code lists to identify study individuals are available on Github: https://github.com/SwanseaUniversityDataScience/1492-LDCP.

    Ethics statements

    Patient consent for publication

    Ethics approval

    The study uses anonymised data only, and ethical approval was not required for this secondary use of data. See the following weblink: https://saildatabank.com/data/.

    Acknowledgments

    The Liver Disease Cymru Partnership (LDCP) received a grant from the National Institute for Health Research (NIHR154876). This project was partly funded by an unrestricted grant from the Liver Disease Implementation Group, Welsh Government (LDIG-22-19). We wish to acknowledge all members of the LDCP who have contributed to the collaborations: Professor Kerry Hood, Professor Deborah Fitzsimmons and Professor Katherine Cullen for advice developing the group and the Liver Research Cymru patient advisory group for their perspective on the research to be undertaken. This study uses anonymised data held in the Secure Anonymised Information Linkage (SAIL) Databank, which is part of the national population data research infrastructure for Wales. We would like to acknowledge all the data providers who make anonymised data available for research.

    References

    1. ↵
      1. Paik JM,
      2. Golabi P,
      3. Younossi Y, et al
      . Changes in the Global Burden of Chronic Liver Diseases From 2012 to 2017: The Growing Impact of NAFLD. Hepatology 2020;72:1605–16. doi:10.1002/hep.31173
      OpenUrlCrossRefPubMed
    2. ↵
      1. Wang H,
      2. Naghavi M,
      3. Allen C
      . Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 2016;388:1459–544. doi:10.1016/S0140-6736(16)31012-1
      OpenUrlCrossRefPubMed
    3. ↵
      1. Office for National Statistics
      . Deaths registered in england and wales (series dr): 2017. 2017. Available: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredinenglandandwalesseriesdr/2017 [Accessed 05 Jun 2024].
    4. ↵
      1. Williams R,
      2. Aspinall R,
      3. Bellis M, et al
      . Addressing liver disease in the UK: a blueprint for attaining excellence in health care and reducing premature mortality from lifestyle issues of excess consumption of alcohol, obesity, and viral hepatitis. Lancet 2014;384:1953–97. doi:10.1016/S0140-6736(14)61838-9
      OpenUrlCrossRefPubMed
    5. ↵
      1. Nobili V,
      2. Carter-Kent C,
      3. Feldstein AE
      . The role of lifestyle changes in the management of chronic liver disease. BMC Med 2011;9:70. doi:10.1186/1741-7015-9-70
    6. ↵
      1. Department of Health and Social Care
      . Advancing our health: prevention in the 2020s – consultation document. GOV.UK; 2019. Available: https://www.gov.uk/government/consultations/advancing-our-health-prevention-in-the-2020s/advancing-our-health-prevention-in-the-2020s-consultation-document [Accessed 05 Jun 2024].
    7. ↵
      1. NICE
      . Diet, nutrition and obesity. Available: https://www.nice.org.uk/guidance/lifestyle-and-wellbeing/diet--nutrition-and-obesity [Accessed 05 Jun 2024].
    8. ↵
      1. Burton R,
      2. Henn C,
      3. Lavoie D, et al
      . A rapid evidence review of the effectiveness and cost-effectiveness of alcohol control policies: an English perspective. The Lancet 2017;389:1558–80. doi:10.1016/S0140-6736(16)32420-5
      OpenUrl
    9. ↵
      1. Welsh Government
      . The quality statement for liver disease: the quality statement describes what good quality liver disease services should look like. 2022. Available: https://www.gov.wales/sites/default/files/pdf-versions/2023/3/1/1679326108/quality-statement-liver-disease.pdf [Accessed 05 Jun 2024].
    10. ↵
      1. Nam YH,
      2. Mendelsohn AB,
      3. Panozzo CA, et al
      . Health outcomes coding trends in the US Food and Drug Administration’s Sentinel System during transition to International Classification of Diseases-10 coding system: A brief review. Pharmacoepidemiol Drug Saf 2021;30:838–42. doi:10.1002/pds.5216
      OpenUrl
    11. ↵
      1. Ratib S,
      2. West J,
      3. Fleming KM
      . Liver cirrhosis in England-an observational study: are we measuring its burden occurrence correctly? BMJ Open 2017;7:e013752. doi:10.1136/bmjopen-2016-013752
    12. ↵
      1. Ford DV,
      2. Jones KH,
      3. Verplancke J-P, et al
      . The SAIL Databank: building a national architecture for e-health research and evaluation. BMC Health Serv Res 2009;9:157. doi:10.1186/1472-6963-9-157
    13. ↵
      1. Lyons RA,
      2. Jones KH,
      3. John G, et al
      . The SAIL databank: linking multiple health and social care datasets. BMC Med Inform Decis Mak 2009;9:3. doi:10.1186/1472-6947-9-3
    14. ↵
      1. Jones KH,
      2. Ford DV,
      3. Ellwood-Thompson S, et al
