There is increasing availability of routinely collected data, including linked and anonymised electronic health records, administrative data, and drug and disease registries. The EHR Research Group at LSHTM comprises a wide range of epidemiologists, statisticians, and clinicians with extensive experience analysing real-world data from around the world.
Our work covers a broad spectrum including disease aetiology, safety and effectiveness of medications, vaccines and surgery, environmental influences, and health services research. We have particular interest in developing and applying optimal research methods, including both observational and interventional approaches.
There is increasing availability of routinely collected data, including linked and anonymised electronic health records (EHRs). The overarching aim of the EHR Research Group is to capitalise on research opportunities offered by the availability of routinely collected data, bringing methodological rigour to provide real-world evidence for important questions regarding human health.
The EHR Research Group at LSHTM comprises a wide range of epidemiologists, statisticians, and clinicians with extensive experience analysing real-world data from around the world. We have a diverse programme of work using a range of anonymised data sources. These include the UK Clinical Practice Research Datalink (CPRD) based on primary care records, Hospital Episode Statistics, administrative data, drug and disease registries, and numerous international data sources.
Our work covers a broad spectrum including disease aetiology, safety and effectiveness of medications, vaccines and other medical products, environmental influences, and health services research. Disease areas are similarly diverse, with major interests in a range of both non-communicable and communicable diseases. We have particular interest in developing and applying optimal research methods, including both observational and interventional approaches.
Here is Liam Smeeth talking about why access to routinely collected health data is so important for the research community.
CPRD data can be accessed and used for studies by LSHTM staff. There is a fee associated with use of the data. If you are not a member of the EHR Research Group at LSHTM and are planning to use CPRD data, please refer to the Clinical Practice Research Datalink website for further details.
There is a monthly EHR Research Group scientific meeting hosted by the group at LSHTM – if you would like to attend this meeting please email firstname.lastname@example.org in order to be added to the distribution list. Please note that these meetings will be internal for the rest of 2020.
Our research is funded by a range of different government bodies, charities and companies. These include the Wellcome Trust, Medical Research Council, National Institute for Health Research, GSK, and the British Heart Foundation.
Cancer affects a huge number of people globally, and new treatments and prevention strategies are constantly being developed. Our research on cancer covers three main themes:
Long term health of cancer survivors
As the number of cancer-survivors increases, the long term health of these individuals is an increasing priority. The Beyond Cancer team, within the EHR Research Group, is examining the impact of cancer history and treatment on future risk of vascular diseases, mental health, quality of life, and other morbidities.
Medicines associated with cancer development
Research in this area aims to determine whether certain drugs have an impact on the risk of developing cancer. For example, our work has examined the impact of drugs such as angiotensin receptor blockers (used to lower blood pressure) on cancer, and PDE5 inhibitors on the risk of melanoma.
Cancer risk factors
Another strand of our work is to better characterise how potential risk factors for cancer are associated with specific cancers. A notable example was our work published in the lancet showing that BMI influences cancer risk in different ways for different cancer sites.
- Cardiometabolic disease
Cardio-metabolic disease is an umbrella term, and in our group it covers the entities of diabetes and cardiovascular disease. Given that many people in the UK, and indeed globally, are affected by cardio-metabolic disease, this research theme is very important from a public health point of view. Our group uses electronic health records to answer questions relating to cardio-metabolic disease that ultimately help to improve our understanding of disease processes and disease outcomes. This is beneficial to patients, care providers, policy makers and the public at large.
- Infections / Vaccines / Immunity
Infectious diseases remain a major cause of ill health among individuals in the UK and elsewhere, particularly among the very young, among older individuals and among those with underlying health conditions. An increasing number of these diseases can be prevented by vaccination, and vaccination programmes are key public health tools for reducing illness due to infectious disease.
We use anonymised electronic health records to look at a wide range of important public health questions relating to infectious diseases – to monitor how common these conditions are, to identify risk factors for developing them, and to determine their outcomes. We are also assessing the effectiveness and safety of preventative measures such as vaccination. Separately, we are interested in other immune-mediated conditions, such as autoimmune diseases.
- Maternal Health
The ability to conduct studies of maternal health using electronic health records depends crucially on identifying pregnancy episodes in the data source. While primary care records provide a rich source of maternity data, the precise timing of pregnancy can often be difficult to ascertain. To address this, we have been working with researchers from the Clinical Practice Research Datalink (CPRD) to develop a new pregnancy identification algorithm which makes optimal use of pregnancy-related records relating to the timing and duration of pregnancy, the type of outcome (live birth, stillbirth or early pregnancy loss) and additional features pertaining to the pregnancy. The output of this joint collaboration will be a pregnancy register which will be made available to CPRD users working in pregnancy-related research in due course. This work is part of the wider programme of research of the NIHR Health Protection Research Unit (HPRU) in Immunisation.
