Our four main areas of research are:

Using real world data held in electronic health records to create robust evidence that will inform clinical guideline development

Working directly with national guideline bodies, we identify areas of uncertainty in healthcare that have not been, or cannot be, addressed by gold standard randomised controlled trials. Using emulated target trial methodology and real world data, we aim to produce robust evidence that can be used in guideline development. We currently use UK primary care data from the Clinical Practice Research Datalink (embed link to CPRD website) and OpenSAFELY (embed link to OpenSAFELY website) along with linked hospital and mortality data. 

How evidence from real world data informs clinical guideline development

We are undertaking a programme of qualitative work to understand how real world data shapes the development of healthcare guidelines, how guidelines shape health practices, and how well real world data represents patient care and experience.

Using routinely collected data from electronic health records to evaluate implementation and impact of clinical guidelines

We work with national guideline bodies and conduct epidemiological studies to assess how guidelines are implemented using real world data from the UK. We also examine health outcomes following guidelines implementation, identify inequalities in implementation and missed opportunities for care, and the health economic impact of these.

Real-world evidence in health economic evaluations

We carry out work to understand how real-world evidence (RWE) can be used in health economic evaluations for medicines. This includes research into how RWE can contribute to the development of economic models and provide components for use in these models, and how RWE can be used to assess the effectiveness of medicines in clinical practice after they have been approved by NICE. 

We apply the target trial emulation framework in this research, using specific randomised efficacy trials as reference trials for emulated target trials. This approach can improve confidence in results obtained when extending the assessment of effectiveness to groups beyond the original trial population. These studies can also be used to learn how to assess and mitigate possible biases in emulated target trials.

Completed work

Impact of the COVID-19 Pandemic on treatment and outcomes among patients with heart failure with reduced ejection fraction: nationwide studies using the OpenSAFELY platform

We examined the impact of the COVID-19 pandemic on treatment and outcomes in patients with heart failure with reduced ejection fraction (HFrEF) in England, using routinely collected NHS data from over 220,000 individuals within the OpenSAFELY platform. We assessed prescribing of all four guideline-recommended therapies across pre-pandemic, pandemic and post-pandemic periods, alongside trends in hospitalisations and mortality.

Prescribing of recommended therapies increased steadily over time, including during the pandemic, but uptake remained low overall (15% by 2024). Inequities persisted, with lower prescribing in women, older adults and in some regions, though no clear disparities were observed by ethnicity or deprivation. Hospital admissions have returned to pre-pandemic levels, but mortality rates remain elevated in the most recent period.

These findings demonstrate progress in HFrEF treatment but highlight the need for ongoing monitoring and targeted interventions to address inequities and improve long-term outcomes.

Read the published work.

Model-based cost-effectiveness analysis of first-line pharmacotherapy combinations in adults with chronic heart failure and reduced ejection fraction

We collaborated with NICE on this study, which evaluated the cost-effectiveness of different first-line drug combinations for people with heart failure with reduced ejection fraction (HFrEF) in England, using a model developed to inform NICE guideline updates. The analysis compared five treatment strategies, ranging from traditional stepwise approaches to early initiation of combination therapy, incorporating evidence from clinical trials alongside real-world NHS data on hospitalisations and mortality.

We found that starting quadruple therapy early (combining an ACE inhibitor or ARNI, beta-blocker, mineralocorticoid receptor antagonist and SGLT2 inhibitor) was the most cost-effective approach, improving survival and reducing hospitalisations at an acceptable cost to the NHS. These findings support a shift away from stepwise treatment escalation towards earlier use of combination therapy in routine care.

INTEGRATE contributed real-world data from large-scale English electronic health records to inform baseline risks of hospitalisation and death, as well as treatment patterns over time. This ensured the model reflected contemporary NHS practice and strengthened its relevance for policy decisions.

These results directly informed NICE guideline development, providing evidence to support earlier use of combination therapy to improve outcomes and optimise healthcare resource use in people with HFrEF.

Read the full paper.

Ongoing work

Measuring inequalities in Type 2 diabetes and glucagon-like peptide-1 receptor agonist treatment in the England: a population based study

Treatments such as glucagon-like peptide-1 receptor agonists (GLP-1 RA) offer both glycaemic control and weight reduction and gaining approval in the UK in 2006. Following recommendations from the National Institute of Health and Care Excellence (NICE) in 2020, GLP-1 RAs are now being prescribed for weight management in non-diabetic individuals, after showing benefits outside the diabetic community. Obesity is a growing public health concern in the UK, being a known risk factor for type 2 diabetes.

The aim of this study is to understand the prevalence and disparities of type 2 diabetes in adults, and the subsequent trends in the prescribing of GLP-1 RA following NICE recommendations. This study will use data from the clinical practice research datalink (CPRD) to carry out a descriptive study to investigate the prevalence of type 2 diabetes, the characteristics among this group, and the characteristics for prescribing of GLP-1 RA, such as age, sex, ethnicity, region, deprivation and morbidities to understand their impact on the receipt of NICE recommended treatments. Descriptive statistics such as means, proportions and counts will be used to describe the study population, and further logistic regression analyses will be used to determine factors associated with GLP-1 RA prescription. Data from hospital episode statistics will be used to understand outcomes following GLP-1 RA prescription. This study will look over time, understanding the impact of guideline publication and health alerts on prescribing patterns.