      . The UK Secure eResearch Platform for public health research: a case study. The Lancet 2016;388:S62. doi:10.1016/S0140-6736(16)32298-X
    15. ↵
      1. Wilkinson T,
      2. Schnier C,
      3. Bush K, et al
      . Drug prescriptions and dementia incidence: a medication-wide association study of 17000 dementia cases among half a million participants. J Epidemiol Community Health 2022;76:223–9. doi:10.1136/jech-2021-217090
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Schnier C,
      2. Wilkinson T,
      3. Akbari A, et al
      . The Secure Anonymised Information Linkage databank Dementia e-cohort (SAIL-DeC). Int J Popul Data Sci 2020;5:1121. doi:10.23889/ijpds.v5i1.1121
    17. ↵
      1. Office for National Statistics
      . Population estimates for the uk, england, wales, scotland, and northern ireland: mid-2022. 2024. Available: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/annualmidyearpopulationestimates/mid2022 [Accessed 05 Jun 2024].
    18. ↵
      1. Pembroke TPI,
      2. John G,
      3. Puyk B, et al
      . Rising incidence, progression and changing patterns of liver disease in Wales 1999-2019. World J Hepatol 2023;15:89–106. doi:10.4254/wjh.v15.i1.89
      OpenUrl
    19. ↵
      1. Office for National Statistics
      . Estimate of the population for england and wales. 2023. Available: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/estimatesofthepopulationforenglandandwales [Accessed 05 Jun 2024].
    20. ↵
      1. von Elm E,
      2. Altman DG,
      3. Egger M, et al
      . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies. PLoS Med 2007;4:e296. doi:10.1371/journal.pmed.0040296
    21. ↵
      1. Benchimol EI,
      2. Smeeth L,
      3. Guttmann A, et al
      . The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement. PLoS Med 2015;12:e1001885. doi:10.1371/journal.pmed.1001885
    22. ↵
      1. Nasr P,
      2. von Seth E,
      3. Mayerhofer R, et al
      . Incidence, prevalence and mortality of chronic liver diseases in Sweden between 2005 and 2019. Eur J Epidemiol 2023;38:973–84. doi:10.1007/s10654-023-01028-x
      OpenUrl
    23. ↵
      1. Tian H,
      2. Zhang K,
      3. Hui Z, et al
      . Global burden of non-alcoholic fatty liver disease in 204 countries and territories from 1990 to 2019. Clin Res Hepatol Gastroenterol 2023;47:102068. doi:10.1016/j.clinre.2022.102068
      OpenUrl
    24. ↵
      1. Liu J,
      2. Tian Y,
      3. Fu X, et al
      . Estimating global prevalence, incidence, and outcomes of non-alcoholic fatty liver disease from 2000 to 2021: systematic review and meta-analysis. Chin Med J 2022;135:1682–91. doi:10.1097/CM9.0000000000002277
      OpenUrl
    25. ↵
      1. Allen AM,
      2. Therneau TM,
      3. Larson JJ, et al
      . Nonalcoholic fatty liver disease incidence and impact on metabolic burden and death: A 20 year-community study. Hepatology 2018;67:1726–36. doi:10.1002/hep.29546
      OpenUrlCrossRefPubMed
    26. ↵
      1. Keaver L,
      2. Xu B,
      3. Jaccard A, et al
      . Morbid obesity in the UK: A modelling projection study to 2035. Scand J Public Health 2020;48:422–7. doi:10.1177/1403494818794814
      OpenUrlPubMed
    27. ↵
      1. Powell R
      . Diabetes prevalence – trends, risk factors, and 10-year projection. Public Health Wales; 2023. Available: https://phw.nhs.wales/services-and-teams/observatory/data-and-analysis/diabetes-prevalence-trends-risk-factors-and-10-year-projection [Accessed 05 Jun 2024].
    28. ↵
      1. Younossi Z,
      2. Anstee QM,
      3. Marietti M, et al
      . Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol 2018;15:11–20. doi:10.1038/nrgastro.2017.109
      OpenUrlCrossRefPubMed
    29. ↵
      1. Loomis AK,
      2. Kabadi S,
      3. Preiss D, et al
      . Body Mass Index and Risk of Nonalcoholic Fatty Liver Disease: Two Electronic Health Record Prospective Studies. The Journal of Clinical Endocrinology & Metabolism 2016;101:945–52. doi:10.1210/jc.2015-3444
      OpenUrl
    30. ↵
      1. Caussy C,
      2. Aubin A,
      3. Loomba R
      . The Relationship Between Type 2 Diabetes, NAFLD, and Cardiovascular Risk. Curr Diab Rep 2021;21:15. doi:10.1007/s11892-021-01383-7
    31. ↵
      1. Tanase DM,
      2. Gosav EM,
      3. Costea CF, et al
      . The Intricate Relationship between Type 2 Diabetes Mellitus (T2DM), Insulin Resistance (IR), and Nonalcoholic Fatty Liver Disease (NAFLD). J Diabetes Res 2020;2020:3920196. doi:10.1155/2020/3920196
    32. ↵
      1. Eslam M,
      2. Newsome PN,
      3. Sarin SK, et al
      . A new definition for metabolic dysfunction-associated fatty liver disease: An international expert consensus statement. J Hepatol 2020;73:202–9. doi:10.1016/j.jhep.2020.03.039
      OpenUrlCrossRefPubMed
    33. ↵
      1. Pryke R,
      2. Guha IN
