16. Reproductive health outcomes
16.1 Aim
The aim of this analysis is to investigate whether male Australian Gulf War veterans report higher than expected adverse reproductive outcomes, or adverse health outcomes in their live children, following the period of the Gulf War.
16.2 Research questions
1. Is there an increased risk of fertility difficulties in Australian Gulf War veterans following the period of the Gulf War?
2. Is there an increased risk of miscarriage, stillbirth or termination amongst pregnancies fathered by male Gulf War veterans following the period of the Gulf War?
3. Are the children of male Gulf War veterans, born after the period of the Gulf War, at greater risk of being born prematurely, having a low birth weight, having a cancer, having a birth defect or dying?
16.3 Methods and materials
Questions relating to fertility difficulties, reproductive outcomes and the health of live children were included in the postal questionnaire.
Fertility difficulties were defined as difficulties getting pregnant or fathering a pregnancy despite trying for at least 12 months. The year that these difficulties began was reported. Participants were asked whether they had sought or undertaken infertility treatment, whether a cause for the fertility difficulties had been found and whether they had since become pregnant or fathered a pregnancy.
For all fathered pregnancies for all participants (irrespective of fertility status), subjects were asked to report whether the pregnancy resulted in a live birth, miscarriage, stillbirth or termination and to provide the date for each of these outcomes. Participants were not asked to provide reasons or causes for miscarriage, stillbirth or terminations. In relation to live births, study participants were asked to report birth weight and duration of pregnancy. Participants were also asked to report whether any live born child had a cancer, birth defect or chromosomal abnormality or other serious health problem and to identify these. Further, participants were also asked to report whether any live born child had subsequently died and the reason for this.
Reported birth defects and chromosomal abnormalities were screened by the study team and excluded if they were clearly misclassified. Reports of ‘other serious health problems’ were also reviewed by the study team. Some of these health problems were identified as probable birth defects and were included as such.
Birth weights were categorised as “low birth weight” if reported to be < 2500 grams. Of those categorised as “low birth weight”, some were further categorised as “very low birth weight” if reported to be < 1500 grams.
Births were categorised as “premature” if the term of pregnancy was reported to be £ 36 weeks.
Reported cancers were checked against the records of the National Cancer Statistics Clearing House (NCSCH) when sufficient information was given regarding the identity of the child and when consent was provided for matching the information with the NCSCH.
To identify those outcomes most likely to have occurred in the period following the Gulf War, fertility difficulties and adverse reproductive outcomes reported as occurring in 1991 or later were identified. Further, the health of live children was assessed and reported only for those born in 1992 or later.
16.4 Results
Reproductive health outcomes were assessed for all 1424 Gulf War veterans and 1548 comparison group subjects who completed the study postal questionnaire. However, there were some missing data for various items within the reproductive health section of the questionnaire, and therefore the total number of participants contributing information varies within the tables presented below.
16.4.1 Fertility difficulties
In total, approximately 14% of Gulf War veterans and 13% of the comparison group reported fertility difficulties in their lifetime. In Table 16.1 these are divided in to those who reported first experiencing fertility difficulties prior to 1991, and those who reported first experiencing fertility difficulties in 1991 or later. Gulf War veterans were no more likely than the comparison group to have reported experiencing fertility difficulties prior to 1991, but more likely to have reported these occurring for the first time in the period 1991 or later. Of those subjects in both groups who reported fertility difficulties in the period 1991 or later, approximately half had sought infertility treatment and, of these, more than half had found causes for their fertility difficulties. Gulf War veterans who reported fertility difficulties in the period 1991 or later, were more likely than comparison group subjects to report that they had since fathered a child.
