Table 4 Summary of estimates of risk for congenital anomalies associated with Agent Orange exposure in Vietnam war
|
|
|
Measure of heterogeneity
|
|
| Subgroup |
Number of studies |
Summary OR or RR (95% CI) |
Q-value |
P-value |
I2 |
Analysis model |
|
| All studies |
22 |
1.95 (1.59–2.39) |
163.0 |
<0.0001 |
0.87 |
RE |
Study design
|
| Cohort |
16 |
2.16 (1.68–2.76) |
119.0 |
<0.0001 |
0.87 |
RE |
| Case–control |
5 |
1.40 (0.95–2.06) |
20.0 |
<0.0001 |
0.80 |
RE |
| Cross-sectional |
1 |
3.00 (2.02–4.47) |
NA |
NA |
NA |
NA |
Study population
|
Vietnamese
|
| Cohort |
10 |
3.11 (2.11–4.61) |
50.26 |
<0.0001 |
0.82 |
RE |
| Case–control |
2 |
3.55 (2.06–6.14) |
0.00 |
0.965 |
. |
FE |
| Cross-sectional |
1 |
1.97 (1.31–2.96) |
NA |
NA |
NA |
NA |
| Total |
13 |
3.00 (2.19–4.12) |
54.6 |
<0.0001 |
0.78 |
RE |
International veterans
|
| Cohort |
6 |
1.45 (1.07–1.96) |
45.63 |
<0.0001 |
0.89 |
RE |
| Case–control |
3 |
0.99 (0.87–1.13) |
0.3 |
0.8607 |
0.00 |
FE |
| Total |
9 |
1.29 (1.04–1.59) |
55.0 |
<0.0001 |
0.85 |
RE |
| Total (Lathrop22 removed) |
8 |
1.28 (1.02–1.61) |
53.5 |
<0.0001 |
0.87 |
RE |
Exposure
|
| International non-Ranch Hand veteransa |
6 |
1.04 (0.93–1.16) |
5.14 |
0.273 |
0.03 |
FE |
| Ranch Hand veteransb |
3 |
1.20 (1.08–1.34) |
2.559 |
0.278 |
0.46 |
FE |
| Ranch Hand veteransb (Lathrop22 removed) |
2 |
1.18 (1.05–1.32) |
1.688 |
0.194 |
0.41 |
FE |
| North Vietnam veteransb |
7 |
2.61 (1.72–3.95) |
29.82 |
<0.0001 |
0.80 |
RE |
| North Vietnam veteransb (Can et al.31 removed) |
6 |
2.75 (2.17–3.49) |
8.32 |
0.1395 |
0.40 |
FE |
| Sprayed civiliansc |
6 |
3.27 (2.54–4.10) |
5.41 |
0.3679 |
0.08 |
FE |
|
RE: Random-effects model; FE: fixed-effects model; I 2: Index of inconsistency.
aNo or minimal exposure.
bFather exposed.
cBoth parents exposed.