Examples of work we have undertaken or are now initiating in the areas of maternal health include:
- a study showing increased risk of pre-eclampsia following specific infections in pregnancy
- studies to investigate social factors associated with lower uptake of vaccines among pregnant women, and the safety of vaccines given in pregnancy
- studies of the effects of autoimmune diseases and their therapies on pregnancy outcomes
Issues such as missing data, measured and unmeasured confounding, measurement being connected to underlying health status, etc. mean that complex statistical techniques can be necessary to obtain reasonable conclusions about causation from routinely-collected data. Assessing how well these techniques work in practice and identifying optimal analysis approaches is critical to allow researchers to obtain robust and valid conclusions from their analyses. This group undertakes a range of methodological research aiming to provide practical guidance to applied researchers about how best to analyse data from electronic health records for the purpose of causal inference.
The effects of medications are not completely known when a drug is first licensed for use by the general population. Important side effects might only become apparent when the drug has been used in very large numbers of patients, or for prolonged periods of time. Some effects might not have been seen because they happen more often in people who are unlikely to be included in randomised trials, such as people who take a lot of other medication, or who have several illnesses at the same time. The natural spread of age and gender are not always evenly represented in trials, so information on drug effectiveness and side effects might be sparse for older age groups and, sometimes, women. Additionally, side effects of drugs such as heart attack and stroke are sometimes referred to as "rare" and the numbers of people in randomised trials are too small for these "rare" effects to show up.
In pharmacoepidemiology we study these effects in large groups of patients to give us a better idea of the overall balance of risks and benefits for a medication in the whole population. Not all effects are harmful; sometimes we might also be interested in finding out about the unintended benefits of a drug. Using primary care and linked electronic health records, we have been investigating these effects for many years, as well as developing better methodologies for this kind of study. Some of our past work includes examining the risk of stroke and myocardial infarction in patients with and without dementia who use anti-psychotics, the association between MMR vaccine and autism, co-prescribing of PPIs (drugs used to treat stomach acid) and drugs which prevent blood clots on risk of heart attack and stroke, and also the risk of cancer in users of angiotensin receptor blockers (drugs used to treat high blood pressure).
We have ongoing interests in improving methodology for pharmacoepidemiology, including:
- Self-controlled methods
- Propensity score methods
- Intervention studies using electronic health records
- Methods for studying drug effectiveness
Although we work on a broad range of drugs and diseases, some key areas of current focus are:
- Long term impact of treatment for cancer
- Treatment of high blood pressure
- Effects of medications on cardiovascular disease
- Effects of medications on renal function
- Our programme of work assessing the effectiveness and safety of vaccines is described in the Vaccines section
We offer training courses in pharmacoepidemiology:
- Renal disease
Our group is conducting a wide portfolio of research studies into aspects of kidney disease. We predominantly use Electronic Health Records and other sources of routine data. We are always interested in new collaborations and if you are interested in working with us please get in touch.
National Chronic Kidney Disease Audit
The National CKD Audit is being implemented in England and Wales to provide a comprehensive picture of management and outcomes for people with CKD stages 3-5 in the region. It aims to improve the quality of patient care and assesses each GP practice against NICE CKD quality guidelines and standards to encourage quality improvement.
Acute Kidney Injury
Funded by the Wellcome Trust, we are investigating the epidemiology of acute kidney injury and the role that comorbidities and drugs play in its development. We are contributing to national organisations and campaigns working to improve the care and outcome of people who develop AKI such as the NHS England ‘Think Kidneys’.
Psychiatric illness and kidney disease
We are studying the prevalence and outcomes of kidney disease among people with psychiatric illness with a particular focus on the side-effects of medications in this population.
Diabetes and kidney disease
Previous work looking at the association between CKD and infections among diabetics has led to a broader programme of work studying the effect of diabetes treatments in slowing the development of kidney disease.
General epidemiology of kidney disease
We are studying the impact of a range of risk factors and medications on the development and outcome of people with kidney disease using a number of routine data sources, including the UK renal registry.
Global epidemiology of kidney disease
In addition to our research using EHRs we have a number of national and international collaborations investigating kidney disease in a global health context.
Links to some key publications:
- Methodological challenges when carrying out research on CKD and AKI using routine electronic health records
- CKD and the risk of acute, community-acquired infections among older people with diabetes mellitus: a retrospective cohort study using electronic health records
- Are pre-existing markers of chronic kidney disease associated with short-term mortality following acute community-acquired pneumonia and sepsis? A cohort study among older people with diabetes using electronic health records
- Inpatient coronary angiography and revascularisation following non-ST-elevation acute coronary syndrome in patients with renal impairment: a cohort study using the Myocardial Ischaemia National Audit Project
- Incomplete reversibility of estimated glomerular filtration rate decline following tenofovir disoproxil fumarate exposure
- Association between glycemia and mortality in diabetic individuals on renal replacement therapy in the U.K
- Baseline kidney function as predictor of mortality and kidney disease progression in HIV-positive patients
- Comparison of CKD-EPI and MDRD to estimate baseline renal function in HIV-positive patients
- Disparities in testing for renal function in UK primary care: cross-sectional study
- Patterns and effects of missing comorbidity data for patients starting renal replacement therapy in England, Wales and Northern Ireland
- Outcomes in patients on home haemodialysis in England and Wales, 1997-2005: a comparative cohort analysis
- Erythropoiesis-stimulating agent dosing, haemoglobin and ferritin levels in UK haemodialysis patients 2005-13
Respiratory diseases are diseases that affect the airways, the bronchi and the lungs. They include acute infections, such as pneumonia and bronchitis, and chronic conditions such as asthma and chronic obstructive pulmonary disease.
