1. Are Australian Gulf War
veterans more or less likely, than the comparison group, to develop a
psychological disorder in he period following the Gulf War?
2. Are Australian Gulf War
veterans more or less likely, than the comparison group, to have psychological
disorders which have been present in the previous 12 months?
3. Do Australian Gulf War veterans
score differently on the 12 item General Health Questionnaire when compared
with the comparison group?
4. Do Australian Gulf War
veterans score differently on the Posttraumatic Stress Disorder Checklist
when compared with the comparison group?
5. Do Australian Gulf War veterans
score differently on the Alcohol Use Disorders Identification Test when
compared with the comparison group?
6. Where differences in risk
of psychological disorders occur between Australian Gulf War veterans
and the comparison group, are these associated with differences in exposures
and experiences that occurred during the Gulf War deployment?
back to top
11.3 Methods and materials
The investigation of psychological health included administration of
a comprehensive, psychologist-administered psychological health interview
and several briefer, self-administered questionnaires.
11.3.1 Measurement of psychological health
A complete description of the instruments used in the assessment of psychological
health is provided in chapter 5. The methods for scoring each instrument
are defined below.
11.3.1.1 Composite International Diagnostic Interview, CIDI-Auto
2.1
The prevalence of several affective, anxiety, somatic and substance-use
disorders was assessed according to diagnostic criteria described in the
4th edition of the Diagnostic and Statistical Manual of Mental
Disorders DSM-IV.[308] The instrument used was an interviewer
administered version of the Composite International Diagnostic Interview,
CIDI-Auto 2.1.[306] For each participant, upon interview,
the CIDI-Auto 2.1 output included whether a disorder had been present
or absent and, for those disorders which had been present, the output
included the age of first onset of symptoms (age of onset) and the age
of last symptoms (age of recency). The CIDI output also included a recency
code categorising the time period, prior to the interview, within which
the most recent symptoms were experienced. These recency codes were:
1. within two weeks
2. two weeks to less than one
month ago
3. one month to less
than six months ago
4. six months to less
than one year ago
5. in the last twelve
months, don’t know when
6. more than one year ago
Using the age of onset, age of recency and recency code the study team
classified the present disorders in to the following categories:
- CIDI defined DSM-IV pre-Gulf War disorder: all diagnosed
disorders where age of onset of first symptoms was less than the subject’s
age at 1 August 1990.
Disorders included in this category were those which were first experienced
by the subject prior to the time of the Gulf War. Age of recency was
not considered in this definition; symptoms may have ceased prior to the
Gulf War, or may have continued after the time of the Gulf War.
- CIDI defined DSM-IV post-Gulf War disorder: all diagnosed
disorders where age of onset of first symptoms was greater than or equal
to the subject’s age at 1 August 1990.
Disorders included in this category were those which were first experienced
by the subject during or after the time of the Gulf War. Age of recency
was not considered in this definition; symptoms may have ceased at any
time since the Gulf War, or may have been ongoing at the time of assessment.
- CIDI defined DSM-IV disorder present in previous 12 months:
any pre-Gulf War or post-Gulf War disorder where symptoms have been
present within twelve months of the interview.
Disorders included in this category were those pre-Gulf War and post-Gulf
War disorders with recency codes 1 to 5; excluding those with code 6.
- CIDI defined DSM-IV current disorder: any pre-Gulf War
or post-Gulf War disorder where symptoms have been present within four
weeks of the interview.
Disorders included in this category were those pre-Gulf War and post-Gulf
War disorders with recency codes 1 and 2; excluding codes 3 to 6.
Age at 1 August 1990 was calculated by the study team and rounded down
to whole years; for example, 35.7 years of age was truncated to 35 years
of age.
It is important to note that the CIDI gives only age of onset of first
symptoms, per person, per diagnosis and does not record remissions and
subsequent onset of symptoms for the same diagnosis. Therefore, if age
of onset for a diagnosis was less than the subject’s age at 1 August
1990, the study team could only classify that disorder as “pre-Gulf
War” regardless of whether new symptoms were experienced after the
time of the Gulf War. For a subject to be categorised as having a particular
“post-Gulf War disorder” they could not have had that same disorder
categorised as “pre-Gulf War”. All subjects with pre-Gulf War
diagnoses were excluded from the analyses of post-Gulf War disorders of
the same type.
back to top
11.3.1.2 12-item General Health Questionnaire (GHQ-12)
The GHQ-12 was scored by coding the four possible responses to each question
as 0 – 0 – 1 – 1 and then summing the binary scores.[218] Respondents
were given a total score ranging from a minimum of zero to a maximum of
twelve, with a higher score indicating poorer psychological health.
