Risk factors for recurrent cough (RC) in childhood, and its relation to asthma were investigated as part of the prospective, longitudinal Tucson Children's Respiratory Study. RC, defined as > or = 2 episodes of cough without a cold in the past year, was assessed by questionnaire in 987 children at age 6. Children having RC without wheeze (n = 154) did not differ from children with neither symptom (n = 610) in serum IgE levels, skin test response, size-corrected forced expiratory flow, or percentage of decline following cold air challenge. In contrast, children with both RC and wheeze (n = 116) had significantly more respiratory illness, more atopy, lower flow at end-tidal expiration (V'maxFRC), and greater declines in lung function following cold air challenge than children with neither symptom. Current parental smoking was a risk for RC without wheeze, whereas male gender, maternal allergy, wheezing lower respiratory tract illness (LRI) in early life, and high IgE were significant risks for RC with wheeze, compared with children having neither symptom. RC early in life resolved in the majority of children, between ages 2-3 yr and age 6, and between age 6 and age 11. High IgE and positive skin prick test were associated with persistence of RC to age 6 among children who wheezed, and markers of allergy were associated with persistence of RC between 6 and 11 yr. These findings suggest that recurrent cough in the absence of wheeze differs in important respects from classic asthma, and using the same label to refer to these distinct syndromes may obscure their diverse pathophysiologies.