|Home | About | Journals | Submit | Contact Us | Français|
The prevalence of eczema, particularly in younger children, increased substantially over the second half of the 20th century. Analysis of primary healthcare data-sets offers the possibility to advance understanding about the changing epidemiology of eczema.
To investigate recent trends in the recorded incidence, lifetime prevalence, prescribing and consulting behaviour of patients with eczema in England.
QRESEARCH is one of the world’s largest national aggregated health databases containing the records of over nine million patients. We extracted data on all patients with a recorded diagnosis of eczema and calculated annual age–sex standardized incidence and lifetime period prevalence rates for each year from 2001–2005. We also analysed the consulting behaviour of these patients when compared with the rest of the QRESEARCH database population. The number of eczema prescriptions issued to people in England was also estimated.
The age–sex standardized incidence of eczema was 9.58 per 1000 person-years in 2001 and increased to 13.58 per 1000 patients in 2005 (p<0.001). By 2005, eczema affected an estimated 5,773,700 (95% confidence intervals [CI] 5,754,100–5,793,400) individuals in England, who, on average, consulted a general practitioner 4.02 (95% CI 4.01–4.03) times a year. During the study period, the number of eczema related prescriptions increased by 56.6% (95% CI 56.6–56.7), so by 2005 an estimated 13,690,300 (95% CI 13,643,200–13,737,600) prescriptions were issued.
Recorded incidence and lifetime prevalence of eczema in England continue to increase. Similar increases have also been observed in the estimated number of eczema prescriptions issued to the English population.
A recent review of UK epidemiological data revealed that there has been an inexorable rise in the prevalence of allergic disorders.1 Allergic pathophyisiology can cause a spectrum of diseases in individuals, which may vary in severity. Eczema is an inflammatory skin disorder often resulting in red, itchy and poorly‐defined patches occurring on flexural surfaces, and is most commonly found in children of preschool age. Atopic eczema has been defined as a chronic, relapsing, inflammatory skin condition associated with epidermal barrier dysfunction, which in turn is in many cases increasingly believed to result from an underlying filaggrin gene defect.2 Individuals can also exhibit eczema without atopic features (i.e. non‐atopic eczema); such patients will respond to treatments such as creams containing corticosteroids and emollients just as well as individuals with atopic eczema. Individuals with atopic eczema are however – unlike those with non‐atopic eczema – more likely to develop other atopic diseases such as asthma or hayfever later in life,3 and their eczema is often severe and more likely to persist into adulthood.4
Survey data provide useful information on variations in lifetime/period prevalence of self‐reported diagnoses of eczema, particularly in children and adolescents. However there are relatively few reliable national data describing clinician‐diagnosed disease incidence; furthermore very few data exist on the overall population trends over time for all ages. Exploitation of large national healthcare data‐sets, with their key strengths of large numbers and representative data, offers an important opportunity to develop insights into the epidemiology of eczema.5 Studying primary care databases provides a picture of overall national trends – something that is not possible with large scale surveys such as ISAAC, which has studied only children,6 and the ECRHS, which has surveyed only adults.7 Large primary care data‐sets such as QRESEARCH, recording information at the point at which the majority of patients with eczema are likely to be managed, do however offer an important opportunity to study changing patterns of disease. Building on previous work,8–10 we sought to describe recent trends in the primary care diagnosis, prescribing and consulting behaviours of patients with eczema in England.
Version 10 of the QRESEARCH database was used for these analyses. This database contains representative anonymized aggregated health data derived from 525 general practices throughout England. Data were available for the period 1 January 1999 to 31 December 2005, these comprising over nine million individual patients who collectively contributed over 30 million patient‐years of observation. The methods used to collect primary care data for the QRESEARCH database have been previously described.8
Patients were included in the analysis year if they were registered for the entire year of study. Patients with incomplete data (i.e. temporary residents, newly-registered patients and those who joined, left or died during the study year) were excluded. Patients were considered to have eczema if they had a relevant computer‐recorded diagnostic Read code in their electronic health record during the time period of interest.
Incidence was defined as the number of patients with a new case of eczema diagnosed in a specific year, with the denominator being the number of patient‐years of observation (calculated from the number of patients registered with practices and their length of registration). Lifetime prevalence was defined as the number of people with eczema ever recorded on at least one occasion in the general practice records; the denominator used to calculate the lifetime prevalence rate was the number of patients registered with the study practices.