Several thresholds, or cut-off scores, for determining GHQ-12 caseness,
or possible psychiatric condition, have been employed in the literature.[264,
271] Using the ANSHWB data, Donath reported an optimum threshold
of one or more symptoms for the Australian population.[269]
Earlier Australian studies have reported optimum cut off scores of two
or more[360] and as high as eight or more.[268]
These extremes have been observed by Goldberg[264] in a comparison
of GHQ-12 screening threshold results across fifteen cities. In their
studies of British Gulf War veterans, the King’s College Group employed
a GHQ-12 caseness score of three or more symptoms.[21, 155]
To determine the most appropriate GHQ-12 caseness score in our study,
Receiver Operating Characteristic (ROC) analysis[361] was conducted
to identify the GHQ-12 score with the best specificity and sensitivity
to detect a person with or without any current psychological disorder
(as defined using the CIDI and present within the previous four weeks),
excluding current substance use disorder, alcohol use disorder and specific
phobia. For each possible cut-point (for example, 1, 2 or 3 symptoms
and so on), the sensitivity (Y-axis) and 1-specificity,[362]
otherwise known as the false positive rate (X-axis), were calculated and
plotted. The area under the resulting plotted ROC curve represented the
probability that a randomly chosen subject with any current psychological
disorder would have a higher GHQ-12 score than a randomly chosen subject
without any current psychological disorder. The method for constructing
the area under the ROC curve, and the confidence interval, is described
by Hanley and McNeil.[363]
Using ROC analysis, and assuming equal importance of sensitivity and
specificity, the optimum cut-off point for the GHQ-12 in our study proved
to be two symptoms, with sensitivity 0.73 (95% CI = 0.67 to 0.77) and
specificity 0.68 (95% CI =0.66 to 0.69). The area under the curve was
0.77 (95% CI 0.74 to 0.80). The method used to calculate confidence intervals
for sensitivity and specificity, was described by McKenzie et al[364]
and was conducted using the software of Mackinnon.[365]
11.3.1.3 Posttraumatic Stress Disorder Checklist-S (PCL-S)
The PCL-S is one of three versions of the Posttraumatic Stress Disorder
Checklist (PCL) available.[278] The same standard scoring
method applies to each version of the PCL with a total score computed
by coding the five possible responses to each question as 1 – 2 –
3 – 4 – 5 and then summing the results. With seventeen questions
in total, possible scores range from 17 to 85. In our study the cut-off
score for PCL-S caseness, predictive of possible posttraumatic stress
disorder, was set at >=50. This threshold was recommended by Weathers[278]
and was shown to have a sensitivity of 0.82 and a specificity of 0.83
in predicting posttraumatic stress disorder cases as measured using the
Structured Clinical Interview for DSM-III-R in Vietnam War veterans.
In their study of the validity of the PCL as a measure of symptomatic
change in Australian veterans of the Vietnam War with current posttraumatic
stress disorder, Forbes et al supported the use of the diagnostic
cut-off of >=50, which had a sensitivity of 0.97[366] (specificity
was not applicable as all subjects had posttraumatic stress disorder).
11.3.1.4 Alcohol Use Disorders Identification Test (AUDIT)
The AUDIT was scored by coding the five possible responses to questions
1-8, as 0 – 1 – 2 – 3 – 4, and coding the three possible
responses to questions 9-10 as 0 – 2 – 4. The ten items were
then summed to result in a total score ranging from 0 to 40.[367]
In our study the cut-off score for AUDIT caseness, predictive of problem
drinking, was set at >=8 as recommended by the WHO.[276]
Bohn et a[367] used ROC analysis to investigate different
AUDIT thresholds for optimal detection of alcohol risk groups. They found
that an AUDIT score of >=8 could detect 98% (0.98) of hazardous drinkers;
that is, drinkers who have not yet experienced alcohol-related problems,
yet consume alcohol in patterns that increase the risk of developing such
problems. Specificity for the >=8 threshold, for hazardous drinkers,
was only 34% (0.34). In the detection of harmful drinkers, defined as
those who experience physical or mental harm due to drinking, but who
are not alcohol dependent, the AUDIT threshold of >=8 showed sensitivity
of 0.77 and specificity of 0.81. Bohn et al, however, recommended
the use of a >=10 threshold, which demonstrated poorer sensitivity
of 0.87 in hazardous drinkers and 0.60 in harmful drinkers, but improved
specificity of 0.75 and 0.87 respectively.[367]
Barry and Fleming[277] reported that an AUDIT score of >=5
represented the optimal balance between sensitivity and 1-specificity,
on their ROC curve, for the detection of drinkers who met lifetime DSM-III
criteria for alcohol misuse and or dependence. To minimise the possibility
of false positive tests, however, these authors recommended raising the
cut-off score to 7 or 8.
11.3.2 Gulf War exposure measures
Where differences in risk of psychological disorders exist between Gulf
War veterans and the comparison group, the analysis will investigate associations
with several measures of exposure.
Specifically, the analysis will investigate whether risk of psychological
disorders in either study group differs across subcategories of age, service
type and rank.
Further, the analysis will investigate whether a difference in risk of
psychological disorders exists between Gulf War veterans and those comparison
group subjects who have been on other active deployments.
Finally the analysis will investigate whether risk of psychological disorders
within the Gulf War veteran group differs according to:
- Military Service Experience questionnaire score
- Deployment completed before or after the air war commenced on 17
January 1991
- Total number of immunisations
The scoring and groupings for these exposures are described in chapter
8.
back to top
11.4 Results
11.4.1 CIDI-defined DSM-IV disorders
Subjects included in the analysis of CIDI-defined DSM-IV psychological
disorders were all those males who completed the interviewer administered
psychological assessment. These included 1381 Gulf War veterans and 1377
comparison group subjects. They represented 99.8% of male subjects, from
both groups, who attended HSA for the study’s medical examination.
The results for Gulf War veterans and comparison group subjects who met
criteria for CIDI-defined pre-Gulf War disorders, are shown in Table 11.1.
Prevalence across groups for most pre-Gulf War disorders were similar,
indicating that the two groups varied little in their overall levels of
psychological morbidity prior to the time of the Gulf War deployment.
The results for Gulf War veterans and comparison group subjects, who
met criteria for CIDI-defined post-Gulf War disorders, are also shown
in Table 11.1. The results for CIDI-defined disorders present within
the previous 12 months are presented in Table 11.2.
Gulf War veterans were more likely than the comparison group to develop
most post-Gulf War psychological disorders. Of particular note, the risk
of several post-Gulf War anxiety disorders including posttraumatic stress
disorder, obsessive compulsive disorder and social phobia, was elevated
three to five times in Gulf War veterans. Increased risk for post-Gulf
War bipolar disorder was almost three fold, and risk for post-Gulf War
major depression, alcohol dependence or abuse and drug dependence or abuse
were all more than one and a half times higher in Gulf War veterans.
These increased risks remained statistically significant after adjustment
for service type, rank, age, education level and marital status. Only
when numbers were small, and prevalences were around one percent or less,
did the increased risk in Gulf War veterans not reach statistical significance
for post-Gulf War disorders.
Alcohol dependence and abuse and major depression were the most prevalent
post-Gulf War disorders in both groups. The overall levels of post-Gulf
War somatic disorders were very low, affecting less than one percent of
all subjects. No subjects recorded a diagnosis of somatisation disorder.
In relation to CIDI defined disorders present within the previous 12
months (Table 11.2), anxiety disorders including posttraumatic stress
disorder, obsessive compulsive disorder, social phobia and panic disorder
and agoraphobia, were three to five times more likely in Gulf War veterans.
These associations persisted after adjustment for the presence of the
disorders prior to the Gulf War, in addition to the adjustments for service
type, rank, age, education level and marital status. Bipolar disorder
was more than two times more likely in Gulf War veterans, and major depression
and alcohol dependence or abuse were more than one and a half times more
likely when assessed as present within the previous 12 months.
Gulf War veterans were also more likely than the comparison group to
have more than one CIDI disorder present within the previous 12 months,
with 7% of Gulf War veterans and 3.5% of the comparison group recording
two or more of these disorders. On average, Gulf War veterans had two
times as many disorders present in the previous 12 months as the comparison
group.
In both groups, subjects with ‘any affective disorder’ primarily
comprised those with a major depression and subjects with ‘any
substance use disorder’ primarily comprised those with alcohol
dependence or abuse.
back to top
Table 11.1 Lifetime CIDI-defined DSM-IV disorders first
present prior to the Gulf War, and those newly present after the Gulf
War, in male Gulf War veterans and comparison group subjects.
go to table 11.1
Table 11.2 CIDI-defined DSM-IV disorders present within
12 months prior to assessment, in male Gulf War veterans and comparison
group subjects.
go to table 11.2
Table 11.3 through to Table 11.6 demonstrate the effects of study group
(Gulf War veterans versus comparison group) upon several categories of
CIDI defined post-Gulf War psychological disorder, across subgroups of
age, service type and rank. The disorders considered were ‘any affective
disorder’, ‘any anxiety disorder’, posttraumatic stress
disorder and ‘any substance use disorder’. Age, service type
and rank serve as proxies for potentially differing levels of experience
and exposures. P values for interaction assessed whether the adjusted
odds ratios were consistent across the subgroups of age, service type
and rank.
Within almost every subgroup of age, service type and rank, Gulf War
veterans had higher prevalences than the comparison group in each of the
presented categories of CIDI defined post-Gulf War disorder. Many of
these differences between the two groups, and within the subgroups of
age, service type and rank, remained statistically significant after adjustment
for the other subgroups, with the exception of differences between the
two groups in substance use disorders. There was only a small amount
of variation in the adjusted odds ratios across subgroups of age, service
type and rank and tests for interaction failed to reach statistical significance.
This indicates that the Gulf War deployment did not affect one subgroup
differently to another. For example, the Gulf War deployment is associated
with increased post-Gulf War affective disorders in Gulf War veterans,
however the Gulf War deployment is not more strongly associated with increasing
affective disorders in < 20 year olds than it is in increasing affective
disorders in >= 35 year olds. This finding held for anxiety disorders,
posttraumatic stress disorder and substance use disorders, and also for
the different subgroups of service type and rank.
Subjects classified as ‘other ranks-non supervisory’, compared
with ‘other ranks-supervisory’ and ‘officer’ ranks,
demonstrated the highest prevalences of psychological disorder, for both
study groups and for all broad categories of post-Gulf War disorder.
For most disorder categories, prevalences amongst non-supervisory ranks
in both groups were approximately twice as high as prevalences amongst
officer ranks. The prevalences of post-Gulf War substance-use disorders,
in both study groups, were more than six times higher in the non-supervisory
ranks than in the officer ranks. Officer ranks were found to have the
lowest prevalences in all but one disorder category, with the single exception
of posttraumatic stress disorder, for which supervisory ranks showed the
lowest prevalence. The Gulf War deployment, however, was no more associated
with increasing risk in the lower rank subgroups, than it was with increasing
risk in higher rank subgroups.
Subjects aged <25 years in both study groups, generally recorded more
substance use disorders and more affective disorders than subjects aged
>= 25 years. For these and other disorders however, the Gulf War deployment
did not increase risk in one subgroup of age differently to another.
