Explore the vast range of titles available.

Background Immortal time bias is common in observational studies but is typically described for pharmacoepidemiology studies where there is a delay between cohort entry and treatment initiation. Methods This study used the Clinical Practice Research Datalink (CPRD) and linked national mortality data in England from 2000 to 2019 to investigate immortal time bias for a specific life-long condition, intellectual disability. Life expectancy (Chiang’s abridged life table approach) was compared for 33,867 exposed and 980,586 unexposed individuals aged 10+ years using five methods: (1) treating immortal time as observation time; (2) excluding time before date of first exposure diagnosis; (3) matching cohort entry to first exposure diagnosis; (4) excluding time before proxy date of inputting first exposure diagnosis (by the physician); and (5) treating exposure as a time-dependent measure. Results When not considered in the design or analysis (Method 1), immortal time bias led to disproportionately high life expectancy for the exposed population during the first calendar period (additional years expected to live: 2000–2004: 65.6 [95% CI: 63.6,67.6]) compared to the later calendar periods (2005–2009: 59.9 [58.8,60.9]; 2010–2014: 58.0 [57.1,58.9]; 2015–2019: 58.2 [56.8,59.7]). Date of entry of diagnosis (Method 4) was unreliable in this CPRD cohort. The final methods (Method 2, 3 and 5) appeared to solve the main theoretical problem but residual bias may have remained. Conclusions We conclude that immortal time bias is a significant issue for studies of life-long conditions that use electronic health record data and requires careful consideration of how clinical diagnoses are entered onto electronic health record systems.

Obesity, depressive disorders and antidepressant drugs are associated with increased mortality, cardiovascular disease, diabetes, fractures and falls. We explored outcomes associated with the most commonly prescribed antidepressants in overweight or obese people with depression. We identified a cohort of overweight or obese adults (≥18 years) in primary care from the UK Clinical Practice Research Datalink, linked with hospital and mortality data, between 1 January 2000 and 31 December 2016 who developed incident depression to January 2019. Cox proportional hazards models and 99% confidence intervals were used to estimate hazard ratios (HR) for mortality, cardiovascular disease, diabetes, and falls/fractures associated with exposure to selective serotonin reuptake inhibitors (SSRIs), tricyclic (TCA)/other, combination antidepressants, citalopram, fluoxetine, sertraline, amitriptyline and mirtazapine, adjusting for potential confounding variables. In 519,513 adults, 32,350 (9.2 per 1,000 years) displayed incident depression and 21,436 (66.3%) were prescribed ≥1 antidepressant. Compared with no antidepressants, all antidepressant classes were associated with increased relative risks of cardiovascular disorders [SSRI HR: 1.32 (1.14-1.53), TCA/Other HR: 1.26 (1.01-1.58)], and diabetes (any type) [SSRI HR: 1.28 (1.10-1.49), TCA/Other: 1.52 (1.19-1.94)]. All commonly prescribed antidepressants except citalopram were associated with increased mortality compared with no antidepressants. However, prescription ≥1 year of ≥40mg citalopram was associated with increased mortality and falls/fractures and ≥1 year 100mg sertraline with increased falls/fractures. In overweight/obese people with depression, antidepressants may be overall and differentially associated with increased risks of some adverse outcomes. Further research is required to exclude indication bias and residual confounding.

BackgroundDetermining the most appropriate treatment in patients with cancer with an acute myocardial infarction (MI) can be challenging. Optimal management requires an understanding of bleeding risk which may be different in this population. This study aimed to investigate the bleeding risk among patients with MI with and without cancer, in the first and second year post-MI.MethodsPatients with MI, with and without cancer were identified from a national cardio-oncology database from England between 2006 and 2019. The outcome was a presentation to hospital for a major bleeding event, with patients followed for a maximum of 24 months. Inverse probability weighting was used to compare cancer and non-cancer cohorts in time-to-event analyses.Results587 279 patients with MI were identified, 9820 (1.7%) had cancer and 577 459 (98.3%) did not. Colorectal, prostate, breast, lung and bladder cancer were the most common types of cancer. The rate of hospital presentation for bleeding in the first year post-MI was higher in patients with cancer than the non-cancer reference population (HR: 1.53, 95% CI 1.45 to 1.62, p<0.001). 506 280 patients with MI were followed up in the second year post-MI. 5666 (1.1%) had cancer and 500 614 (98.9%) did not. The bleeding rate in patients with MI with cancer remained elevated in the second year post-MI (HR: 1.42, 95% CI 1.29 to 1.56, p<0.001). There were marked differences in bleeding between cancer types.ConclusionIn this real-world observational study, patients with cancer had an increased bleeding risk in the first year post-MI which decreased but persisted in the second year after MI. Bleeding risk in patients with cancer must be carefully assessed post-MI.

