Combined Impact of Smoking and Early-Life Exposures on Adult Lung Function Trajectories

Abstract

Rationale: Both adverse early-life exposures and adult smoking can negatively influence adult lung function trajectory, but few studies consider how the impact of early-life exposures may be modified by subsequent smoking.

Methods: The Medical Research Council National Survey of Health and Development is a nationally representative cohort, initially of 5,362 individuals, followed since enrollment at birth in March 1946. Using data collected prospectively across life and multilevel modeling, we investigated how the relationships between early-life exposures (infant lower respiratory infection, manual social class, home overcrowding, and pollution exposure) and FEV₁ and FVC trajectories between ages 43 and 60-64 years were influenced by smoking behavior.

Measurements and main results: Among 2,172 individuals, there were synergistic interactions of smoking with infant respiratory infection (P = 0.04) and early-life home overcrowding (P = 0.009), for FEV₁ at 43 years. Within smoker-stratified models, there were FEV₁ deficits among ever-smokers associated with infant lower respiratory infection (-108.2 ml; P = 0.001) and home overcrowding (-89.2 ml; P = 0.002), which were not evident among never-smokers (-15.9 ml; P = 0.69 and -13.7 ml; P = 0.70, respectively). FVC modeling, including 1,960 individuals, yielded similar results. FEV₁ decline was greater in smokers (P < 0.001), but there was no effect of any early-life exposure on FEV₁ decline. Neither smoking nor early-life exposures were associated with FVC decline.

Conclusions: Besides accelerating adult FEV₁ decline, cigarette smoking also modifies how early-life exposures impact on both midlife FEV₁ and FVC. These findings are consistent with smoking impairing pulmonary development during adolescence or early adulthood, thereby preventing catch-up from earlier acquired deficits.

Keywords: COPD development; childhood respiratory infections; chronic obstructive pulmonary disease; infancy.

Figure 1.

The relationship between infant lower respiratory infection and father’s occupational class, home overcrowding, and high pollution exposure during early life among those individuals providing complete data during early life (n = 2,172) within the Medical Research Council National Survey of Health and Development. (A) Venn diagram representing the overlapping prevalence (%) of infant lower respiratory infection, father’s occupational class, home overcrowding, and high pollution exposure during early life. Numbers shown indicate percentages of the included population (n = 2,172). *Represents those with no lower respiratory infection, father from nonmanual occupational class, no home overcrowding, and low pollution exposure during early life. (B) Associations of infant lower respiratory infection, by father’s occupational class, home overcrowding, and pollution exposure adjusting for male sex and birth weight. Odds ratios (95% confidence intervals) of having an infant lower respiratory infection according to the presence of each factor were calculated by multiple logistic regression. CI = confidence interval; OR = odds ratio.

Figure 2.

A comparison of the estimated average (A) FEV₁ and (B) FVC trajectories between ages 43 and 60–64 years for males of average height at age 43 years (175 cm), average weight at age 43 years (78 kg), and average birth weight (3.5 kg) according to adult smoking behavior and early-life disadvantage. Estimated level and slope between 43 and 60–64 years of age was calculated using multilevel models (Tables 1 and 2). Predicted FEV₁ and FVC values according to age were calculated as per Reference . Never-smoker = never smoked up to age 60–64 years. Adult smoker = smoked 20 cigarettes per day from age 20 until 60–64 years. Nondisadvantaged early life = no lower respiratory infection, father in nonmanual social class, nonovercrowded home, and low pollution exposure during early life. Maximally disadvantaged early life = lower respiratory infection present, father in manual social class, overcrowded home, and high pollution exposure during early life.

Figure 3.

A comparison of the estimated pattern of FEV₁ decline in relation to FVC decline between ages 43 and 60–64 years for males of average height at age 43 years (175 cm), average weight at age 43 years (78 kg), and average birth weight (3.5 kg) according to adult smoking behavior and early-life disadvantage. Estimated level and slope between 43 and 60–64 years of age were calculated using multilevel models (Tables 1 and 2). Blue-shaded area indicates zone within which FEV₁ values would meet traditional airflow limitation criteria (FEV₁/FVC < 0.7) associated with chronic obstructive pulmonary disease diagnosis. See Figure E5 for accompanying FEV₁/FVC plots. Never-smoker = never smoked up to age 60–64 years; adult smoker = smoked 20 cigarettes per day from age 20 until 60–64 years; nondisadvantaged early life = no lower respiratory infection, father in nonmanual social class, nonovercrowded home, and low pollution exposure during early life; maximally disadvantaged early life = lower respiratory infection present, father in manual social class, overcrowded home, and high pollution exposure during early life.

Figure 4.

The relationship between the deficit in adult FEV₁ associated with early-life respiratory infection and the prevalence of ever-smoking among study members. The data shown are taken from the National Survey of Health and Development and five other major prospective studies. The graph suggests that the adult FEV₁ deficit associated with infant pneumonia (red markers) increases in magnitude as the prevalence of smoking within a study increases. A similar trend is seen across studies reporting the decrement associated with a milder definition of infant respiratory infection (blue markers), including “bronchitis” and “respiratory tract infection” rather than just pneumonia. For details of included studies, see Figure E5. Lines were fitted by metaregression. NSHD = National Survey of Health and Development.