Table 1.
Prospective birth cohort studies examining the effects of childhood pneumonia upon adult lung function
| Study | Country | Year(s) of inception | Year(s) of study | Source of diagnosis | No. with pneumonia | No. with pneumonia studied (%) | No. without pneumonia studied | Ages (years) | Median follow-up (years) | Main findings |
|---|---|---|---|---|---|---|---|---|---|---|
| Barker et al. [23]a | UK | 1920–30 | 1989 | Health visitors prospective reports of illness in the first yr, and between 1 and 5 yrs of life | 122b | 122 (100) | 703 | 59–67 | 60 | Bronchitis or pneumonia in male infants was associated with significantly reduced adult values for FEV1 (−0.17 L; 95% CI −0.32, −0.02) and FVC (−0.24 L; 95% CI −0.24, −0.07) and increased odds of wheezing (OR 1.83 [95% CI 1.05, 3.20]) independent of smoking, birth weight and social class. |
| Shaheen et al. [24]a | UK | 1921–35 | 1985–86 | Doctors’ records for the first 5 yrs of life | 18 | 10 (53) | 229 | 57.6 ± 4.3d | 50 | Pneumonia in the first 2 yrs of life was associated with significantly reduced adult values for FEV1 (−0.39 L; 95% CI −0.67, −0.11; p = 0.007) and FVC (−0.60 L; 95% CI −0.92, −0.28); p < 0.001) independent of age, sex, height, smoking and other illnesses before age 2 yrs. |
| Johnston et al. [25]a | UK | 1958 | 1992–93 | Parents report when child aged 7 yrs | 350 | 193 (55) | 1,199 | 34–35 | 30 | History of pneumonia in the first 7 yrs of life was significantly associated with reduced adult FEV1 (−0.102 L; 95% CI −0.175, −0.029; p = 0.006) and FVC (−0.173 L; 95% CI −0.243, −0.067; p = 0.001) without a change in FEV1/FVC ratios, which were unaltered by bronchodilators and were independent of a history of wheezing. |
| Shaheen et al. [26] | UK | 1917–22 | 1990–91 | Health visitors’ reports of illness collected 3 monthly in first yr, then once yearly | 20 | 20 (100) | 598 | 67–74 | 70 | In men only, pneumonia before age 2 yrs was significantly associated with lower adult FEV1 (−0.65 L; 95% CI −1.02, −0.29, p < 0.001), FVC (−0.51 L; 95% CI −0.94, −0.08; p = 0.02) and FEV1/FVC ratios (−10.9%; 95% CI −16.9, −4.8; p < 0.001) independent of smoking and asthma. |
| Chan et al. [27] | USA | 1980–84 | 1996–2003 | Clinical with CXR (infiltrates, bronchopneumonia or pneumonia) in the first 3 yrs of life | 66 | 44 (67) | 308 | 26–29 | 26 | Those with pneumonia before 3 yrs of age had significantly lower FEV1/FVC ratios (−3.23%; p = 0.034)), FEF25-75 (−0.508 L/s; p = 0.02) and FEF25-75/FVC ratios (−11.14%; p = 0.024) at 26 yrs of age, independent of wheeze and smoking, and which were only partially reversible with bronchodilators. Pneumonia was also associated with a two-fold increased risk of actively managed asthma at age 29 yrs (OR 1.95 [95%CI 1.11, 3.44]; p = 0.02). |
| Lopez-Bernal et al. [28] | UK | 1972–74 | 1997 | Study team or maternal report of respiratory illness in the first 5 yrs of lifec | NR | NR | NR | 25.0 ± 0.7d | 25 | 679 (71%) of original cohort underwent lung function testing. After adjusting for potential confounding factors, lower respiratory tract infections in the first yr of life were negatively associated with all spirometry values, except for FVC, and showed a significant dose-response effect where a two-fold increase in lower respiratory tract infections was associated with reduced FEV1 (−0.078 L; 95% CI −0.153, −0.03), and FEV1/FVC ratios (−1.23%; 95% CI −2.22, −0.25). |
| Colley et al. [29] | UK | 1946 | 1966 | Parent questioned when child aged 2 yrs (pneumonia, bronchopneumonia or bronchitis) | 820 | 820 (100) | 2,502 | 20e | 20 | In ‘never smokers’, prevalence of day/night cough was 9.1% in group with chest infections vs 5.2% without. In smokers, values were 16.5% and 13.5%, respectively. Difference between groups (with and without chest infections) was significant (p < 0.025). Social class and air pollution had no significant effect. |
CI, confidence interval; CXR, chest radiograph; FEV1, forced expiratory volume in one second; FVC, forced vital capacity; FEF25-75, forced expiratory flow between 25% and 75% of FVC; NR, not reported; OR, odds ratio; UK, United Kingdom; USA, United States of America
aIncluded in review by Edmonds et al. [18]
bMales only
cSupplemented by a doctor’s diagnosis when available
dMean ± Standard Deviation
eCohort was born in the last week in March 1946. In 1966, they were all sent a questionnaire at the same time