IntroductionCardiovascular disease and cancer are common causes of morbidity and mortality. Advancements in treatment strategies for both diseases have resulted in a growing population who live with both conditions. Myocardial infarction (MI) represents approximately 20% of all cardiovascular disease admissions in cancer patients and 10% of patients who present with an acute MI have cancer. Managing MI patients with cancer require careful balancing of their ischaemic and bleeding risks. While dual antiplatelet therapy (DAPT) increases a patient’s bleeding risk, cancer patients are at further increased risk due to a range of direct and indirect cancer effects. While our recent studies demonstrated an increased bleeding risk in cancer patients in the 1st year after MI (the period of intensive APT), it is currently not known if this risk is sustained beyond this period. VICORI is the world’s first whole-country cardio-oncology research platform, linking data from the National Cancer Registration and Analysis Service, National Institute for Cardiovascular Outcomes Research and Hospital Episode Statistics. We investigated the risk of bleeding among MI patients in the 2nd year post-MI, where a large majority of patients would have completed a course of DAPT, stratified by the presence or absence of cancer.MethodsIn this retrospective observational study, we investigated the risk of bleeding following an MI between 2006–2019, in patients with and without cancer. Cancer was defined as a diagnosis of cancer in the 1-year preceding MI. Patients who lived past the 1st year after MI were followed-up from 12 to 24 months after their MI. We used inverse probability weighting based on propensity scores to produce a balanced cohort of patients with and without cancer. Cox proportional hazard and flexible parametric modelling were used to investigate cancer as a predictor of bleeding. Subgroup analyses were performed on stent status, MI and cancer type.ResultsOf 506280 patients presenting with MI, 5666 (1.1%) had cancer while 500614 (98.9%) did not. Prostate (n=1377), colorectal (n=1024), lung (n=490), breast (n=448) and bladder (n=334) cancers were among the most common types of cancers. The rate (hazard) of bleeding in MI was higher in patients with cancer in the 2nd year post-MI (HR:1.42, 95%CI 1.29–1.56) compared to those without. From the flexible parametric survival analysis, the risk of bleeding in cancer patients in the 2nd year post-MI was lower compared to the 1st year but remained statistically significant compared to patients without cancer.ConclusionIn this large real-world study, cancer patients have a persistent, increased bleeding risk beyond the 1-year post-MI period, compared to those without cancer, and this risk differed by type of cancer. Further work is needed to identify specific patient characteristics in each patient sub-group which alters one’s ischaemic or bleeding risk so they can be balanced favourably with personalised strategies.Abstract 45 Figure 1Probability of bleeding and 95% CI by cancer status using flexible parametric modelling after inverse probability weighting in the first year after MIAbstract 45 Figure 2Probability of bleeding and 95% CI by cancer status using flexible parametric modelling after inverse probability weighting in the second year after MI. This only includes patients who survived past year 1 or not lost to follow upConflict of InterestNone

BackgroundAlthough evidence shows that patients with end stage renal disease (ESRD) experience a high symptom burden which impacts on quality of life (QoL), less is known about patients with earlier stages of chronic kidney disease (CKD). This study aimed to explore symptom burden and potential contributing factors in patients with CKD Stage 1-5 not requiring renal replacement therapy (RRT).MethodsPatients with CKD Stage 1-5 and not on RRT were asked to report their symptoms using the Leicester Uraemic Symptom Score (LUSS), a questionnaire which assesses the frequency and intrusiveness of 11 symptoms commonly reported by kidney patients.ResultsSymptoms were assessed in 283 CKD Stage 1-5 patients: 54% male, mean age 60.5 standard error± 1.0, mean eGFR 38ml/min/1.73m2. Some 96% (95% confidence interval 93.2–98.0) of participants reported experiencing at least one symptom, the median reported being six. Excessive tiredness (81%;76.0–85.6), sleep disturbance (70%;64.3–75.3) and pain in bones/joints (69%;63.4–74.6) were reported most commonly. Overall, few significant associations were found between biochemical markers of disease severity and symptom burden. Men tended to report fewer symptoms than women and South Asian patients often described experiencing symptoms with a greater severity. Older patients found musculoskeletal symptoms more intrusive whereas younger patients found reduced concentration more intrusive.ConclusionsOur findings suggest that patients with CKD stages 1–5 experience a multitude of symptoms that could potentially impact QoL. Using multidimensional tools like the LUSS, more exploration and focus could provide a greater opportunity for patient focussed symptom control from the earliest stages of CKD.