Using large electronic databases, we study the long-term effects and outcomes of treatment for these chronic respiratory diseases. Asthma and COPD are very common illnesses, in which patients suffer from episodes of acute deteriorations of respiratory function termed exacerbations. There is increasing understanding that asthma and COPD are part of a spectrum of obstructive airway diseases consisting of complex subtypes which share certain characteristics (phenotypes) but differ in terms of prognosis.
Past work includes validation studies of COPD and acute COPD exacerbations in UK primary care electronic healthcare records and an ongoing validation study of asthma using the same data source. Other work focuses on the cardiovascular risk profile of COPD patients.
Currently, we are studying treatment patterns and control of different asthma phenotypes in the UK primary care electronic records.
Links to publications:
- The risk of myocardial infarction (MI) and death following MI in people with chronic obstructive pulmonary disease: A systematic review and meta-analysis
- Predicting mortality after acute coronary syndromes in people with chronic obstructive pulmonary disease
- COPD disease severity and the risk of venous thromboembolic events: a matched case-control study
- Changes in the incidence, prevalence and mortality of bronchiectasis in the UK from 2004 to 2013: a population-based cohort study
- Skin Disease
Our research addresses an important research gap, the causes, consequences and treatment of skin disease. Little is known about the epidemiology of skin diseases. Our programme of works uses routinely-collected health data to provide answers to important research questions. These data include Electronic Health Records, Administrative data and Registry data.
Our work includes research on:
Herpes zoster - Examples include:
- Risk factors for herpes zoster
- Risk factors for post-herpetic neuralgia
- The role of zoster in triggering acute cardiovascular events
- Effectiveness of the zoster vaccine in routine use
We have well-developed national and international collaborations, involving colleagues in Canada the Netherlands, Denmark, and the USA. We are increasingly using data from different countries concurrently to answer the same research questions. We are always interested in new collaborations and if you are interested in working with us please get in touch. For further information on Dermato-Epidemiology, please see the European Dermato-Epidemiology Network.
Eczema – Examples include:
- Validating the diagnosis of eczema in UK general practice data (collaboration with colleagues at UPENN and UCSF)
- Long-term adverse outcomes with eczema
Other skin diseases - Examples include:
- Role of drugs in risk of malignant melanoma
We have well-developed national and international collaborations, involving colleagues in Canada the Netherlands, Denmark, and the USA. We are increasingly using data from different countries concurrently to answer the same research questions. We are always interested in new collaborations and if you are interested in working with us please get in touch.
For further information on Dermato-Epidemiology, please see the European Dermato-Epidemiology Network.
Study findings suggest that this group will also have a greater chance of severe COVID-19 outcomes
There are nearly 2 million cancer survivors in England. Although, people under current treatment for cancer were recognised early on as being a high-risk group for severe COVID-19, it was unclear whether this much larger population of medium- to long-term cancer survivors might be considered vulnerable to severe disease. Researchers in the Beyond Cancer Team, part of the LSHTM Electronic Health Records Group, addressed this by looking at historical data on another similar epidemic respiratory virus, influenza.
The researchers compared the risk of influenza hospitalisation or death between cancer survivors and cancer free controls. The study included data from 108,215 cancer survivors and 523,641 people with no prior cancer. Survivors of blood cancers had substantially elevated risks of influenza hospitalisation or death persisting for at least 10 years after cancer diagnosis, while risk was doubled for survivors of other cancers for up to 5 years from diagnosis. Cancer survivors were also more likely to have other illnesses that have been associated with higher risk of severe COVID-19 disease, such as diabetes and heart disease.
Flu vaccination has been historically recommended for people with some active cancers or with current or recent cancer treatment. The results of this study suggest that a wider population of longer-term cancer survivors might need to be prioritised for vaccination against both influenza and COVID-19.
Read the full paper on EClinicalMedicine.
Nationwide observational study on pre-exposure use of hydroxychloroquine could help recruitment into ongoing clinical trials
Early in the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, it was suggested that hydroxychloroquine - a commonly used treatment for rheumatic diseases - might have benefits for the treatment and prevention of COVID-19. Hydroxychloroquine has since been investigated in several randomised clinical trials and observational studies. Most studies to date have shown no evidence of a benefit of hydroxychloroquine as a treatment for patients admitted to hospital who already have COVID-19.
A separate question remains: does routine ongoing use of hydroxychloroquine in people without coronavirus infection protect against new infections or severe COVID-19 outcomes? We studied a large number of people who were prescribed hydroxychloroquine as part of normal clinical care and followed them up to look for signals of benefit or harm in mortality from COVID-19 and other causes. We found no evidence of benefit or harm after adjusting for important differences between people with the same health conditions who were prescribed hydroxychloroquine or not.
The scientific community should endeavour to encourage recruitment to trials like COPCOV since, in the absence of a vaccine, there are no current preventative treatments available. This paper adds to evidence of scientific equipoise and can help with recruitment. The use of hydroxychloroquine for prevention of COVID-19 mortality outside trial settings is currently not justified.
