XClose

UCL Institute of Health Informatics

Home
Menu

Using health record data to find out about the role of gender in the onset of symptomatic cardiovascular disease

Principal investigator: Harry Hemingway
Investigators: Julie George, Eleni Rapsomaniki, Mar Pujades-Rodriguez, Anoop Dinesh Shah, Spiros Denaxas, UCL and Farr London; Emily Herrett and Liam Smeeth, London School of Hygiene & Tropical Medicine and Farr London; and Adam Timmis, Queen Mary University of London and Farr London.

The Challenge

Cardiovascular disease is the leading cause of death from non-communicable diseases worldwide, accounting for 17.5 million deaths worldwide in 2012. However, the pattern of onset of cardiovascular disease, at least in developed countries, appears to be changing. The number of deaths from heart attacks and strokes has declined by about a third in the last decade, but more chronic presentations like stable angina or heart failure do not seem to have followed the same pattern. We therefore did not know what the current pattern of the onset of cardiovascular disease or how it might differ between men and women. Most researchers carrying out research into the prevention of cardiovascular disease have studied people who have had heart attacks and strokes. However, since these life-changing outcomes are what need to be prevented, it is actually necessary to investigate factors that occur before onset of the disease. Farr Institute researchers realised electronic health records contained a wealth of information leading up to the late events, particularly about what could be seen when patients first attended clinic. The challenge was to analyse data linked from a number of sources to find useful patterns for improved early disease prevention strategies.

The Research

Researchers from UCL, London School of Hygiene & Tropical Medicine and Queen Mary University London used innovative methods to analyse linked anonymised health record data from general practices, hospitals, a specialist register of admissions for heart attacks and mortality records to investigate the onset of cardiovascular disease (CVD) in 1.9 million patients with no prior history of diagnosed CVD. We investigated which of 12 CVDs affecting the heart, head, abdomen and legs appeared first during the follow-up period and how this varied by age and gender of the patients. We took into account known risk factors for CVD such as smoking, raised blood pressure, cholesterol, obesity and social deprivation.

The Results

The study followed up patients for six years between 1997 and 2010. During this time about 115,000 people experienced their first cardiovascular diagnosis. Interestingly two thirds of diagnoses (66%) were not heart attack or stroke. This was at odds with the fact that currently most cardiovascular disease prevention strategies focus on these two events. Diagnoses included such diseases as peripheral arterial disease, stable angina, heart failure and abdominal aortic aneurysm. Being male was strongly associated with a first lifetime presentation of heart attack, abdominal aortic aneurysm, or death from coronary disease with no prior clinical diagnosis, especially for men under 60 years of age. With increasing age, the proportion of patients whose initial lifetime presentation of CVD was either heart failure or different types of stroke increased, while differences between men and women declined.

The Impact

The results of the study have been published in a journal widely read by doctors and researchers and used in continuing medical education resources. The study, which could not have been done without the scale and detail available with anonymised linked electronic health records, provided important new information on the differences and similarities between men and women in the onset of CVD, which has the potential for a more personalised approach to prevention of cardiovascular disease.

For more information: George J et al. (2015) Circulation Oct 6;132(14):1320-8.