Background The aim of equality, equity, diversity, and inclusion (EDI) is to ensure fair treatment, equal opportunities, equitable outcomes, and representation. The NIHR Research Design Service (RDS) EDI toolkit ( https://www.rssleicesterresources.org.uk/edi-toolkit ) helps researchers embed EDI throughout their work. This study evaluated the applicability of the RDS EDI toolkit for statistical methodology research and proposed adaptations to enable statistical methodologists to embed EDI in their research. Methods A full-day meeting was held to consider how the RDS EDI toolkit could inform the inclusion of EDI principles in statistical methodology research. Twelve individuals attended from the University of Leicester and the NIHR Research Support Service (RSS) Hub delivered by the University of Leicester and Partners. At the meeting, definitions of statistical methodology research and EDI were agreed. The RDS EDI Toolkit was interrogated to identify relevant aspects and additional considerations for statistical methodology research. Results Overall, the RDS EDI toolkit was valuable for incorporating EDI in statistical methodology research. Five recommendations to supplement the toolkit are proposed to reflect specific EDI challenges for statistical methodology research. Statistical methodology researchers should: Perform formal assessments of the required resources for maximising EDI from the outset of statistical methodology research projects, including consideration of research team, training, patient and public involvement, and appropriate budgeting. Conduct prospective and retrospective context-specific evaluations of the impact of their methodological research on exacerbating or reducing inequalities. Evaluate the selection of data sets, work with multiple, diverse databases, or use data sets that have undergone equality impact assessments. Clearly communicate EDI assessments and limitations, including which data sets are used and their purpose, such as illustrating or comparing methods, informing simulations, or guiding clinical practice. Incorporate EDI in dissemination activities and advocate for EDI principles in the peer review process. Conclusions Embedding EDI principles throughout statistical methodology research will improve its relevance and quality, better serve the public, and build public trust. It is essential that statistical methodologists strive towards equity in all aspects of their work. This paper demonstrates the value of the NIHR RDS EDI toolkit for statistical methodology research and encourages methodologists to adopt the recommendations in this paper. Further extensions to this work are needed to seek the wider views and experiences of statistical methodologists and public contributors from diverse and under-represented groups.

IntroductionPeople with a learning disability face significant health and mortality inequalities as well as wider systemic inequities. Challenges in palliative and end of life care (PEOLC) include communication difficulties, lack of involvement in decision-making and multimorbidity. Early identification of PEOLC needs is challenging, impacting timely care planning. The study aims to (1) understand barriers and enablers to providing high-quality, accessible PEOLC for people with a learning disability, and identify effective service delivery models and interventions and (2) improve PEOLC quality and accessibility by developing robust guidance for health and social care services.Methods and analysisThis is a mixed-methods study guided by the NHS England 2021 Ambitions Framework and adopting the Social Model of Disability. There are four workstreams: (1) a retrospective cohort analysis of the Clinical Practice Research Datalink; (2) a rapid scoping review; (3) field work in four study sites across England, involving (a) interviews with senior leaders and commissioners (n=up to 16) and informal stakeholder engagement conversations; (b) ethnographic case studies with people with a learning disability at the end of life (n=up to 20) and retrospective case reviews of people with a learning disability who have died (n=up to 40), using family and staff interviews and (c) development and piloting of methods for enabling systematic identification of PEOLC need, using experience-based co-design and (4) patient and public involvement (PPI) activities and a co-production group of 10 people with a learning disability to support data analysis and outputs. Data will be analysed using adapted framework analysis methodology. This is an inclusive, co-produced study with significant involvement of advisors and researchers with a learning disability as part of the study team.Ethics and disseminationEthical approval has been obtained for workstreams 1, 3a and 3b. Significant attention has been paid to ensuring informed consent, making adjustments for capacity. Accessible information and consent forms will be used, involving consultees and adhering to the Mental Capacity Act for participants who lack capacity. Data security will follow General Data Protection Regulation rules. Dissemination will include patient exemplars, guidance and various resources, engaging stakeholders through multiple formats.Study registrationresearchregistry10500.