|
GWV
|
Comp grp
|
|||||||
|---|---|---|---|---|---|---|---|---|
|
n
|
(%)
|
n
|
(%)
|
Crude OR
|
Adj OR
|
95% CI
|
P value
|
|
|
N=1378
|
N=1504
|
|||||||
| Experienced difficulty getting pregnant pre-1991 |
65
|
(4.7)
|
92
|
(6.1)
|
1.1
|
-
|
-
|
-
|
|
N=1313*
|
N=1412*
|
|||||||
| First experienced difficulties in 1991 or later |
130
|
(9.9)
|
102
|
(7.2)
|
1.4
|
1.4
|
1.0-1.8
|
0.032
|
| Sought infertility treatment |
53
|
(4.0)
|
47
|
(3.3)
|
1.2
|
1.2
|
0.8-1.8
|
0.406
|
| Cause for infertility found |
31
|
(2.4)
|
25
|
(1.8)
|
1.3
|
1.2
|
0.7-2.1
|
0.431
|
| Fathered a pregnancy since then |
85
|
(6.5)
|
52
|
(3.7)
|
1.8
|
1.8
|
1.3-2.6
|
0.001
|
* Persons reporting fertility difficulties prior to 1991 are excluded.
back to top16.4.2 Pregnancy outcomes
Live births, miscarriages, stillbirths and terminations reported to have occurred in 1991 or later are shown in Table 16.2. The pattern of these pregnancy outcomes in the two study groups is very similar. Live births were the outcome for more than 80% of pregnancies in each group. Still births and miscarriages combined represented approximately 14% of the pregnancy outcomes in both groups and terminations represented just under 5% of the pregnancy outcomes in both groups.
The 1170 and 1272 live births were reported by 684 Gulf War veterans (48%) and 732 comparison group subjects (47%) (data not shown). These two groups both reported an average of 1.7 live births.
|
GWV
|
Comp grp
|
||||||||
|---|---|---|---|---|---|---|---|---|---|
|
Pregnancy outcomes in 1991 or later
|
n
|
(%)*
|
n
|
(%)*
|
Crude OR†
|
Adj OR
|
95% CI†
|
P value‡
|
|
| Number of pregnancies |
N=1448
|
N=1555
|
|||||||
| Live births |
1170
|
(80.8)
|
1272
|
(81.8)
|
-
|
-
|
-
|
-
|
|
| Miscarriages |
204
|
(14.1)
|
197
|
(12.7)
|
 |
1.1
|
1.1
|
0.8-1.3
|
0.709
|
| Stillbirths |
5
|
(0.4)
|
14
|
(0.9)
|
|||||
| Terminations |
69
|
(4.8)
|
72
|
(4.6)
|
1.0
|
1.0
|
0.7-1.5
|
0.960
|
|
* All percentage values are derived from N=1448 pregnancies
in the Gulf War veterans column and N=1555 pregnancies in the comparison
group column.
† Odds ratios are for miscarriages/stillbirths or terminations compared
to live births. Odds ratios obtained by polytomous logistic regression
adjusting for age, rank, service type, education, marital status, smoking
and alcohol.
‡ CI and P values are adjusted for clustering of multiple pregnancies
within individuals.
16.4.3 Health of live born children
Health related outcomes for children born in 1992 or later are shown in Table 16.3.
|
Gulf War veterans
|
Comp. group
|
|||||||
|---|---|---|---|---|---|---|---|---|
|
Children born in 1992 onwards
|
n
|
(%)
|
n
|
(%)
|
Crude OR
|
Adj OR
|
95% CI‡
|
P value‡
|
| Number of children |
N=1096
|
-
|
N=1145
|
-
|
||||
| Low birth weight |
56
|
(6.3)*
|
65
|
(6.9)*
|
0.9
|
0.9
|
0.6-1.3
|
0.503
|
| Very low birth weight |
9
|
(1.0)*
|
9
|
(1.0)*
|
1.1
|
1.0
|
0.4-2.8
|
0.992
|
| Premature birth |
71
|
(7.3)†
|
94
|
(9.4)†
|
0.8
|
0.7
|
0.5-1.1
|
0.097
|
| Reported cancer |
4
|
(0.4)
|
1
|
(0.1)
|
4.2
|
-§
|
0.5-38.0
|
0.196
|
| Reported birth defect |
40
|
(3.6)
|
38
|
(3.3)
|
1.1
|
1.0
|
0.6-1.6
|
0.967
|
| Reported death |
2
|
(0.2)
|
4
|
(0.3)
|
0.5
|
-§
|
0.1-2.9
|
0.463
|
* These percentages are derived from N= 885 live births
for Gulf War veterans and N= 936 live births for the comparison group
for whom birth weight was provided.
