From experience of working with patients with severe eczema in a tertiary clinic, we know the major impact that eczema has on individuals’ health and quality of life. However, in clinic we only see the most severe patients, and we have suspected that there is a large unmet need and many unanswered questions that are relevant to patients who are seen in general practice.
Previous studies have shown variable findings for the association between eczema and cardiovascular outcomes, but many had limitations that we thought our data could help clarify. For example, most previous studies had limited assessment of the effects of disease severity. No previous research focused on the concept of disease activity over time, both of which are heterogeneous between patients, and clinically important to patients and physicians.
As eczema is so common (affecting up to 10% of adults), even a small increase in cardiovascular disease risk could be important from a public health perspective. As Professor Geoffrey Rose has highlighted, “a large number of people at a small risk may give rise to more cases of disease than the small number who are at high risk”.
One of the key difficulties of conducting our research was capturing eczema disease severity and activity. Eczema is an episodic disease, and individuals may experience periods of disease activity as well as times when their disease is well controlled. It can also vary widely in its severity, a characteristic which is not routinely recorded in primary care records. Severity therefore needed to be inferred from other data points – specifically the medications that patients received. Deciding on a robust algorithm to identify eczema patients, their disease activity, and their disease severity, was challenging.
There were several potential sources of bias within the study, including the possibility of misclassification of the exposure. This presented two questions: firstly how to robustly address these sources of bias within the analysis. We overcame this by conducting eight sensitivity analyses. Secondly, how to effectively communicate in our research the justification for sensitivity analyses and the approach that we took. We found that displaying this in a a table really helped to clarify this.
We found higher risks of cardiovascular disease than expected among eczema patients relative to people without eczema, as well as consistent findings across the cardiovascular outcomes when assessing severe and persistently active eczema. If these results are robustly replicated, it would support targeted screening and a focus on primary prevention strategies to reduce cardiovascular disease among such patients.
Sinead Langan is a Wellcome Senior Clinical Fellow and an Associate Professor in the Faculty of Epidemiology and Population Health at the London School of Hygiene and Tropical Medicine. She also works as an honorary consultant dermatologist at St John’s Institute of Dermatology, Guy’s and St Thomas’ hospital London in a busy tertiary service for patients with eczema.
Richard Silverwood is an assistant professor of Medical Statistics in the Faculty of Epidemiology and Population Health and Co-Director of the Centre for Statistical Methodology at London School of Hygiene & Tropical Medicine.