Read the full paper on The Lancet Rheumatology.
Tune into the King’s Fund Digital Health and Care Conference, where our very own Rohini Mathur and Liam Smeeth, along with Ben Goldacre and Caroline Morton of the Oxford DataLab, and Chris Bates from the GP software company TPP discuss the development of OpenSAFELY – a data analytics platform, built to deliver large-scale COVID-19 research in just 42 days.
In this roundtable session, they will discuss the technical aspects of creating OpenSAFELY, the philosophy of open and collaborative working underpinning the platform, and early findings from their studies.
Full conference runs 2-5 November 2020. The team's session is 3 November 11am-12pm GMT (Session 5).
Link to sign up: https://www.kingsfund.org.uk/events/digital-health-and-care
We are pleased to invite applications for two PhD studentships in vaccine epidemiology, funded for 3 years by the UK National Institute for Health Research (NIHR), starting in January or April 2021.
Full details can be found by clicking here.
We are pleased to invite applications for two PhD studentships in pharmacoepidemiology, funded for 3.5 years by GlaxoSmithKline (GSK), starting in January or April 2021.
Full details can be found by clicking here.
New estimates suggest strategies to reduce the risk of cardiovascular disease are needed for the growing population of cancer survivors
Survivors from a wide range of cancers could experience increased risks of heart disease and blood circulation problems compared to those who have never had cancer, according to new estimates published in the Lancet.
In one of the largest studies of its kind, the research team, led by members of the EHR Research Group at the London School of Hygiene & Tropical Medicine, analysed the medical records of more than 630,000 people in the UK, including over 100,000 survivors of a range of cancers.
They found raised risks of blood clots forming in the veins among survivors of most cancers, ranging from approximately a two- to ten-fold increase in risk for different cancer sites. Risks decreased over time but were still elevated for more than five years after the cancer diagnosis. Survivors of half of the 20 cancers studied also faced increased risks of heart muscle damage, known as cardiomyopathy, and heart failure.
For example, survivors of blood, oesophagus, lung, kidney and ovarian cancers were more than one and a half times more likely to experience cardiomyopathy or heart failure than people with no prior cancer.
An increased risk of coronary artery disease and stroke was found in some cancer survivors, including those with prior blood cancers.
The researchers stress that for many people the overall risks will still be low, particularly for younger cancer survivors, but say the findings highlight the need for new strategies to prevent and manage heart and circulation problems in cancer survivors, including a need to raise awareness among GPs of the raised risks.
While the researchers could not definitively identify the causes of the increased risks, the team’s analysis suggests that exposure to cancer treatments such as chemotherapy is likely to play a key role.
Lead author Helen Strongman from the London School of Hygiene & Tropical Medicine said: “Over recent decades cancer treatment and management have improved substantially, resulting in a large and growing population of long-term cancer survivors. Around half of those diagnosed with cancer in developed countries are now expected to survive for more than 10 years. However, there are concerns that there may be increased long-term risks of cardiovascular disease following cancer diagnosis, driven by treatment side-effect effects and the potential impact of the cancer itself.”
While previous studies have demonstrated the short- to medium-term increased risk of cardiovascular disease from some specific cancer treatments, there have been limited data on the overall and long-term differences in cardiovascular risk between cancer survivors and those who have never had cancer.
To address this, the team brought together anonymised data from primary care, hospitals, cancer registries, and death certificates, to quantify the absolute and relative risks of a comprehensive range of cardiovascular diseases in survivors of the 20 most common adult cancers1, compared with cancer-free general population controls. The research also accounted for other risk factors for cardiovascular disease, such as older age, smoking and body mass index.
The team found large increases in risk of venous thromboembolism (blood clots in the veins) among survivors of 18 of the 20 cancers studied compared to cancer-free controls. Survivors of breast cancer had double the risk of venous blood clots one year after diagnosis compared with similar women with no prior cancer, though the overall level of risk remained low, especially for younger survivors.
For female breast cancer survivors aged under 60 years, six extra blood clot cases per year were observed per 1,000 women, while among older women (80+ years) 12 extra cases per year were observed per 1,000 women. These risks decreased over time but remained elevated for at least five years after diagnosis with most cancers.
The team also observed long-term increased risks of heart failure and heart muscle damage among survivors of 10 of the 20 site-specific cancers studied. Among people who had non-Hodgkin lymphoma five years previously, there were four extra cases of heart failure per year for every 1,000 people aged under 60, rising to 21 extra cases per year for every 1000 people aged over 80 years.
The senior author of the study Professor Krishnan Bhaskaran from the London School of Hygiene & Tropical Medicine said: “With treatment for cancer becoming more effective, we must start thinking about living beyond cancer and improving the health of survivors. Our work revealed raised risks of various cardiovascular diseases in cancer survivors. These findings are important because those affected might benefit from preventative measures and earlier interventions to reduce the impact.
"If patients, primary care doctors, and specialists are more aware of the potential for increased cardiovascular risks in this patient group, we might detect problems earlier and improve outcomes. There is limited guidance at present to help doctors manage cardiovascular risk in cancer survivorship, and further evidence is needed on whether routine monitoring or additional preventative measures would benefit those with the highest risks."
The authors acknowledge limitations of their study including that there was only limited information available about the anti-cancer treatments that patients had received. Although the research team had some information on whether patients had received chemotherapy, radiotherapy, and surgery, there was no detailed information on specific chemotherapy drugs, radiotherapy doses, or surgery procedures.
The study was funded by the Wellcome Trust and the Royal Society.
Helen Strongman, Sarah Gadd, Anthony Matthews, Kathryn E Mansfield, Susannah Stanway, Alexander R Lyon, Isabel dos-Santos-Silva, Liam Smeeth, Krishnan Bhaskaran. Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: a population-based cohort study using multiple linked UK electronic health records databases. The Lancet. DOI: 10.1016/S0140-6736(19)31674-5
The Electronic Health Records (EHR) Research Group at the London School of Hygiene & Tropical Medicine are pleased to announce they are now accepting applications for two longstanding pharmacoepidemiology and pharmacovigilance short courses: Professional Certificate in Pharmacoepidemiology & Pharmacovigilance and Practical Pharmacoepidemiology. Details about this year’s courses are provided below.
Accreditation: Both face-to-face courses have been approved by the Federation of the Royal Colleges of Physicians of the United Kingdom for Category 1 (external) Continuing Professional Development (CPD) credits (Professional Certificate in Pharmacoepidemiology & Pharmacovigilance = 30 credits; Practical Pharmacoepidemiology = 20 credits).
Professional Certificate in Pharmacoepidemiology & Pharmacovigilance
Course dates: 4 – 7 November 2019, 17 – 21 February 2020, 6 – 9 April 2020 (also available via distance learning)
This course is a 30-week examined introductory training course that aims to equip students with a basic understanding of the concepts and practice of pharmacoepidemiology, pharmacovigilance and drug safety. By the end of the course, students will be able to:
- Demonstrate an understanding of, and critically evaluate, issues surrounding the risks and benefits of drug use in humans
- Gain an understanding of, and reflect upon, important pharmacoepidemiological concepts and methods
- Assess and critically analyse the results of pharmacoepidemiological studies (other investigators'), including critical appraisal of the study question, study design, methods and conduct, statistical analyses and interpretation
Find out more about the Professional Certificate in Pharmacoepidemiology & Pharmacovigilance course. Anyone with questions about the course content/suitability can contact Dr. Kevin Wing or Dr. Rohini Mathur at Kevin.Wing@lshtm.ac.uk or Rohini.Mathur@lshtm.ac.uk.
Course dates: 16 – 19 September 2019
The course is designed for students with a basic grounding in epidemiological methods and concepts and/or some prior knowledge of pharmacoepidemiology (which could include people who have already completed the Professional Certificate in Pharmacoepidemiology & Pharmacovigilance at LSHTM). During this course students will:
- Develop their knowledge of pharmacoepidemiological concepts and methods, with a particular focus on database studies.
- Gain practical experience of testing study feasibility and performing analyses in STATA, using primary care data from the Clinical Practice Research Datalink (CPRD) (prior experience of STATA not required).
- Gain an understanding of biases and other sources of error that can occur in pharmacoepidemiology studies, and strategies to avoid them.
Find out more about the Practical Pharmacoepidemiology. Anyone with questions about the course content/suitability can contact Dr. Christopher Rentsch or Dr. Harriet Forbes at Christopher.Rentsch@lshtm.ac.uk or Harriet.Forbes@lshtm.ac.uk.
Finally, if you have any colleagues that you think would benefit from the course, do please let them know. And if there is an appropriate email group and/or noticeboard at your place of work, it would be greatly appreciated if you could circulate/post the course details to help us keep the course recruiting strongly.
The theatre performance ECZEMA! which is part of Professor Sinéad Langan’s Wellcome Trust funded public engagement project is now open for booking on Tuesday 4 and Wednesday 5 June. The performance takes place at the South London Gallery (65-67 Peckham Rd, London, SE5 8UH), starting at 7.30pm and has a running time of 35 minutes. To find out more, and to book tickets, please see the SLG website:
The performance, originally commissioned by National Theatre Wales, was updated to include outcomes from a creative writing workshop "Understanding how adult eczema affects lives" held at LSHTM in collaboration with the artist Maria Fusco in February. It is a dark comedy performed by Welsh actor Rhodri Meilir (Pride, Doctor Who) who is accompanied by organist, John Harris, who composed the score using motion-capture to translate scratching gestures (from the artist/director who has eczema herself) into music. The performance is written and directed by artist Maria Fusco and explores eczema.
The first annual lecture in memory of our greatly missed colleague Adrian Root will take place at LSHTM on Thursday 23 May. The science writer, epidemiologist and physician Ben Goldacre will be providing his unique insights into key issues around evidence-based medicine. Full details can be found on the event page.