† These percentages are derived from N= 978 live births for Gulf
War veterans and N= 997 live births for the comparison group for whom
duration of pregnancy was provided.
‡ Standard errors adjusted for clustering of multiple births within
the same individual.
§ Due to the small numbers of events, this odds ratio adjusting for confounders
and clustering was not computed. The associated 95% CI is for the crude
odds ratio adjusted for clustering.
The 1096 and 1145 live births were reported by 665 Gulf War veterans (47%) and 687 comparison group subjects (44%). The children in both study groups were reported to have a very similar pattern of birth weight and duration of gestation and a very similar pattern of birth defects. The quality of the self-reported birth defect data was not high enough to confidently categorise the birth defects into major or minor congenital defects. The total numbers of cancers and deaths were very small in both groups.
back to top16.4.4 Validation of reports of children with cancer
The validation of reports of children with cancer included all children regardless of their year of birth. In total, Gulf War veterans reported six children with cancer and the comparison group reported eight children with cancer.
One comparison group subject did not give consent for his child’s details to be matched with the NCSCH. Therefore the identification details for the six Gulf War children, and for seven of the comparison group children, were sent to the Australian Institute of Health and Welfare (AIHW) for matching with the NCSCH.
The NCSCH confirmed cancers in five of the six Gulf War children, and in four of the seven comparison group children.
Of the four Gulf War children born in 1992 or later and reported to have cancer, three were confirmed by the NCSCH match. The one comparison group child born in 1992 or later and reported to have a cancer, was not confirmed by the NCSCH match.
16.5 Discussion
Gulf War veterans reported more fertility difficulties than the comparison group in the period following the Gulf War, but this group also reported greater success in subsequently fathering a child. The two study groups reported similar rates of pregnancies and live births in the period since the Gulf War, and there appear to be no apparent differences in the risk of still births, miscarriages or terminations reported for that time period. Similarly, there appear to be no differences in risk of birth defects or congenital malformations reported between the two study groups. The numbers of reported cancers and deaths in children were too few in both groups to be meaningfully interpreted.
Using a similar, though stricter definition for fertility difficulties, a study of Western Australian couples reported the prevalence of lifetime ‘infertility’ to be approximately 19%;[392] 5%-6% higher than that reported by both of our study groups. In comparison with self reported data from US Gulf War veterans,[239] Australian Gulf War veterans reported similar live birth-to-pregnancy ratios but lower rates of still births, premature birth and birth defects. When compared with US Gulf War veteran data collected from birth registries,[190, 191] Australian veterans have similar or lower rates of low birth weight, premature birth and birth defects. There are few other sources of comparable data on infertility and other reproductive outcomes in the Australian population or international Gulf War populations.
All findings in relation to reproductive health outcomes in this study should be interpreted with some caution. The numbers of adverse reproductive outcomes reported by the study groups for the period following the Gulf War are quite small, limiting the power of the study to identify small differences in risk. The conclusions are also limited by the study’s primary reliance on self-reported data, an inability to directly assess the female partners of study participants, an inability to control for various genetic, health and psychosocial factors involved in fertility and reproductive outcomes and a paucity of comparable Australian normative data.
The self-reported data, provided in relation to reproductive outcomes in this study, was often only partially complete. Many participants did not provide clear information about dates of pregnancy outcomes or types of cancers or birth defects. Even after thorough follow-up attempts to re-contact study participants about missing data, there often remained large numbers of cases where the information provided was of insufficient quality to be included in the study results.
