7.0 Female Vietnam Veterans Survey Findings and Community Comparisons
7.1 General
7.2 Comparative Community Data
7.3 Interpretation of Community Data
7.4 Statistical Method
Part A: Medical History
Part B: Marital History
Part C: Health of Partner(s)
Part D: Children
Part E: General
7.1 General
Back to top
This section of the report contains the female veterans survey data shown in comparison to expected community data. The results are presented in the same order as the questions appeared in the survey questionnaire. Where appropriate, recommendations are made for action to be taken in regard to particular findings. This report will be referred to the Repatriation Medical Authority (RMA) for information.
The findings in this section are the result of initial general analysis. Analysis of the survey data to follow more specific lines of inquiry has not yet been undertaken. Generally the results of this study should be considered to be indicative rather than definitive. The comparisons of community data with the survey results should be interpreted with caution.
Any recommendations made in this section in regard to compensation refer only to the operation of the VEA.
Reference is made in some recommendations to Statements of Principles (SoPs). SoPs are instruments produced by the RMA under the auspices of the VEA. They delineate, for the purpose of claiming compensation, causal factors that must exist and be linked to the service of a veteran for a particular disability (or death) to be accepted as war caused.
It should be noted that the US Department of Veterans Affairs has commissioned a study into the health of US female Vietnam veterans. The US study is to be released in 1999. Once released, the results of the US study will be considered by DVA. This may lead to additional research into the health of Australian female Vietnam veterans.
7.2 Comparative Community Data
Comparative community data for this section were obtained by Dr Charles Guest of the National Centre for Epidemiology and Population Health at the Australian National University.
Dr Guest analysed the data and provided comparisons between the number of veterans reporting each condition and the expected prevalence in a representative community sample of the same age and sex.
7.3 Interpretation of Community Data
7.3.1 Sampling Difficulties
In considering the results of the comparison, readers need to recognise the sampling difficulty encountered in this survey. The difficulties in obtaining current addresses for female veterans meant that locations were found for only 57% of the total cohort.
There is a higher than expected number of single rather than married/ de facto female Vietnam veterans in the sample. As health status is linked to marital status, this has had an unknown but probably major impact on the outcome of the survey. The relatively small size of the total female veteran cohort further amplifies the uncertainties of interpretation introduced by this possible bias.
The results of this study are drawn from the 223 completed questionnaires that were received from participants. This number represents only 46% of the female Vietnam veterans recorded on the Nominal Roll 54. In considering the comparisons in this section, it is important to note that the sample may provide a less accurate picture than would have been found if a greater proportion of the female veteran cohort had been surveyed.
7.3.1.1 Implications of Sampling Difficulties
In recognition of the sampling difficulties outlined above, it is recommended that DVA develop a register of information on all female Vietnam veterans. This register would provide a sound basis for any further studies into the health of Australian female Vietnam veterans.
Development of a register would involve locating, to the extent possible, all female Vietnam veterans. This process would enable the profile for the cohort to be expanded, providing the mailing addresses of female Vietnam veterans who were not found for this current survey. It would also identify the number of deceased female Vietnam veterans and their causes of death.
7.3.2 Accuracy of Comparisons
A number of difficulties were encountered when comparing data from other community surveys with the data from the female veterans survey.
The first is that of relevance of community data. The most relevant comparisons would come from surveys where the definition of a condition is compatible with that used in the veteran survey, and where a match in the age of participants, their socio-economic status (including occupation), and the time period under consideration could be made. These criteria were generally not able to be fully satisfied.
Another difficulty is that Australian data were not always available. When Australian data were not found, international comparisons have been offered where possible, or it is noted that comparative community data were simply not available.
The questions asked in other community health surveys were often not directly comparable with questions from the female veterans questionnaire. This survey asked female veterans if 'since your first day of service in Vietnam, you [have] ever been told by a doctor that you have any of the following medical conditions...'. Thus the veteran survey results show the overall prevalence of conditions post-Vietnam, including an unknown proportion which would have been cured since the time of diagnosis. Community surveys typically do not seek such a long-term prevalence, and instead record a present-time 'snapshot' of prevalence, or the prevalence over a specific interval, often 12 months. Where possible, allowances have been made for this by selecting comparable data, by calculations (for example by age adjustment or adjusting incidence to prevalence), or in the interpretation offered.
Another point of contention is that some of the medical conditions in the survey can be defined or measured in a number of different ways. The varying definitions used in the reference surveys and the veterans health study make direct comparisons of some of the results problematic. Care has been taken to ensure that the comparisons presented are based on compatible definitions, but in cases where the most appropriate definition is uncertain, a number of alternative comparisons have been listed. Inconsistency of definition undoubtedly leads to biases which are unable to be quantified.
Further discussion of the possible biases in the comparative estimates of prevalence can be found with the comments on particular conditions.
7.3.3 Demographic Factors
The women who responded ranged in age from 44 to 79 years (median 56, mean 57, standard deviation 7 years). Their median year of birth was 1941. Fifty per cent responded that they were currently married or living in a de facto relationship and just under one third had never married. Half of the women reported having children. Thirty per cent responded that they were in good, very good or excellent health.
The branches of the armed services to which these veterans belonged are known, but specific information about occupational duties was not elicited in this survey. Moreover, information on occupation prior to, or since leaving, Vietnam is not available.
7.4 Statistical Method
To allow for the small size of the veteran sample and the varying accuracy of the community data, ranges have been calculated for both the observed and expected numbers of the various conditions. These ranges, corresponding to 95% confidence intervals, give a better indication of the prevalence than simply considering the number found in the sample and the number calculated for the comparison populations. If the range is wide, this indicates that variation by chance is important.
The statistical comparisons of the observed and expected rates are made by inspecting the ranges for each rate. The difference between what is observed and expected has been interpreted as statistically significant when the 95% confidence intervals do not overlap.
A more complete explanation of the statistical method used in interpreting the comparative data is at Appendix D.
| Return to Contents | | Go to Part A 
|