In relation to service type, post-Gulf War psychological disorders were
generally least prevalent amongst Air Force participants in both study
groups. Post-Gulf War affective disorders and anxiety disorders were
most prevalent in Army participants in both groups when compared with
the other services. In contrast, post-Gulf War substance use disorders
were most prevalent in Navy subjects in both study groups when compared
with the other services. There were no differences in the effect of the
Gulf War deployment across the service types.
back to top
Table 11.3 Any post-Gulf War affective disorder: The effects
of study group across subgroups of age, service type and rank.
| |
GWV |
Comp grp |
|
|
|
|
|
| |
N |
(%) |
n |
(%) |
Crude
OR |
Adj
OR† |
95% CI |
|
P value‡ |
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
38
|
(22.4)
|
13
|
(11.7)
|
2.2
|
2.2
|
1.1-4.5
|
 |
0.756 |
| 20-24 |
92
|
(23.7)
|
53
|
(15.7)
|
1.7
|
1.7
|
1.1-2.5
|
| 25-34 |
94
|
(15.0)
|
76
|
(11.0)
|
1.4
|
1.5
|
1.1-2.1
|
| >= 35 |
26
|
(16.5)
|
22
|
(11.1)
|
1.6
|
1.9
|
1.0-3.5
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
218
|
(18.6)
|
123
|
(12.4)
|
1.6
|
1.6
|
1.3-2.1
|
 |
0.921 |
| Army |
17
|
(21.3)
|
21
|
(14.2)
|
1.6
|
1.6
|
0.8-3.4
|
| Air Force |
15
|
(16.7)
|
20
|
(10.1)
|
1.8
|
1.9
|
0.9-4.0
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
32
|
(12.9)
|
27
|
(8.3)
|
1.6
|
1.8
|
1.1-3.2
|
|
0.368 |
| Other rank-supervisory |
104
|
(16.1)
|
83
|
(12.6)
|
1.3
|
1.4
|
1.0-1.9
|
|
Other rank-non supervisory
|
113
|
(25.1)
|
54
|
(15.3)
|
1.9
|
2.0
|
1.4-2.9
|
† Odds ratios are adjusted for service type, rank
and age (< 20, 20-24, 25 to 34, >= 35 years), education and marital
status
‡ P value for interaction
back to top
Table 11.4 Any post-Gulf War anxiety disorder: The effects of
study group across subgroups of age, service type and rank.
| |
GWV |
Comp grp |
|
|
|
|
|
| |
n |
(%) |
n |
(%) |
Crude
OR |
Adj
OR* |
95% CI |
|
P value‡ |
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
15
|
(9.3)
|
2
|
(1.8)
|
5.5
|
4.8
|
1.1-44.6
|
|
0.215 |
| 20-24 |
36
|
(9.9)
|
13
|
(4.0)
|
2.6
|
2.6
|
1.3-5.5
|
| 25-34 |
39
|
(6.6)
|
22
|
(3.3)
|
2.1
|
2.2
|
1.2-3.9
|
| >= 35 |
15
|
(9.8)
|
3
|
(1.6)
|
6.8
|
6.1
|
1.6-34.6
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
90
|
(8.2)
|
30
|
(3.1)
|
2.7
|
2.7
|
1.7-4.2
|
|
0.934 |
| Army |
12
|
(15.4)
|
7
|
(5.0)
|
3.5
|
3.9
|
1.3-12.6
|
| Air Force |
3
|
(3.4)
|
3
|
(1.5)
|
2.3
|
2.2
|
0.3-17.1
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
19
|
(7.9)
|
7
|
(2.1)
|
3.9
|
4.1
|
1.5-12.0
|
|
0.540 |
| Other rank-supervisory |
38
|
(6.3)
|
20
|
(3.2)
|
2.0
|
2.1
|
1.2-4.0
|
| Other rank-non supervisory |
47
|
(11.1)
|
13
|
(3.8)
|
3.1
|
3.1
|
1.6-6.5
|
* These odds ratios are adjusted for service type, rank
and age (<25 vs =25 years) only. CI values for these adjusted odds
ratios were GWV obtained using exact methods for stratified 2x2 tables
‡ P value for interaction
back to top
Table 11.5 Any post-Gulf War posttraumatic stress disorder: The
effects of study group across subgroups of age, service type and rank.
| |
GWV |
Comp grp |
|
|
|
|
|
| |
n |
(%) |
n |
(%) |
Crude
OR |
Adj
OR* |
95% CI |
|
P value‡ |
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
9
|
(5.3)
|
1
|
(0.9)
|
6.2
|
4.9
|
0.6-222
|
|
0.100 |
| 20-24 |
25
|
(6.4)
|
5
|
(1.5)
|
4.6
|
4.0
|
1.5-13.6
|
| 25-34 |
25
|
(3.9)
|
12
|
(1.7)
|
2.3
|
2.2
|
1.0-4.9
|
| >= 35 |
14
|
(8.5)
|
1
|
(0.5)
|
18.7
|
16.9
|
2.4-729
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
63
|
(5.3)
|
16
|
(1.6)
|
3.5
|
3.4
|
1.9-6.3
|
|
0.871 |
| Army |
5
|
(6.0)
|
2
|
(1.3)
|
4.7
|
4.3
|
0.7-46.9
|
| Air Force |
5
|
(5.4)
|
1
|
(0.5)
|
11.8
|
8.3
|
0.9-40.6
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
11
|
(4.4)
|
2
|
(0.6)
|
7.6
|
6.6
|
1.4-63.1
|
|
0.538 |
| Other rank-supervisory |
28
|
(4.3)
|
10
|
(1.5)
|
2.9
|
3.0
|
1.4-6.9
|
| Other rank-non supervisory |
33
|
(7.3)
|
7
|
(2.0)
|
3.9
|
3.9
|
1.7-10.6
|
* These odds ratios are adjusted for service type, rank
and age (<25 vs >=25 years) only. CI values for these adjusted odds
ratios were obtained using exact methods for stratified 2x2 tables
‡ P value for interaction
* These odds ratios are adjusted for service type, rank and age (<25
vs >=25 years) only. CI values for these adjusted odds ratios were
obtained using exact methods for stratified 2x2 tables.
‡ P value for interaction
back to top
Table 11.6 Any post-Gulf War substance-use disorder: The effects
of study group across subcategories of age, service type and rank.
| |
GWV |
Comp grp |
|
|
|
|
|
| |
n |
(%) |
n |
(%) |
Crude
OR |
Adj
OR† |
95% CI |
|
P value‡ |
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
67
|
(42.9)
|
32
|
(34.0)
|
1.5
|
1.5
|
0.9-2.6
|
|
0.884 |
| 20-24 |
77
|
(26.0)
|
43
|
(16.3)
|
1.8
|
1.7
|
1.1-2.7
|
| 25-34 |
60
|
(13.0)
|
47
|
(9.5)
|
1.4
|
1.4
|
0.9-2.1
|
| >= 35 |
10
|
(8.5)
|
7
|
(5.3)
|
1.7
|
1.8
|
0.7-5.1
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
202
|
(22.8)
|
110
|
(15.7)
|
1.6
|
1.6
|
1.2-2.1
|
|
0.382 |
| Army |
9
|
(14.1)
|
9
|
(8.6)
|
1.7
|
1.6
|
0.6-4.4
|
| Air Force |
3
|
(3.7)
|
10
|
(5.7)
|
0.6
|
0.6*
|
0.1-2.7
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
12
|
(5.5)
|
11
|
(4.0)
|
1.4
|
1.3
|
0.6-3.0
|
|
0.678 |
| Other rank-supervisory |
66
|
(14.4)
|
44
|
(10.4)
|
1.5
|
1.4
|
0.9-2.1
|
| Other rank-non supervisory |
135
|
(38.0)
|
74
|
(26.2)
|
1.7
|
1.7
|
1.2-2.5
|
† Odds ratios are adjusted for service type, rank
and age (< 20, 20-24, 25 to 34, >= 35 years), education and marital
status
‡ P value for interaction
* Where numbers are small, odds ratios are adjusted for service type,
rank and age (<25 vs >=25 years) only. CI values for these adjusted
odds ratios were obtained using exact methods for stratified 2x2 tables
back to top
11.4.2 12 item General Health Questionnaire (GHQ-12)
1422 (99.9%) Gulf War veterans and 1544 (99.7%) comparison group subjects
completed the GHQ-12 in the postal questionnaire. Table 11.7 presents
the prevalence of GHQ-12 cases (GHQ-12 symptom score >=2) for all Gulf
War veteran and comparison group subjects, and for all subjects within
subgroups of age, service type and rank. Gulf War veterans were more
likely than comparison group subjects to be suffering psychological morbidity
as measured by this instrument. The difference between study groups remained
statistically significant after adjustment for age, service type, rank,
education and marital status.
There was a strong age effect, with younger Gulf War veterans having
a greater prevalence of GHQ caseness than older Gulf War veterans, whilst
in the comparison group prevalence was greatest in the oldest subjects.
This indicates a differential age effect of Gulf War service upon increasing
psychological morbidity, with the association being greatest in the youngest
age group and least in the oldest subgroup.
Remaining tests for interaction were not statistically significant, indicating
that the Gulf War deployment was not more strongly associated with increased
psychological morbidity in one subgroup of service type or rank, when
compared with other subgroups.
In both study groups, psychological morbidity was greatest in the Army,
compared with the Navy and Air Force.
Table 11.7 GHQ-12 cases: The effects of study group across subgroups
of age, service type and rank.
| GHQ-12 cases (GHQ-12 >=2) |
| |
GWV
(N=1422) |
Comp grp
(N=1544) |
|
|
|
|
|
| |
n |
(%)† |
n |
(%)† |
Crude OR |
Adj OR§ |
95% CI |
|
P value |
|
All subjects
|
564
|
(39.6)
|
502
|
(32.5)
|
1.4
|
1.4
|
1.2-1.6
|
|
<0.001
|
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
74
|
(42.8)
|
29
|
(22.8)
|
2.5
|
2.6
|
1.5-4.3
|
|
0.040‡
|
| 20-24 |
164
|
(40.6)
|
123
|
(31.3)
|
1.5
|
1.5
|
1.1-2.0
|
| 25-34 |
267
|
(39.7)
|
266
|
(34.1)
|
1.3
|
1.3
|
1.0-1.6
|
| >= 35 |
59
|
(34.1)
|
84
|
(34.4)
|
1.0
|
1.1
|
0.7-1.6
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
488
|
(39.7)
|
355
|
(31.7)
|
1.4
|
1.5
|
1.2-1.7
|
|
0.484‡
|
| Army |
41
|
(47.1)
|
69
|
(40.4)
|
1.3
|
1.3
|
0.7-2.1
|
| Air Force |
35
|
(33.3)
|
78
|
(30.8)
|
1.1
|
1.1
|
0.7-1.8
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
94
|
(35.1)
|
117
|
(30.0)
|
1.3
|
1.3
|
0.9-1.8
|
|
0.065‡
|
| Other rank-supervisory |
267
|
(39.0)
|
260
|
(35.2)
|
1.2
|
1.2
|
1.0-1.5
|
| Other rank-non supervisory |
202
|
(43.2)
|
125
|
(30.0)
|
1.8
|
1.8
|
1.4-2.4
|
† Percentage of subjects within each subgroup of
age, service type or rank
§ Odds ratios are adjusted for service type, rank and age (< 20, 20-24,
25 to 34, >= 35 years), education and marital status
‡. P value for interaction
back to top
11.4.3 Posttraumatic Stress Disorder Checklist (PCL-S)
1339 (94.0%) Gulf War veterans and 1452 (93.7%) comparison group subjects
completed the PCL-S in the postal questionnaire.
Table 11.8 presents the prevalence of PCL-S cases (PCL-S score >=50)
for all Gulf War veteran and comparison group subjects, and for all subjects
within subgroups of age, service type and rank. Gulf War veterans were
more likely, than comparison group subjects, to be suffering symptoms
indicative of posttraumatic stress disorder, as measured by this instrument.
This difference between the two study groups was statistically significant
within most age and rank categories. The association between Gulf War
service and increased psychological morbidity, however, did not differ
across subgroups of age, service type or rank.
In both study groups, PCL-S caseness was most common in the Army, compared
with the Navy and Air Force. Posttraumatic stress disorder, as assessed
using this measure, was also most common in the oldest subjects in both
groups, and in the lowest ranks in the Gulf War veteran group.
Table 11.8 PCL-S cases: The effects of study group across subgroups
of age, service type and rank.
| PCL-S cases (PCL-S >=50) |
| |
GWV (N=1339) |
Comp grp (N=1452) |
|
|
|
|
|
| |
n |
(%)† |
n |
(%)† |
Crude OR |
Adj OR§ |
95% CI |
|
P value |
|
All subjects
|
105
|
(7.9)
|
66
|
(4.6)
|
1.8
|
2.0
|
1.5-2.9
|
|
<0.001
|
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
13
|
(7.9)
|
3
|
(2.6)
|
3.2
|
3.1*
|
0.8-17.7
|
|
0.584‡
|
| 20-24 |
31
|
(8.2)
|
14
|
(3.8)
|
2.2
|
2.2
|
1.1-4.3
|
| 25-34 |
40
|
(6.4)
|
28
|
(3.8)
|
1.7
|
1.9
|
1.2-3.2
|
| >= 35 |
21
|
(12.5)
|
21
|
(9.0)
|
1.5
|
1.7
|
0.9-3.4
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
86
|
(7.5)
|
45
|
(4.3)
|
1.8
|
2.1
|
1.4-3.1
|
|
0.930‡
|
| Army |
13
|
(15.1)
|
13
|
(7.9)
|
2.1
|
2.2
|
0.9-5.1
|
| Air Force |
6
|
(6.1)
|
8
|
(3.4)
|
1.9
|
1.7
|
0.6-5.2
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
9
|
(3.6)
|
11
|
(3.0)
|
1.2
|
1.5
|
0.6-3.7
|
|
0.350‡
|
| Other rank-supervisory |
49
|
(7.6)
|
37
|
(5.3)
|
1.5
|
1.8
|
1.1-2.9
|
| Other rank-non supervisory |
47
|
(10.6)
|
18
|
(4.7)
|
2.4
|
2.9
|
1.6-5.2
|
† Percentage of subjects within each subgroup of
age, service type or rank
§ Odds ratios are adjusted for service type, rank and age (< 20, 20-24,
25 to 34, >= 35 years), education and marital status
‡. P value for interaction
* Where numbers are small odds ratios are adjusted for service type, rank
and age (<25 vs >=25 years) only. Confidence intervals for these
adjusted odds ratios were obtained using exact methods for stratified
2x2 tables
back to top
11.4.4 Alcohol Use Disorders Identification Test (AUDIT)
1421 (99.8%) Gulf War veterans and 1546 (99.9%) comparison group subjects
completed the AUDIT in the postal questionnaire.
Table 11.9 presents the prevalence of AUDIT cases (AUDIT score >=8)
for all Gulf War veteran and comparison group subjects, and for all subjects
within subgroups of age, service type and rank. Gulf War veterans were
more likely, than comparison group subjects, to be problem drinkers as
measured by this instrument.
The risk of problem drinking, associated with the Gulf War deployment,
was greatest for the youngest Gulf War veterans, those in the Navy and
those in the lowest ranks. These differences in risk across subgroups
of age, service type and rank reached statistical significance.
Table 11.9 AUDIT cases: The effects of study group across subgroups
of age, service type and rank
| AUDIT cases (AUDIT >=8) |
| |
GWV (N=1421) |
Comp grp (N=1546) |
|
|
|
|
|
| |
n |
(%)† |
n |
(%)† |
Crude OR |
Adj OR§ |
95% CI |
|
P value |
|
All subjects
|
517
|
(36.4)
|
464
|
(30.0)
|
1.3
|
1.2
|
1.0-1.4
|
|
0.014
|
|
Age
|
|
|
|
|
|
|
|
|
|
| < 20 |
75
|
(43.4)
|
41
|
(32.5)
|
1.6
|
1.7
|
1.0-2.8
|
|
0.037‡
|
| 20-24 |
166
|
(41.1)
|
112
|
(28.6)
|
1.7
|
1.6
|
1.2-2.2
|
| 25-34 |
222
|
(33.1)
|
231
|
(29.5)
|
1.2
|
1.0
|
0.8-1.3
|
| >= 35 |
54
|
(31.0)
|
80
|
(32.7)
|
0.9
|
0.9
|
0.6-1.4
|
|
Service Type
|
|
|
|
|
|
|
|
|
|
| Navy |
483
|
(39.6)
|
374
|
(33.4)
|
1.3
|
1.3
|
1.1-1.6
|
|
0.025‡
|
| Army |
24
|
(27.6)
|
49
|
(28.5)
|
1.0
|
0.9
|
0.5-1.7
|
| Air Force |
10
|
(9.5)
|
41
|
(16.2)
|
0.5
|
0.5
|
0.2-1.0
|
|
Rank
|
|
|
|
|
|
|
|
|
|
| Officer |
53
|
(19.8)
|
78
|
(19.9)
|
1.0
|
0.8
|
0.5-1.2
|
|
<0.001‡
|
| Other rank-supervisory |
254
|
(37.1)
|
265
|
(35.8)
|
1.0
|
1.0
|
0.8-1.3
|
| Other rank-non supervisory |
209
|
(44.7)
|
121
|
(29.2)
|
2.0
|
2.0
|
1.5-2.7
|
† Percentage of subjects within each subgroup of
age, service type or rank
§ Odds ratios are adjusted for service type, rank and age (< 20, 20-24,
25 to 34, >= 35 years), education and marital status
‡. P value for interaction
The AUDIT total score is derived from subcategories of questions which
are representative of hazardous drinking (related to quantity and frequency
of drinking), alcohol dependence (drinking behaviour indicative of an
addiction) and harmful drinking (consequences of drinking which suggest
harm to self or others). When analysed on the basis of these subcategories
(data not shown), approximately 80% of the AUDIT score is being derived
from high levels of hazardous drinking in this population, with only low
levels of alcohol dependence and harmful drinking evident from responses
to this questionnaire.
back to top
11.4.5 All Gulf War veterans versus comparison group subjects
who had been on active deployments
Approximately one third of the comparison group (N=514) reported that
they had been on at least one active, war like deployment. Of these,
all 514 completed the AUDIT, 513 completed the GHQ-12, 488 completed the
PCL-S and 450 completed the psychologist administered CIDI. The prevalence
of CIDI defined psychological disorders present in the previous 12 months,
and GHQ-12, PCL-S and AUDIT caseness is shown in Table 11.10 for Gulf
War veterans and for those comparison group subjects who had been on an
active deployment.
Table 11.10 CIDI disorders present within 12 months and GHQ-12,
PCL-S and AUDIT caseness in Gulf War veterans and comparison group subjects
who had been on active deployments
| |
Gulf War veterans |
Comparison group |
|
|
|
| |
n |
(%) |
n |
(%) |
Adj OR‡ |
95% CI |
P value |
|
CIDI disorders
|
N=1381
|
N=450
|
|
|
|
| Any affective disorder |
144
|
(10.4)
|
28
|
(6.2)
|
1.5
|
1.0-2.4
|
0.065
|
| Any anxiety disorder |
177
|
(12.8)
|
37
|
(8.2)
|
1.9
|
1.1-3.1
|
0.015
|
| Posttraumatic stress disorder |
71
|
(5.1)
|
11
|
(2.4)
|
2.2
|
1.1-4.6
|
0.032
|
| Any substance use disorder |
67
|
(4.9)
|
12
|
(2.7)
|
1.6
|
0.9-3.1
|
0.140
|
| Any CIDI disorder |
284
|
(20.6)
|
65
|
(14.4)
|
1.4
|
1.0-2.0
|
0.030
|
| |
N=1422
|
N=513
|
|
|
|
|
GHQ-12 cases
|
564
|
(39.6)
|
185
|
(36.1)
|
1.1
|
0.9-1.4
|
0.370
|
| |
N=1339
|
N=488
|
|
|
|
|
PCL-S cases
|
105
|
(7.9)
|
22
|
(4.5)
|
1.9
|
1.1-3.1
|
0.015
|
| |
N=1421
|
N=514
|
|
|
|
|
AUDIT cases
|
517
|
(36.4)
|
152
|
(29.6)
|
1.1
|
0.9-1.4
|
0.481
|
‡ Odds ratios are adjusted for service type, rank
and age (< 20, 20-24, 25 to 34, >= 35 years), education, marital
status and pre-Gulf War disorders of the same type
When the statistical analysis is restricted to only those comparison
group subjects who have been on active deployments, the statistical power
of the analysis to detect small differences in psychological health outcomes
between the two study groups is reduced. Despite this reduction in statistical
power, Table 11.10 shows that when Gulf War veterans are compared with
only those comparison group subjects who have been on active deployments,
risk of CIDI disorders in the previous twelve months and risk of PCL-S
caseness remains heightened in the Gulf War veteran group. For most of
the measures of recent psychological morbidity shown in Table 11.10, the
adjusted odds ratios are only slightly reduced compared with those presented
in previous tables where Gulf War veterans were compared with all comparison
group subjects. The exception is the adjusted odds ratio for ‘any
substance disorder’ which actually remains the same. As expected,
however, the confidence intervals are slightly wider than those previously
presented. The elevated risk of psychological morbidity in Gulf War veterans
remains statistically significant for ‘any CIDI disorder’, ‘any
anxiety disorder’, posttraumatic stress disorder and PCL-S caseness,
but the risk is no longer statistically significantly elevated for ‘any
affective disorder’, ‘any substance disorder’, GHQ-12 caseness
and AUDIT caseness, when only those comparison group subjects who have
been on active deployments are included in the analysis.
back to top
11.4.6 Gulf War veteran group subanalysis: the effects of Gulf
War-related exposures on psychological disorders in Gulf War veterans
For Gulf War veterans, Table 11.11 through to Table 11.17 present the
effects of Gulf War service related Military Service Experience (MSE)
questionnaire scores, total number of immunisations and deployment era
(deployment completed before or after the commencement of the air war
on 17 January 1991) upon CIDI defined post-Gulf War psychological disorders
and GHQ-12, PCL-S and AUDIT caseness. These analyses are confined to
Gulf War veterans only.