Large US studies on birth defects in children of Gulf War veterans have shown differing results depending on whether the data was self-reported or collected from medical records or birth defect registries. The self-reported data collected by Kang et al[239] showed an increased risk of birth defects associated with Gulf War service, whereas Cowan et al[190] and Araneta et al[191] found no increased risk when data was drawn from medical records and birth defect registries. Kang et al concluded that there was some over reporting of birth defects in both study groups, possibly because the study did not predefine or provide examples of birth defects. The rate of over reporting, however, was assessed to be slightly higher in the Gulf War veterans group.[239]
Self reported information about reproductive outcomes has been shown to be poorer in men than in women. A study of 857 US couples concluded that husbands substantially misreported their wives reproductive histories.[393] The couples only recorded 88.5% agreement on numbers and dates of live births. Men were shown to misreport the prevalence of low birth weight (sensitivity 74%), spontaneous abortion (sensitivity 71%) and induced abortion (sensitivity 35%). The quality of the reproductive health data provided by male participants in our study is unknown, however it is unlikely misreporting of this nature would be different between Gulf War veteran males and comparison group males. The extent to which male participants sought the advice of their female partners when completing the reproductive history section of the postal questionnaire is also unknown.
Whilst our study attempted to validate self-reported cancers in children of study participants, this included only five children born following the period of the Gulf War, of which three cancers were confirmed. Where cancers were not confirmed, reasons for this could include insufficient information about the child’s name, date of birth, subsequent change of name, State and date of diagnosis. The diagnosis could have been too recent to be found in the NCSCH which, at the time of matching, only held data on known cancers to the end of 1998. In general, cancers reported in children of Gulf War veterans were more likely to be confirmed than cancers reported in children of the comparison group, implying that Gulf War veterans are not over reporting this health outcome.
Similar to the findings reported by Kang et al[239] some over-reporting of birth defects could be expected in both of our study groups as the postal questionnaire did not pre-define or provide examples of birth defects. Australian national data estimates the rates of birth defects or congenital malformations between 1981 and 1996 to have been 160.1 per 10,000 live births (1.6%),[394] a rate half of that reported by our study groups. However this national data set only included major birth defects. Whilst the study team attempted to identify major birth defects amongst all defects reported, the quality of the self-reported data was too poor for this to be undertaken with certainty. Therefore all self-reported birth defects were included in the results unless they were assessed as being clearly not a birth defect. Our study did not seek medical confirmation of the reports of birth defects.
Multiple factors may contribute to infertility and adverse reproductive health outcomes and these include the age and health of female partners. Other factors include any history of conditions such as endometriosis, pelvic inflammatory disease, disorders of ovulation, early menopause or congenital abnormalities. The inability of the study to collect this type of information substantially limits our ability to compare our results with other studies or to draw firm conclusions about possible differences between the two study groups.
There are other possible avenues of research which would more conclusively assess the risk of adverse reproductive health in Gulf War veterans. Information about female partners could only be effectively sought with a further study of randomly selected participants and their partners. Medical records could be sought and birth defects registers accessed in an attempt to validate birth defects and other adverse reproductive outcomes or health outcomes in children. The process of matching reports with the state-based birth defects registers is complicated and the reported data is probably of insufficient quality to allow this. Further, there is not a central national birth defects register that contains identifying details which can be used, and the various state and territory registers which do exist have not employed a standardised approach to data collection or coding.
16.5.1 Summary of findings
With due consideration of the limitations placed on interpretation of our data, the study can make some general statements in relation to the research questions posed.
Male Gulf War veterans appear to be at slightly greater risk of fertility difficulties following the period of the Gulf War. However, this group is more successful at subsequently fathering a child and the two groups report similar rates of pregnancies and live births.
There appears to be no increased risk of miscarriage, stillbirth or termination amongst pregnancies fathered by male Gulf War veterans following the period of the Gulf War.
Children of male Gulf War veterans, born after the period of the Gulf War, do not appear to be at greater risk of being born prematurely, having a low birth weight or having a birth defect or chromosomal abnormality. The numbers of cancers and deaths in children are too small to make any firm conclusions at this stage, however early indications are that there is no increased risk in the children of Gulf War veterans.
back to top