Background Life expectancy is a simple measure of assessing health differences between two or more populations but current life expectancy calculations are not reliable for small populations. A potential solution to this is to borrow strength from larger populations from the same source, but this has not formally been investigated. Methods Using data on 451,222 individuals from the Clinical Practice Research Datalink on the presence/absence of intellectual disability and type 2 diabetes mellitus, we compared stratified and combined flexible parametric models, and Chiang’s methods, for calculating life expectancy. Confidence intervals were calculated using the Delta method, Chiang’s adjusted life table approach and bootstrapping. Results The flexible parametric models allowed calculation of life expectancy by exact age and beyond traditional life expectancy age thresholds. The combined model that fit age interaction effects as a spline term provided less bias and greater statistical precision for small covariate subgroups by borrowing strength from the larger subgroups. However, careful consideration of the distribution of events in the smallest group was needed. Conclusions Life expectancy is a simple measure to compare health differences between populations. The use of combined flexible parametric methods to calculate life expectancy in small samples has shown promising results by allowing life expectancy to be modelled by exact age, greater statistical precision, less bias and prediction of different covariate patterns without stratification. We recommend further investigation of their application for both policymakers and researchers.

Background Patient and Public Involvement and Engagement (PPIE) is important to all aspects of health research. However, there are few examples of successful PPIE in statistical methodology research. One of the reasons for this relates to challenges in the identification of individuals interested in statistical methodology research projects, and ambiguities over the importance of PPIE to these projects. Methods This project was conducted between August 2022 and August 2023. The aim is to report the process of the development of an accessible animation to describe what statistical methodology is and the importance of PPIE in statistical methodology research projects. For this, we combined storyboarding and scriptwriting with feedback from PPIE members and researchers. Results After three stages that incorporated feedback from the relevant stakeholders, we produced a final animation about PPIE in statistical methodology. The resulting animation used minimal text, simple animation techniques and was of short duration (< 3 min) to optimise the communication of the key messages clearly and effectively. Conclusions The resulting animation provides a starting point for members of the public to learn about PPIE in statistical methodology research and for methodologists who wish to conduct PPIE. We recommend further work to explore ways in which members of the public can be more meaningfully involved in methodology research. Plain English summary Patient and public involvement and engagement (PPIE) is when members of the public are directly involved in carrying out research projects. This is important because we as researchers want to make sure we are focusing on what matters most to patients, so that the research has as large an impact as possible. PPIE has typically been used in more applied research projects, such as clinical trials, but is equally as important in statistical methodology research, where we focus on making sure the statistical tools that we use in the applied projects are as good as possible. The aim of this project was to create a short animation that helps to explain the importance of PPIE in statistical methodology research projects. Researchers sometimes incorrectly assume that PPIE is less important in these projects as this type of research has a less obvious benefit to patients. The animation helps to further explain these concepts. It describes what statistical methodology research is and why involving members of the public is still important. This paper explains the process of developing the animation, including receiving feedback from members of the public to make sure the animation is accessible to as many people as possible. The result is a short, 3-min animation that is free to view on the NIHR website. This can be used by other researchers to help them when recruiting members of the public to their research projects.

BackgroundThe epidemiology of autism in adults has relied on untested projectionsusing childhood research.AimsTo derive representative estimates of the prevalence of autism and keyassociations in adults of all ages and ability levels.MethodComparable clinical diagnostic assessments of 7274 Adult PsychiatricMorbidity Survey participants combined with a population case-registersurvey of 290 adults with intellectual disability.ResultsThe combined prevalence of autism in adults of all ages in England was11/1000 (95% CI 3–19/1000). It was higher in those with moderate toprofound intellectual disability (odds ratio (OR) = 63.5, 95% CI27.4–147.2). Male gender was a strong predictor of autism only in thosewith no or mild intellectual disability (adjusted OR = 8.5, 95% CI2.0–34.9; interaction with gender, P = 0.03).ConclusionsFew adults with autism have intellectual disability; however, autism ismore prevalent in this population. Autism measures may miss more womenwith autism.