A cohort of 2,289 children, previously studied at the age of 6-8 yr, were followed up by means of a postal questionnaire when aged 14 -16 yr to examine the association between potential risk factors and the natural history of respiratory symptoms. Children with current symptoms, persistent symptoms, and late-onset symptoms were identified and multivariate analyses were performed to determine the independent association between risk factors and these various symptom-based subgroups. Personal and family history of atopy was significantly associated with all symptom groups and with the presence of doctor-diagnosed asthma. Smoking, either active or passive, was shown to be significantly associated with current, persistent, and late-onset symptoms. Other factors shown to be significantly associated with certain symptom groups were gender (late-onset wheeze), single-parent households (current cough, persistent cough), social class (late-onset wheeze), number of children in the household (persistent wheeze, late-onset cough), number of furry pets in the household (current wheeze), birth weight (late-onset wheeze), and gas cookers (current wheeze, persistent wheeze). In a subgroup of children studied in more detail in 1987, bronchial hyperresponsiveness in 1987 was positively associated with persistent wheeze in 1995, whereas positive skin-prick testing in 1987 was not.