In 2018, the EHR group launched a memorial prize, to honour the work of our colleague, Dr. Adrian Root. The prize will be awarded annually to an LSHTM MSc epidemiology student who has worked on an Electronic Health Records project. The inaugural prize was awarded to Ms. Amanda Clery at the LSHTM graduation ceremonies on March 5 2019. Amanda completed a cohort study investigating the effects of quality of antenatal care on the risk of stillbirth in the UK using the data from the Clinical Practice Research Datalink. She was supported by EHR group members Rohini Mathur (project supervisor), Caroline Minassian, and Liam Smeeth.
“We wanted to connect with people”.
As researchers we work with data; millions of lines of code, statistics and figures, tables and graphs, funnel plots indicating patterns and outliers. But we always need to remind ourselves that every line of data, every entry in a medical record or hospital database, relates to an individual journey. We got involved with cancer research for different reasons but ultimately we are all passionate about improving outcomes for cancer patients, the individuals behind the data that we use to investigate and understand the disease.
Unlike biomedical research into cancer, involving drugs or treatments, clinical trials or patient level studies, there is less of an obvious way for epidemiologists and statisticians to engage with the public and involve patients in our research. What we do is often not well understood and the value of health data for research is sometimes overlooked.
For all these reasons, we wanted to step away from our computer screens and the data, and connect with those affected. We wanted to interact with cancer patients and carers, to understand their experiences with the disease, and have the opportunity to share with them what we do, and how it involves them and their data, and why this is important. We wanted to open a dialogue and encourage mutual understanding, and we wanted to do this by creating a safe and open space for sharing, for reflecting and for expressing what can sometimes feel ‘inexpressible’.
What better way to do this, to bring people, strangers, together than through art? Anthony, a Research Fellow in the Electronic Health Records group at the London School of Hygiene & Tropical Medicine came up with the idea after discussing the effectiveness of art as therapy with Jayne Dent, an artist currently based in his hometown of Newcastle. She works with many artistic forms but was interested in the use of mono-printing as a form of expression. Mono-printing is a simple technique that creates a one-off ‘mono’ print on paper. It is an artistic form that does not need prior experience of art, specialist equipment or careful planning. It is an instant form of expression where images and text can be combined to produce a unique piece of work.
Anthony brought in Camille and Yuki, two other staff from the Department of Non-Communicable Disease Epidemiology, and together we put in a proposal which was awarded funding from the School’s public engagement small grants scheme. This enabled us to go ahead and prepare a workshop, facilitated by Jayne, and an exhibition in Newcastle.
With the help of the Cancer Research UK Centre in Newcastle and Maggie’s Centre at the Freeman Hospital in Newcastle, we were able to reach individuals in the local area that responded to flyers and social media postings advertising the event. The one-day workshop took place in late November 2018.
The 8 participants, including 7 cancer patients and 1 carer, that came along to the workshop were a special group of people who, through their courage and openness, and their willingness to fully engage with the experience, made the event a unique experience for everyone involved. We were delighted with the wonderful, expressive and moving prints they produced, as well as their observations of the day:
Being in the space, meeting other participants, being able to talk (and cry) freely, and being pushed to express feelings visually – all of this had a profound effect on me.
Coming together with strangers to share and learn something new was surprisingly calming …. a great way to cope with life’s traumas!
For us, the experience brought the human element to our daily work, the connection we were looking for. It provided the contact and the opportunity to talk to the participants about their individual journeys. It was a unique and valuable insight into the living with the disease.
Throughout the day we were able to tell them about the work that we do, how it relates to them and how their willingness to share their experiences, and their data, makes the research that we do both possible and relevant. It also gave us a fresh perspective on research questions and proposals we are considering, by reflecting and discussing the patients’ journey and their interactions with the healthcare system.
We also learned about the power of creative expression and the ability of art to act as both a form of release and of bringing people together. Before the project, we did not quite realise the power that art had, but watching the participants grow throughout the day, finding new ways of expressing their feelings, was extremely eye opening and gratifying. It changed us all in ways we might not have expected, and was a day we will never forget.
Anthony Matthews, Camille Maringe and Yuki Alencar
Exhibition: Our Cancer Journey – artistic expressions of living with cancer
28th March – 3rd April 2019
Newcastle City Library
Charles Avison Building
33 New Bridge Street West
Newcastle upon Tyne
Herpes zoster, caused by reactivation of the chickenpox virus, is linked to an increased risk of stroke, according to a recent systematic review by Harriet Forbes and colleagues. In the review, the risk of stroke approximately doubled in the week after herpes zoster and gradually returned to baseline over the next six months. While recent infection with, or reactivation of, other members of the herpesvirus family such as cytomegalovirus and herpes simplex virus may increase stroke risk, further high quality evidence is needed to confirm these findings. The paper, which included 41 studies, has recently been published in PLoS ONE.
The findings are explained in the following infographic:
Two systematic reviews assessing long-term outcomes in survivors of breast cancer have recently been published by members of the EHR research group.