In the UK, where the prevalence of diabetes is high, understanding the disparities among disease prevalence, and patterns of GLP-1 RA use is crucial in shaping future health policy. Disparities in access to these medications could exacerbate existing health inequalities. Investigating these patterns is essential to ensure equitable healthcare delivery and to inform policy decisions on the appropriate use of GLP-1 RA across different patient groups.

Estimating the real-world effectiveness of GLP-1 receptor agonists on renal outcomes in patients with type 2 diabetes: an emulated target trial study using UK electronic health records

The NHS faces significant pressures including increasing costs. A substantial proportion of annual growth in UK public health spending is due to cost pressures, including medical technology costs (1). It is important to assess both the clinical- and cost-effectiveness of medicines, and real-world evidence (RWE) can provide valuable contributions.

This study aims to estimate the effectiveness of GLP-1 receptor agonists (GLP1-RAs) at preventing kidney failure and cardiovascular death in adults with type 2 diabetes (T2D) in the UK. It will use the target trial emulation framework (2) and will emulate a hypothetical target trial informed by the population and design of the FLOW (Evaluate Renal Function with Semaglutide Once Weekly) clinical trial of semaglutide in patients with T2D and CKD (3), although it will not seek to exactly replicate FLOW. The hypothetical trial the source population will differ from FLOW as it will consist of patients receiving T2D treatment at a later stage and with higher

BMI, in line with NICE guidance (4).

The exposure of interest will be extended from semaglutide to all GLP1-RAs. Eligible patients who start GLP-1RA treatment will form an initiator group, and eligible patients who do not start treatment will be assigned to a comparator group. Eligibility and treatment status will be assessed at the start of 1-month periods throughout study time. Survival time analysis will be used to estimate the effect on kidney function and mortality outcomes, with propensity scores as a balancing score to adjust for confounding.

This study will also investigate whether the effectiveness of GLP-1RAs is subject to effect modification by concomitant use of sodium-glucose cotransporter-2 inhibitors (SGLT2is), and will estimate effectiveness in populations that were not included or under-represented in FLOW. Including in adults with T2D but without renal impairment, those with poorly controlled diabetes, and with heart failure.

References

1. Appleby J, Leng G, Marshall M. NHS funding for a secure future. BMJ. 2024 Mar 20;384:e079341. doi: 10.1136/bmj-2024-079341. PMID: 38508676.

2. Hernán MA, Robins JM. Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available. Am J Epidemiol. 2016 Apr 15;183(8):758-64. doi: 10.1093/aje/kwv254. Epub 2016 Mar 18. PMID: 26994063; PMCID: PMC4832051.

3. Perkovic V, Tuttle KR, Rossing P, Mahaffey KW, Mann JFE, Bakris G, Baeres FMM, Idorn T, Bosch-Traberg H, Lausvig NL, Pratley R; FLOW Trial Committees and Investigators. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes. N Engl J Med. 2024 Jul 11;391(2):109-121. doi: 10.1056/NEJMoa2403347. Epub 2024 May 24. PMID: 38785209.

4. NICE 2022. Type 2 diabetes in adults: management. NICE guideline NG28. Published: 02 December 2015. Last updated: 29 June 2022 https://www.nice.org.uk/guidance/ng28

Liver disease in England: patterns of burden, risk, and inequalities to guide prevention and policy

Metabolic dysfunction-associated steatotic liver disease (MASLD, formerly non-alcoholic fatty liver disease, NAFLD) is the most prevalent chronic liver disease in the UK, affecting ~25% of adults. A subset (~6%) develop metabolic dysfunction-associated steatohepatitis (MASH), which carries a higher risk of progression to fibrosis, cirrhosis, hepatocellular carcinoma (HCC), and liver failure. Disease progression is typically asymptomatic and slow, but incidence and premature mortality are rising. Between 2004–2022, MASLD diagnoses and advanced liver outcomes nearly tripled, and liver-related deaths increased by 40%. Economic costs are considerable, with diagnosed MASH estimated at £2.3–4.2 billion in 2018, alongside high treatment costs for HCC and cirrhosis.

To inform the NICE MASLD guideline committee, we will leverage linked primary care (CPRD), secondary care and mortality data to provide robust, contemporary estimates of MASLD burden, progression, comorbidities, and inequalities.

Our objectives are to: (1) estimate the incidence and prevalence of MASLD, MASH and other chronic liver disease in the general population; (2) describe fibrosis and advanced liver disease outcomes (cirrhosis, decompensation, hepatocellular carcinoma, liver-related and all-cause mortality, liver transplant) in population, MASLD, MASH and other chronic liver disease cohorts; (3) characterise prevalence of risk factors and combinations of risk factors, and assess associations with advanced outcomes using Cox and competing risks models; (4) examine adherence to testing and monitoring recommendations; (5) quantify the burden of comorbidities and impact on outcomes; and (6) estimate patient eligibility for emerging treatments under alternative screening strategies, including scenario modelling of risk-factor combinations.

Analyses will be stratified by age, sex, ethnicity, deprivation, region, and comorbidities to capture inequalities. Results will generate key descriptive parameters for economic models, support NICE appraisal of novel therapies, and inform updates to clinical guidelines. This work will provide an essential evidence base for reducing the burden and inequalities of advanced liver disease in England.