In order to describe trends in prescribing of eczema medication, we extracted prescribing data and estimated numbers of eczema prescriptions issued to all patients listed in chapters 13.2, 13.4 and 13.5 of the British National Formulary (emollient and barrier preparations, topical steroids, and psoriasis and eczema treatments). Although estimated prescriptions issued to the whole English population may equate with the numbers of prescriptions dispensed nationally, these figures are not directly comparable to Prescribing Analysis and Cost Tabulation (PACT) data.
In order to compare our results on the rates and trends of eczema in England with other published data, a structured literature review was carried out. We conducted our search using Medline and Embase (from 1951 to December 2008), and also Google Scholar. The search terms used included: eczema, trends, prevalence and incidence.
Eczema was defined as patients who have Read codes M11 and below and M12z and below (see Box 1 for Read codes used).
As a result of known age and sex variations, rates of disease and prescribing were standardized by sex and five‐year age bands. The mid‐year population estimates for England in each year of study were used as the reference population. These results were then used to estimate the numbers of people with eczema in England. The Mantel‐Haenszel χ2 test was used to investigate trends over time, this analysis being undertaken using EpiInfo2000 (World Health Organization, Geneva, Switzerland).
During the five‐year study period the number of incident cases per 1000 population increased by 41.8% (p<0.001; Table 1). In 2005, approximately 1 in every 74 people in England was newly diagnosed as having eczema.
Between 2001 and 2005, there was a significant increase in the number of registered patients with a diagnosis of eczema at some point in their lives ( Figure 1). The age–sex standardized lifetime prevalence rate of eczema increased for the whole population over each of the five years of interest, with an overall 48.2% increase (p<0.001; Table 2). In 2005, eczema affected an estimated 5,773,700 (95% confidence intervals [CI] 5,754,100–5,793,400) or about 1 in 9 people in England. It was more common in girls/women than in boys/men, except in children aged under five years or those aged over 75 years. In 2005, the highest lifetime prevalence rate occurred in boys aged 5–9 years (Figure 1).
Table 3 compares general practitioner (GP) and nurse consultation rates for patients with eczema. Figure 2 compares overall consultation rates for the whole QRESEARCH population with those for patients with eczema broken down by age and sex. This includes all GP and nurse consultations in 2005, regardless of the reason for the encounter. Consultation rates for women tended to be higher than for men, and consultation rates for patients with eczema were higher than overall consultation rates. For example, for men aged 85–89 years, the GP and nurse consultation rate for patients with a diagnosis of eczema was 1.3 times higher than the corresponding overall consultation rate in that group of patients.
Figure 3 shows consultation rates per patient (regardless of the reason for the consultation) for eczema broken down by age and sex. The highest consultation rate occurred in women aged 85–89 years.
Table 4 shows the estimated number of eczema prescriptions issued to the whole population in the years 2001 to 2005. Overall there was an increase of 56.6% (95% CI 56.6–56.7) in the number of prescriptions issued (emollient and barrier preparations increase: 78.7% (95% CI 78.6–78.7); topical corticosteroids: 36.4% (95% CI 36.3–36.4); and psoriasis and eczema: 20.7% (95% CI 20.7–20.8)).
This study, using routine data from one of the world’s largest national data‐sets, has revealed that eczema occurs very commonly in childrenand adults, and that, in the beginning of the new millennium, a large increase has occurred in the recorded incidence and lifetime prevalence (in all ages) of these problems. The number of consultations for eczema and eczema-related prescriptions issued in primary care in England has also increased during the study period. The consultation rate also seemed to be higher than the overall population and persisted into adulthood, when, in the majority of cases, eczema is no longer likely to be as problematic.
The main strengths of this study include our interrogation of an extremely large nationally representative data‐set, the fact that all contributing practices used the same computing systems for electronically recording clinical data, and the approach used to ensure that all contributing practices were accustomed to electronically recording routine data. The study design employed ensured that there was no risk of selection bias due to non‐responders or recall bias. Another strength of this study was the use of contemporaneous clinician recording of a diagnosis of eczema as opposed to patient self‐reporting of historical diagnoses or symptoms.6,7
There are a number of limitations related to the use of large routinely collected data from primary care, including the dependence on clinician‐recorded diagnosis of eczema and possible improvements in recording over the study time period. The relatively short time window over which trends were studied is another limitation, but this does also have the advantage of confining analysis to a period during which there were no changes in disease definition or classification. Data regarding childhood prevalence may be underestimated, as the ascertainment of disease present in the community will be dependent on parents bringing their children for consultation.11 Underestimates of eczema prevalence have been compounded by some individuals with mild disease not consulting, opting to use either no treatment or over‐the‐counter preparations. The inadequacy of Read codes for allergy has been previously reported12 and although it is difficult to quantify the precise effect of this, it is likely to have contributed to an underestimate in relation to the actual population prevalence of eczema. However this inadequacy is unlikely to have had any effect on the increasing trends of eczema, as no changes in Read codes for eczema have been introduced during the study period.