Anthony Matthews and colleagues collated substantial randomised controlled trial and observational evidence on the effect of endocrine therapies on several specific cardiovascular diseases. 26 studies were identified, with results for seven specific cardiovascular disease outcomes. Results suggested an increased risk of venous thromboembolism in tamoxifen users compared with both non-users and aromatase inhibitor users. Results were also consistent with a higher risk of the vascular diseases myocardial infarction and angina in aromatase inhibitor users compared with tamoxifen users, but there was also a suggestion that this may be partly driven by a protective effect of tamoxifen on these outcomes. Data were limited, and evidence was generally inconsistent for all other cardiovascular disease outcomes. This review shows that although the choice of aromatase inhibitor or tamoxifen will primarily be based on the effectiveness against the recurrence of breast cancer, the individual patient’s risk of venous or arterial vascular disease should be an important secondary consideration. The full article was published in the BMJ.
Helena Carreira and colleagues looked at adverse mental health outcomes in women with a history of breast cancer. The authors reviewed 60 studies that compared adverse mental health outcomes in women with a history of breast and in women who never had cancer. The results showed compelling evidence of an increased risk of anxiety, depression and suicide, and neurocognitive and sexual dysfunctions in breast cancer survivors compared with women with no prior cancer. Evidence for other outcomes is scarcer, but breast cancer survivors have also been reported at increased risk of sleep disturbance and stress-related disorders including post-traumatic stress disorder. This review also highlighted that further population-based and longitudinal research would help to better characterize these associations, as approximately one-half of the studies were at high risk of selection bias and confounding by socio-economic status. The full article was published in the Journal of the National Cancer Institute.
ACE Inhibitors and ARBs are drugs that have transformed clinical care. There is strong randomised trial evidence for improved clinical outcomes from these drugs for patients with heart failure with reduced left ventricular function, and for proteinuric kidney disease. They are also recommended for use for the treatment of hypertension and after ischaemic heart disease. These multiple indications have led to them being one of the most commonly prescribed drug groups. However, in recent years there has been growing concern about their potential for ‘nephrotoxicity’. They are widely believed to be associated with AKI, particularly in patients who are hypovolaemic or septic.
Multiple guidelines recommend dose reduction or cessation of these drugs for patients who have developed or are at risk of AKI. However, it is often unclear whether, and when, the drugs should be restarted, particularly when patients are at risk of further AKI. We do know that patients are at high risk of readmission with cardiac failure after AKI but it is not clear how much this is due to cessation of ACEI/ARB drugs as part of AKI management. In addition, AKI itself is associated with increased risk of future adverse outcomes including mortality and development of chronic kidney disease. Do these drugs offer specific benefits after AKI? These issues are at the heart of daily clinical work, and yet there is minimal evidence to help us make these important decisions. In this nephrology journal club blog, Laurie Tomlinson discusses whether a recently published observational study on this topic provides evidence that can be applied to routine practice.
Body Mass Index (BMI), a measure of body fat, is linked to risk of death from every major cause except transport accidents, according to new research by the LSHTM Electronic Health Records Research Group.
The study is one of the largest of its kind to look at how BMI is associated with the risk of death both overall, and from a full spectrum of different causes – 3.6 million people and 367,512 deaths were included in the analysis. Overall, both low and high BMI were associated with an increased risk of death. BMI of between 21-25kg/m2 was associated with the lowest risk of dying from cancer and heart disease.
The research team used anonymised data from the UK Clinical Practice Research Datalink (CPRD) which includes data on BMI from general practitioners’ primary care records covering about 9% of the UK population. This is linked to data from the Office of National Statistics mortality database, which includes information on causes of death as recorded on death certificates. Risks of death from each major cause was calculated according to BMI, adjusting for other important factors such as age, sex, smoking status, alcohol use, and socioeconomic status.
Lead author Krishnan Bhaskaran said: “BMI is a key indicator of health. We know that BMI is linked to the risk of dying overall, but surprisingly little research has been conducted on the links to deaths from specific causes. We have filled this knowledge gap to help researchers, patients and doctors better understand how underweight and excess weight might be associated with diseases such as cancer, respiratory disease and liver disease.
“We found important associations between BMI and most causes of death examined, highlighting that body weight relative to height is linked to risk of a very wide range of conditions. Our work underlines that maintaining a BMI in the range 21-25kg/m2 is linked to the lowest risk of dying from most diseases.”
The authors acknowledge limitations of the study including that there was no information was available on the diet or physical activity levels of people included in the study so it was not possible to look at the interplay between BMI and these related factors. The full article was published in The Lancet Diabetes & Endocrinology.
Dr Charlotte Warren-Gash from the Electronic Heatlh Records Research group at LSHTM recently contributed to an episode of the BMJ Heart podcast, speaking with Dr James Rudd about the links between respiratory infections and subsequent heart attacks and strokes. Listen to the podcast.
Adults with eczema could face an increased risk of experiencing non-fatal cardiovascular disease, according to a recent study carried out by members of the Electronic Health Records Research Group at LSHTM.
The observational study involving almost 2 million people found that patients with eczema were 10-20% more likely to experience non-fatal cardiovascular disease than people without eczema, and that the risk increased with more severe disease. Using UK electronic health records from the Clinical Practice Research Datalink, Hospital Episode Statistics and data from the Office for National Statistics between 1998–2015, patients diagnosed with eczema were matched to those without on age, gender, general practice and calendar time. After adjusting for confounders such as socioeconomic status and age, the differences in cardiovascular disease risk between the two groups was analysed. Risk in severe eczema persisted even after adjusting for traditional cardiovascular risk factors that might mediate this association, including smoking and body mass index.
