Table 5.
Summary estimates, 95% CIs and data descriptions for household SHS exposure comparisons grouped by type of asthma.a
Asthma type
|
Percentile
|
|||||||
---|---|---|---|---|---|---|---|---|
Type of combined estimate of RRb | No. | RR | 95% CI | p-Value | Min | 25th | 75th | Max |
Current | ||||||||
Uncorrected | 15 | 1.30 | 1.22–1.39 | 0.249* | 0.86 | 1.24 | 1.60 | 2.17 |
Correctedc | 11 | 1.25 | 1.21–1.30 | 0.528 | 0.88 | 1.23 | 1.38 | 2.17 |
Ever | ||||||||
All studiesc,d | 23 | 1.48 | 1.32–1.65 | <0.001* | 0.57 | 1.45 | 1.74 | 5.81 |
Household SHS | 17 | 1.51 | 1.31–1.75 | <0.001* | 0.94 | 1.47 | 1.73 | 5.81 |
Maternal smoking | 6 | 1.29 | 1.15–1.45 | 0.190* | 1.21 | 1.21 | 1.98 | 2.77 |
Incident | ||||||||
All studies | 8 | 1.21 | 1.08–1.36 | 0.225* | 0.84 | 1.08 | 1.33 | 1.49 |
Household SHS | 4 | 1.13 | 0.89–1.44 | 0.544 | 0.84 | 0.92 | 1.26 | 1.27 |
Maternal smoking | 4 | 1.24 | 1.06–1.45 | 0.074* | 1.08 | 1.08 | 1.45 | 1.49 |
Abbreviations: CI, confidence interval; Max, maximum value of the distribution; Min, minimum value of the distribution; p-value, value from the p distribution for the null hypothesis that the rate ratio is constant across studies.
We report random effects estimates.
We report corrected estimates unless otherwise indicated.
We removed four studies from the analysis of current asthma (Azizi et al. 1995; Daigler et al. 1991; Murray and Morrison 1990; Palmieri et al. 1990) and one study from the analysis of ever asthma (Pokharel et al. 2001) for lack of enough information to convert odds ratios into estimates of RR.
Excluded six studies already included in the current asthma analysis.
Heterogeneity.