Longitudinal study of childhood wheezy bronchitis and asthma: outcome at age 42

BMJ 2003; 326 doi: (Published 22 February 2003) Cite this as: BMJ 2003;326:422
  1. Elisabeth Horak, research fellowa,
  2. Anna Lanigan, research assistanta,
  3. Mary Roberts, research assistanta,
  4. Liam Welsh, research assistanta,
  5. John Wilson, associate professorc,
  6. John B Carlin, deputy directorb,
  7. Anthony Olinsky, directora,
  8. Colin F Robertson, associate professor (colin.robertson{at}
  1. a Department of Respiratory Medicine, Royal Children's Hospital, Parkville 3052, Australia
  2. b Clinical Epidemiology and Biostatistics Unit, Murdoch Children's Research Institute, Parkville 3052, Australia
  3. c Department of Respiratory Medicine, Alfred Hospital, Melbourne, Australia
  1. Correspondence to: C F Robertson
  • Accepted 7 November 2002

Longitudinal studies have reported that asthma in childhood has a good prognosis. However, most of these studies have not taken into account the severity of childhood symptoms.1 The Melbourne Epidemiological Study of Childhood Asthma recruited children at age 7 years and followed them up through adolescence to adulthood.25 This report describes outcome at age 42 years in relation to symptoms in childhood.

Participants, methods, and results

In 1964, 401 children (295 with asthma and 106 controls) were randomly selected from a total of 30 000 7 year olds living in metropolitan Melbourne. A further 83 children with severe asthma were included from the same cohort in 1967, at age 10. 2 3 Original data were available for 479 participants.

At recruitment, 105 children were classified as controls (children who had never wheezed); 74 had mild wheezy bronchitis (<5 episodes of wheezing associated with respiratory tract infection); 104 had wheezy bronchitis (≥5 episodes of wheezing associated with respiratory tract infection); 113 had asthma (wheezing unassociated with respiratory tract infection); and 83 had severe asthma (onset of asthma symptoms before 3 years of age, persistent symptoms at age of 10, and barrel chest deformity or ratio of forced expiratory volume in one second to forced vital capacity ≤50%).

At each review from the age of 21, participants were classified as follows: no recent asthma (no wheeze in past three years); infrequent asthma (wheezing in past three years but none in past three months); frequent asthma (wheezing in past three months, but less than once a week); or persistent asthma (wheezing in past three months, more than once a week).

Fifteen of the original cohort had died at follow up, one from asthma. Of the remaining 464, 403 participated in the current review, giving a continuing participation rate of 87%. In all, 267 participants attended the laboratory for measurement of lung function. We calculated mean values of lung function using standard two sample t tests and confidence intervals of the mean by standard methods.

The table shows the clinical expression of asthma at age 42 according to severity of disease at recruitment. The distribution of severity at age 42 has not changed from that at age 35.5 The proportion of cases with no recent asthma has increased steadily from 20% at age 14 years to 40% (126/317) at age 42.

Distribution of asthma and lung function in participants aged 42 according to severity of asthma at age 7 or 10

View this table:

Lung function was similar to that of controls in participants who had had wheezy bronchitis in childhood (table). Participants who had had asthma aged 7 had reduced lung function at age 42.


Our study shows that the pattern of asthma during childhood predicts outcome. Most children with persistent asthma had continuing symptoms into adult life and reduced lung function. However, children who had intermittent symptoms associated with respiratory tract infections generally had complete resolution of symptoms in adult life. The small number of participants who still had mild, intermittent symptoms at age 42 had normal lung function. This good outcome was achieved despite the fact that anti-inflammatory treatments were not available for most of their childhood.


Contributors: CFR, AO, and JW initiated the project and, together with EH, AL, MR and LW, developed the protocol. EH, AL, MR, and LW were responsible for recruitment, data collection, and data analysis. JBC was the statistician. The manuscript was jointly written and reviewed by all of the authors. CFR is the guarantor.


  • Funding National Health and Medical Research Council of Australia. EH was funded by Nationalbank, Austria.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract