Intended for healthcare professionals


Changes in atopy over a quarter of a century, based on cross sectional data at three time periods

BMJ 2005; 330 doi: (Published 19 May 2005) Cite this as: BMJ 2005;330:1187
  1. Malcolm Law ({at}, professor1,
  2. Joan K Morris, reader1,
  3. Nicholas Wald, professor1,
  4. Christina Luczynska, honorary senior research fellow2,
  5. Peter Burney, professor2
  1. 1 Centre for Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, Barts and The London School of Medicine and Dentistry, London EC1M 6BQ
  2. 2 Department of Public Health Sciences, Division of Primary Care and Public Health Sciences, Guy's, King's, and St Thomas' School of Medicine, Guy's Hospital, London SE1 3QD
  1. Correspondence to: Malcolm Law
  • Accepted 16 March 2005


Evidence that the prevalence of atopic diseases, including asthma and hay fever, has increased over the past 20-30 years comes mainly from questionnaire based surveys; objective measurements are limited.14 We therefore measured serological markers of atopic sensitisation in stored serum samples that had been collected from a population of men over about 25 years.

Participants, methods, and results

We used frozen (−40°C) serum samples from men aged 40-64 years who had attended the British United Provident Association (BUPA) Medical Centre, London, for a routine medical examination. An advantage of this cohort is its socioeconomic homogeneity—almost all were professionals or businessmen. We matched (by age and month of attendance) the 513 samples from men who attended during 1996-8 to samples from 513 men seen in 1981-2 and 513 seen in 1975-6. The serum samples had not previously been thawed; previous work has shown negligible decay in IgE during storage of serum.5

What is already known on this topic

The prevalence of atopic diseases is increasing

Cross sectional studies show that the prevalence of atopy decreases with increasing age

What this study adds

Serological measurements at three time periods provide objective evidence that earlier birth cohorts were less likely to have become atopic than more recent ones; this accounts for the apparent decreasing prevalence of atopy with age

We first tested the samples using Phadiatop (Pharmacia & Upjohn), a standard qualitative serological marker of atopy based on a mixture of 11 common, inhaled allergens (grass, tree, nettle, mugwort, mould, and olive; cat, dog, and horse dander; and two house dust mites). The positive samples were then tested for specific IgE to grass (timothy grass), tree mix (elder, silver birch, hazel, oak, and plane), and cat dander (Pharmacia CAP system); we did not test the Phadiatop negative samples because Phadiatop includes all these allergens. The samples were also tested for IgG antibody to hepatitis A (DiaSorin) and to Helicobacter pylori (samples from 1996-8 only; Axis-Shield Diagnostics).

The table shows that we found highly significant increases over time in the proportion of men positive to Phadiatop and with specific IgE to the three inhaled allergens. The average rate of increase was equivalent to an additional 4.5% of men becoming Phadiatop positive each decade.

Proportions of men aged 40-64 years positive to Phadiatop* and with specific IgE to three inhaled allergens, in three time periods, and mean annual increase in prevalence. Values are percentages (95% confidence interval)

View this table:

We compared the data on men born in 1932-42 who had had serum taken in the earlier periods (1975-6 or 1981-2, at age 40-50) and on men born in the same years (1932-42) who had had serum taken in the later period (1996-8, at age 54-64). The proportions of samples positive to Phadiatop were similar for both the earlier and later periods (35% and 34% respectively). The proportions of samples positive for specific IgE to grass, tree, and cat dander were also similar for the two time periods. In multiple regression, year of birth was associated with the prevalence of atopy (P < 0.001), but current age was not.

Our results do not support previous data suggesting an inverse association between two common enteric infections and adult atopy. The proportions of samples that were Phadiatop positive were similar in men with and without serological evidence of past hepatitis A infection (37% v 34%) and Helicobacter pylori infection (40% v 36%). Two other markers of childhood infection available from questionnaire data—number of siblings and whether a man went to boarding school—also showed no association with Phadiatop positivity.


Our data show that atopy in middle aged men has increased during the last quarter of the 20th century and that the prevalence of atopy does not decline with increasing age; rather, more recent birth cohorts are more likely to have become atopic. Reports of increases over time in specific IgE to inhaled allergens in adults from Nordic countries and Swiss and Japanese schoolchildren14 involved younger subjects and shorter time periods than ours.

The reason for the increase in atopy is unknown. Our data suggest that it is unlikely to be an increased exposure to specific allergens (because sensitivity to both indoor and outdoor allergens increased). Our results also do not support the increase being due to declining childhood infections.

This article was posted on on 15 April 2005:


We thank the BUPA Foundation for support for maintaining the serum samples and database.


  • Contributors ML conceived the study, JKM conducted the statistical analyses, CL did the serological measurements, and all authors contributed to the paper through shared discussion and data interpretation. ML and NW are the guarantors.

  • Funding The Wellcome Foundation and the BUPA Foundation funded this study.

  • Competing interests None declared.

  • Ethical approval Not needed.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.