Longitudinal analysis of the effects of cigarette smoking in children

We performed FVC maneuvres on an 8-L, water-filled spirometer in 674 children who completed standardized respiratory illness questionnaires for at least 2 consecutive visits in years 4 through 7 of the study. FEV₁ was obtained using standard methods. At any given year, 85-90% of children were <14 years old, and approximately 52% were male. Frequency of current smoking increased from 5.7% in year 4 to 9.9% in year 7. A Markov-type, autoregression model was used to analyze the data. This model permits analysis of time-dependent (eg, height, smoking) as well as fixed (eg, sex) variables that may effect change of FEV₁ over time. Based on this model, personal smoking by the children significantly (p<0.001) decreased the rate of increase of FEV₁ after adjustment for height in any year, annual change in height, sex, age and level of FEV₁ in the previous year. In particular, the level of FEV₁ achieved by a smoking child after 2.5 and 5 years was predicted to be lower by 10% and 15%, respectively, compared to that achieved by a comparable nonsmoking child. At age 20, a child who began smoking at age 15 would be expected to have an FEV₁ 0.6 L lower than that of comparable nonsmoker. Data concerning the validation of the model are forthcoming. These data indicate that relatively short periods of cigarette smoking in children may have profound effects on the rate of increase of FEV₁.