Increasing number of psychological stressors, as indicated by increasing
score on the MSE questionnaire, was strongly associated with increasing
risk for all psychological disorder measures. Differences in odds ratios
across MSE score categories were statistically significant after adjustment
for service type, rank, age, education and marital status. The statistically
significant dose response slopes indicate that the expected increase in
the odds of any disorder, per unit increase in MSE score, varied from
7% for the AUDIT measure to more than 20% for the two measures of posttraumatic
stress disorder.
The effect of total number of immunisations upon risk of psychological
disorders was mixed. Compared with Gulf War veterans who reported no
immunisations, those who reported a small number of immunisations typically
had better psychological health, while those who reported large numbers
of immunisations typically had similar or poorer psychological health.
Gulf War veterans who did not know how many immunisations they received,
were also likely to have similar or poorer psychological health than those
who reported no immunisations. Amongst Gulf War veterans who reported
any immunisations, and for most measures of psychological health with
the exception of the AUDIT, statistically significant dose response slopes
indicated that the odds of psychological morbidity increased with every
unit increase in number of reported immunisations. The dose response
slope was steepest for the CIDI measures of any anxiety and posttraumatic
stress disorder and for the PCL.
PCL-caseness was statistically significantly more prevalent amongst Gulf
War veterans whose deployments continued after the air war commenced on
17 January 1991, than amongst Gulf War veterans who had completed their
deployments by this time. Post-Gulf War posttraumatic stress disorder
was also higher in the Gulf War veterans whose deployments continued after
the air war, however this difference did not reach statistical significance.
There were no other notable differences in psychological health outcomes
as a function of deployment era.
back to top
Table 11.11 Gulf War veterans with any post-Gulf War affective
disorder
|
Gulf War exposure
|
n |
(%) |
Crude OR |
Adj OR† |
95% CI |
|
P value§ |
|
MSE questionnaire score
|
|
|
|
|
|
|
|
| 0 – 4 |
(N=320) |
33
|
(11)
|
1.0
|
1.0
|
-
|
|
<0.001
|
| 5 – 8 |
(N=415) |
41
|
(10)
|
0.9
|
0.9
|
0.6-1.5
|
| 9 – 12 |
(N=316) |
55
|
(18)
|
1.9
|
1.9
|
1.2-3.0
|
| > 12 |
(N=369) |
120
|
(35)
|
4.5
|
4.5
|
2.9-7.1
|
| Dose response slope‡ |
-
|
-
|
1.13
|
1.14
|
1.11-1.17
|
|
<0.001
|
|
Immunisations
|
|
|
|
|
|
|
|
| None |
(N=115) |
24
|
(21)
|
1.0
|
1.0
|
-
|
|
-
|
| Any |
(N=907) |
157
|
(17)
|
0.8
|
0.7
|
0.4-1.1
|
|
0.124
|
| 1 – 4 |
(N=251) |
29
|
(12)
|
0.5
|
0.4
|
0.2-0.8
|
|
|
| 5 – 9 |
(N=533) |
102
|
(19)
|
0.9
|
0.8
|
0.5-1.3
|
|
|
| >= 10 |
(N=123) |
26
|
(21)
|
1.0
|
0.8
|
0.4-1.6
|
|
|
| Don’t know |
(N=318) |
67
|
(21)
|
1.0
|
0.8
|
0.5-1.4
|
|
0.409
|
| Dose response slope‡ |
-
|
-
|
1.06
|
1.05
|
1.00-1.11
|
|
0.059
|
|
Deployment completed before
the air war
|
|
|
|
|
|
|
|
| Yes |
(N=315) |
51
|
(16)
|
1.0
|
1.0
|
-
|
|
-
|
| No |
(N=1028) |
199
|
(19)
|
1.2
|
1.1
|
0.8-1.5
|
|
0.642
|
†. Odds ratios are adjusted for service type, rank
and age (< 20, 20-24, 25 to 34, >= 35 years), education and marital
status
‡. Dose response slope is the expected proportionate increase in
the odds ratio per unit increase in the MSE questionnaire score or per
unit increase in the number of immunisations amongst those who reported
‘any’ immunisations
§ With the exception of the P values for the dose response slopes, remaining
P values assess whether any odds ratios within each exposure variable
differ from unity
back to top
Table 11.12 Gulf War veterans with any post-Gulf War anxiety
disorder
|
Gulf War exposure
|
n |
(%) |
Crude OR |
Adj OR† |
95% CI |
|
P value§ |
|
MSE questionnaire score
|
|
|
|
|
|
|
|
| 0 – 4 |
(N=320) |
6 |
(2) |
1.0 |
1.0 |
- |
|
<0.001 |
| 5 – 8 |
(N=415) |
11 |
(3) |
1.4 |
1.3 |
0.5-3.6 |
| 9 – 12 |
(N=316) |
20 |
(7) |
3.6 |
3.6 |
1.4-9.2 |
| > 12 |
(N=369) |
67 |
(21) |
12.7 |
13.0 |
5.4-31.3 |
| Dose response slope‡ |
- |
- |
1.19 |
1.19 |
<