Table 5 compares previous published epidemiological data for eczema. We have found using data from our work that the lifetime clinician‐recorded prevalence peaks in younger children mirrored results from a birth cohort,13 and the prevalence in our older adults was similar to that found in survey data.7 Although little data on the trend of eczema prevalence exist prior to World War II (1939–1945), the prevalence of eczema increased substantially in the latter half of the 20th century, with eczema in school‐aged children being found to increase between the late 1940s and 2000.14–16 In contrast to our results, survey studies (see Box 2 for questions used in surveys) from the period immediately prior (1995–1996 to 2000–2001)17 and intersecting our study period (1995–1996 to 2002–2003)6 found only a moderate increase in eczema in children (2–15 years), a decrease in older children (13–14 years) and no increase in adults (aged >15 years) over time. Also in contrast to our data, a study using general practice electronic data which intersected our study period (1995–2004) found a steep decline over time in young children (<5 years) presenting with eczema,18 but similar to our study found increases in adults (>45 years).
There may be several possible reasons for the increases in eczema diagnosis found in QRESEARCH. One possible explanation is that changes in environmental factors over time have favoured the expression of allergic disease in those who are genetically susceptible.2,19,20 There may also have been increases in sensitization over time, which may then in turn predispose the development of any of a number of allergic disorders. Supporting evidence for such a possibility comes from an important study by Law et al., which found significant increases in atopic sensitization in the UK over a 25-year window.21 Increased predisposition to atopy,22 possibly reflecting changing exposure to known and unknown risk factors,23 may also be important. Increases in the rate of these conditions could however result from increased clinician awareness of allergic problems, which may then have led to improved identification and recording of eczema. Similarly, increased patient awareness, or parental awareness of the potential of accessing effective treatments, may have resulted in increased case presentation and prescribing in primary care.
Given the high prevalence of eczema, particularly in younger children when compared with adults, the overall numbers of people in England with eczema is, for the present at least, likely to continue to increase. A key related important unanswered question concerns the quality of care and symptom control for parents and carers of younger children,24 particularly as eczema is most symptomatic at this younger age and is likely to herald the onset of other allergic conditions.25 Efforts must therefore be directed into investigating effective methods of primary prevention and symptom control, particularly as high levels of consultation rates for this growing population will persist into older age, with an associated substantial impact on the NHS, and in particular primary care. The House of Lords Allergy Inquiry published in 2007 has identified several issues highlighted by this work and other previous research that require further attention.26
This large national study reveals that the recorded incidence and lifetime prevalence of patients with eczema increased in England. With almost 1 in 9 of the population having experienced the condition at some point in their lives, eczema is now one of the most common chronic conditions to effect the English population, and therefore continuing monitoring of trends is very important. Whether these findings reflect a genuine increase in the incidence of eczema, improved awareness, diagnosis and recording in primary care, or, perhaps most plausibly, a combination of genuine increases and improved identification and recording, is a question with important public health implications and one therefore that warrants detailed further enquiry.
Competing interests JHC is Director of QRESEARCH (a not‐for‐profit organization owned by the University of Nottingham and EMIS, commercial supplier of computer systems for 60% of GP practices in the UK)
Funding NHS Health and Social Care Information Centre
Ethical approval Not applicable
Contributorship AS, JHC and JN were involved in designing the study and CS contributed to literature searches and led the drafting of the paperwith all co‐authors commenting on drafts of the manuscript
We would like to record our thanks to the contributing EMIS practices and patients and for EMIS for providing technical expertise in creating and maintaining QRESEARCH. We thank QRESEARCH staff (Govind Jumbu, Alex Porter, Justin Fenty, Mike Heaps and Richard Holland) for their contribution to data extraction, analysis and presentation. These findings have been reported in Primary care epidemiology of allergic disorders: analysis using QRESEARCH database 2001–2006, which is published by the NHS Health and Social Care Information Centre