      . Time to focus on chronic liver diseases in the community: A review of primary care hepatology tools, pathways of care and reimbursement mechanisms. J Hepatol 2023;78:663–71. doi:10.1016/j.jhep.2022.10.010
      OpenUrl
    34. ↵
      1. Deleuran T,
      2. Vilstrup H,
      3. Becker U, et al
      . Epidemiology of alcoholic liver disease in Denmark 2006-2011: a population-based study. Alcohol Alcohol 2015;50:352–7. doi:10.1093/alcalc/agv003
      OpenUrlCrossRefPubMed
    35. ↵
      1. Office for National Statistics
      . Adult drinking habits in great britain: 2017. 2018. Available: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/drugusealcoholandsmoking/bulletins/opinionsandlifestylesurveyadultdrinkinghabitsingreatbritain/2017 [Accessed 05 Jun 2024].
    36. ↵
      1. StatsWales
      . Adult lifestyles by year, 2016-17 to 2019-20. Available: https://statswales.gov.wales/Catalogue/National-Survey-for-Wales/Population-Health/Adult-Lifestyles/adultlifestyles-by-year [Accessed 05 Jun 2024].
    37. ↵
      1. Home Office
      . Alcohol strategy. GOV.UK; 2012. Available: https://www.gov.uk/government/publications/alcohol-strategy [Accessed 05 Jun 2024].
    38. ↵
      1. Welsh Health Circular
      . Attaining the who targets for eliminating hepatitis (b and c) as a significant threat to public health. welsh government. Welsh Government; 2017. Available: https://www.gov.wales/sites/default/files/publications/2019-07/attaining-the-who-targets-for-eliminating-hepatitis-b-and-c-as-a-significant-threat-to-public-health.pdf [Accessed 05 Jun 2024].
    39. ↵
      1. Public Health Wales
      . Prevention, diagnosis and treatment of blood borne viruses in wales: hepatitis b, hepatitis c and hiv annual report 2023 (data to end 2022). 2023. Available: https://phw.nhs.wales/publications/publications1/prevention-diagnosis-and-treatment-of-blood-borne-viruses-in-wales-hepatitis-b-hepatitis-c-and-hiv/ [accessed 05 Jun 2024]
    40. ↵
      1. Pley CM,
      2. McNaughton AL,
      3. Matthews PC, et al
      . The global impact of the COVID-19 pandemic on the prevention, diagnosis and treatment of hepatitis B virus (HBV) infection. BMJ Glob Health 2021;6:e004275. doi:10.1136/bmjgh-2020-004275pmid:33402334
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Welsh Government
      . Together for health -liver disease delivery plan: a delivery plan for nhs wales and its partners to 2020. 2015. Available: https://www.gov.wales/sites/default/files/publications/2018-12/liver-disease-delivery-plan-2015-to-2020.pdf [Accessed 05 Jun 2024].
    42. ↵
      1. Eslam M,
      2. Newsome PN,
      3. Sarin SK, et al
      . A new definition for metabolic dysfunction-associated fatty liver disease: An international expert consensus statement. J Hepatol 2020;73:202–9. doi:10.1016/j.jhep.2020.03.039pmid:32278004
      OpenUrlCrossRefPubMed
    43. ↵
      1. Pratt DS,
      2. Kaplan MM
      . Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000;342:1266–71. doi:10.1056/NEJM200004273421707pmid:10781624
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. Welsh Government
      . 2021-22 health board and public health wales nhs trust allocations. Cardiff Welsh Government; 2021. Available: https://www.gov.wales/2021-22-health-board-and-public-health-wales-nhs-trust-allocations.pdf [accessed 05 Nov 2024]

    Footnotes

    • X @AshleyAkbari, @harryahmed01

    • Contributors This project was a collaborative effort among multiple team members, each contributing significantly to various aspects of the research. TPIP, AA, AD, AY and HA acquired the funding. TPIP, AA, AD, AY, HA and JG contributed to the design of the methodology. TPIP, AA, AD, AY and JG supervised the study and validated the research. JG and AA were responsible for project administration and data management. JG curated the data, performed data analysis, created visual representations of the data and wrote the original draft of the paper. TPIP acted as guarantor. All authors reviewed and edited the manuscript. We would like to acknowledge the use of ChatGPT for grammar corrections and language enhancement in the manuscript, as English is not the first language of the primary author.

    • Funding The Liver Disease Cymru Partnership (LDCP) receives a grant from the National Institute for Health Research (NIHR154876). This project was partly funded by an unrestricted grant from the Liver Disease Implementation Group, Welsh Government (LDIG-22-19). This output will be accessible as Open Access, and the authors have applied a CC BY licence to any Author Accepted Manuscript (AAM) version arising from this submission.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research. Refer to the Methods section for further details.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.