Patients with severe eczema were found to experience a 20% increased risk of stroke, 70% increased risk of heart failure, and 40–50% increased risk of the remaining cardiovascular outcomes, including unstable angina, myocardial infarction, atrial fibrillation and cardiovascular death. Eczema can vary over time and patients whose eczema was active for most of their follow up were also at greater risk of cardiovascular outcomes. The risk was mainly confined to patients with severe eczema, including people on oral immunosuppressive drugs, receiving phototherapy treatment for eczema or who were referred to dermatologists.
Eczema, also known as atopic eczema or atopic dermatitis, is a common systemic inflammatory condition which affects up to 10% of adults and is becoming more common globally. Symptoms include intense itch, pain and sleeplessness. It’s estimated that around 30% of eczema patients would be classified as having moderate-severe eczema.
Dr Sinead Langan said: “Eczema is a debilitating common condition. Increasing evidence suggests that severe eczema could be associated with a wider range of health problems than originally thought. Previous studies on the link between eczema and cardiovascular disease have reported mixed findings. However, these have lacked data on specific risk factors and have not assessed levels of eczema activity over time, points which our research addressed.”
The absolute risk of people with eczema experiencing a cardiovascular event is low. However, the links uncovered in this research, if robustly replicated by future studies, would support targeted screening and focus on primary prevention strategies to reduce cardiovascular disease among patients with eczema.
The full article was published in the British Medical Journal.
Some of the EHR group have just completed a London to Paris bike ride in memory of Adrian Root, a member of the group who sadly died at the end of 2017. We raised over £20,000 to set up a new prize for the MSc epidemiology in Adrian’s name. The trip included a gruelling (=hilly, & lots of angry motorists!) first 65 mile day from London to Newhaven, before a few more leisurely days through the French countryside. Arrival in Paris was made even more exciting by being just prior to the World Cup Final kick-off!
Asthma and COPD share many characteristics and symptoms, and the differential diagnosis between the two diseases can be difficult in primary care. Francis Nissen and colleagues have recently done a study to quantify how commonly patients with chronic obstructive pulmonary disease (COPD) have a concomitant diagnosis of asthma, and how commonly patients with asthma have a concomitant diagnosis of COPD in UK primary care. They found that a concurrent asthma and COPD diagnosis only affects a relative minority of patients with COPD (14.5%) or asthma (14.8%), and that asthma diagnoses may be over-recorded in people with COPD.
The study was based upon data from previous validation studies, including 400 COPD patients and 352 asthma patients from the Clinical Practice Research Datalink (CPRD), with disease status confirmed by review of questionnaires sent to general practitioners (GPs). The prevalence of concurrent asthma and of COPD in validated cases of either disease was examined based on CPRD coding, GP questionnaires and requested additional information. More than half (52.5%) of validated COPD patients had ever received a diagnostic asthma Read code. However, when considering additional evidence to support a diagnosis of asthma, concurrent asthma was only likely in 14.5% of validated COPD patients. Of the validated asthma patients, 15.1% had a diagnostic COPD Read code, although COPD was only likely in 14.8% of validated asthma patients. The full paper has just been accepted for publication in the British Journal of General Practice.
Study with us
We are pleased to announce a number of funded PhD projects led by EHR Research Group members. Click here for more details, including eligibility and how to apply.
Understanding ethnic differences in the comparative effectiveness of antidiabetic medications using high- dimensional propensity scores in electronic health records (Supervisors: Dr Rohini Mathur & Dr Elizabeth Williamson)
Sleep disturbance and dementia: exploring the relationship between common causes of sleep disturbance and long-term cognitive changes using large linked datasets (Supervisors: Dr Charlotte Warren-Gash & Dr Helen Strongman & Prof Liam Smeeth)
Improving the ability to study epilepsy in big data: development and validation of an epilepsy case classification algorithm (Supervisors: Dr Kevin Wing & Dr Elizabeth Williamson & Dr Charlotte Warren-Gash)
Mode of study: Online
Duration of study: 4 days
Click here for more details, including dates, registration, and contact information.
Mode of study: Online
Duration of study: 3 weeks spread out over 1 year
Click here for more details, including dates, registration, and contact information.
Mode of study: Face-to-face in London
Duration of study: 3 weeks spread out over 1 year
Click here for more details, including dates, registration, and contact information.
Mode of study: Online
Duration of study: 4 days
Prerequisites: Applicants will normally have a science, biomedical or biostatistical background, hold a second class honours degree of a United Kingdom university (or equivalent) in a science, medical, statistical or related subject and will have some post-graduate experience in the area of pharamacoepidemiology, including an understanding of methodological concepts such as bias and confounding. It is not expected that applicants will actually be conducting pharmacoepidemiology studies.
Click here for more details, including dates, registration, and contact information.
Please check back soon or follow us on Twitter (@ehr_lshtm) for up-to-date information on other funding opportunities and other enhanced virtual short courses currently in development.