An official website of the United States government
Here's how you know
Official websites use .gov
A
.gov website belongs to an official
government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you've safely
connected to the .gov website. Share sensitive
information only on official, secure websites.
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with,
the contents by NLM or the National Institutes of Health.
Learn more:
PMC Disclaimer
|
PMC Copyright Notice
Professor Lee and colleagues (October 2002
JRSM1)
presented their findings in 1000 Gulf War veterans who came to their programme
for medical advice, and combined the data with those from two previous
surveys, each of 1000 veterans, by Coker et
al.2 and their
own group3. I
question the methodology and statistical models adopted by Lee et al.
in their two
papers1,3.
First, why did they not adhere to the reviewing criteria adopted in the
first paper by Coker? Introduction of the term ‘wellness’ serves
to confuse. There is no differentiation between individuals who were well when
they attended and those who considered that they were unwell. Second, would it
not have been better to adhere to Chapter 18 diagnosis (International
Statistical Classification of Diseases) throughout the series, with
appropriate comment? To change the criteria between papers prevents comparison
of like with like. Third, whilst chronic fatigue syndrome (CFS) is included in
the initial paper—87 diagnosed with the syndrome—it has been
abandoned by Lee et al., despite the predominance of relevant
symptoms. ‘Multiple unexplained physical symptoms’ and somatic
disorders are included, which have no accepted criteria.
The major concern is the inclusion of 1057 individuals who were well by any
definition and attended for reassurance, information, etc. This fundamentally
distorts the statistics: overall, the percentages should be taken not from
3000 but from the 1943 individuals who attended because they felt ill. If the
Chapter 21 individuals are extracted from Table 2, this leaves 837 ill in the
first 1000, 494 in the second 1000 and 612 in the third 1000. Affective
symptoms are then, respectively, 59% (from 50%), 98% (from 49%) and 62% (from
38%); fatigue 50% (from 42%), 91% (from 45%) and 42% (from 26%). This is
reflected throughout. Table 3 then becomes uninterpretable. Table 5 should be
based on 1943 individuals and is effectively distorted by 50%—thus,
post-traumatic stress disorder (PTSD) becomes 19% (from 12%), depression 9%
(from 6%) and so on.
Even if we allow for the self-selection in the veterans seen by Lee et
al. there is a major discrepancy regarding PTSD between this cohort and
the series reported a few weeks later by Wessely's group in the
BMJ4. In
Lee's paper the prevalence is said to be 12% but should be 19%, as I have
shown; in Wessely's paper it is no more than 3%. Could the explanation be that
Lee et al. dropped CFS from the reviewing criteria used by Coker
et al.? Should many of their PTSD diagnoses have been CFS?
Because of these aberrations, the two papers by Lee et al.
contribute little to the understanding of illnesses affecting Gulf War
veterans.
References
1.Lee HA, Gabriel R, Bolton JPG, Bale AJ, Jackson M. Health status
and clinical diagnoses of 3000 Gulf War veterans. J R Soc
Med 2002;95:
491-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Coker WJ, Bhat BM, Blatchley NF, Graham JT. Clinical findings for
the first 1000 Gulf War veterans in the Ministry of Defence's medical
assessment programme. BMJ
1999;318:
290-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Lee HA, Gabriel R, Bale AJ, Bolton P, Blatchley NF. Clinical
findings of the second 1000 UK Gulf War veterans who attended the Ministry of
Defence's medical assessment programme. JRAMC
2001;147:
153-60 [DOI] [PubMed] [Google Scholar]
4.Ismail K, Kent K, Brugha T, et al. The mental health of UK
Gulf War veterans: phase 2 of a two phase cohort study.
BMJ
2002;325:
576-81 [DOI] [PMC free article] [PubMed] [Google Scholar]
As is made clear in our paper, the intent was to review the diagnoses of
all 3000 who attended and to make some assessment of wellbeing, as detailed by
Lee et al1.
Those attending the Gulf Veterans' Medical Assessment Programme (GVMAP) are
referred by their medical attendants on the basis that they have health
concerns relating to their service in the Gulf. It is therefore appropriate to
include all 3000 referred, as the population in which the prevalence of
diseases found is expressed rather than restricting the population denominator
to just those found to be ill. When one discusses the prevalence of disease in
a population, one usually includes the whole population as the denominator
rather than just the unwell portion. This addresses Wilson-Ing's concern that
we have fundamentally distorted the statistics.
Wilson-Ing further suggests that we have changed diagnostic criteria
because of the reduction of Chapter 18 diagnoses. The first point to make is
that the bridging 100 exercise in our first
paper1 was designed
to compare the diagnostic criteria between that paper and that of Coker et
al2. Secondly,
it is hardly surprising that the diagnostic pattern is different, given both
the time that has elapsed and the considerable research effort that has gone
into Gulf veterans' illnesses without finding any convincing evidence of
unique Gulf-related illness. To suggest that the diagnosis of chronic fatigue
syndrome (CFS) has been abandoned by us is disingenuous, since Wilson-Ing is
not in a position to make clinical diagnoses on our series. We, who have seen
the patients, are satisfied that the symptoms presented do not amount to
chronic fatigue
syndrome3,4.
We have certainly not included medically unexplained physical symptoms or
somatic disorders in our diagnoses, but have discussed these aspects in the
discussion section of the paper. We would suggest that the references therein
provide the background and criteria for our discussion.
In his comments about PTSD, Wilson-Ing is falling into the trap of not
comparing like with like. Since PTSD is associated with exposure to trauma, it
is likely that we would see more of it, since we were reporting upon the
diagnoses of a group of military veterans who had health concerns as a result
of their service in a conflict, than Ismail would find in a study examining
the mental health of a sample of the whole Gulf veteran population which
includes those with no health
concerns5. The
analogy is to claim that there is a higher prevalence of lung cancer among
patients attending a chest clinic than amongst the general population. The
suggestion that the alleged excess of PTSD may be misdiagnosed CFS rather
misses the point, explicitly stated in our paper, that the suggested diagnoses
of PTSD were confirmed by consultant psychiatrists.
References
1.Lee HA, Gabriel R, Bale AJ, Bolton P, Blatchley NF. Clinical
findings of the second 1000 UK Gulf War veterans who attended the Ministry of
Defence's Medical Assessment Programme. JRAMC
2001;147:
153-60 [DOI] [PubMed] [Google Scholar]
2.Coker WJ, Bhatt BM, Blatchley NF, Graham JT. Clinical findings for
the first 1000 Gulf War veterans in the Ministry of Defence's medical
assessment programme. BMJ
1999;318:
290-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Chronic Fatigue Syndrome. Report of a Joint Working
Group of the Royal Colleges of Physicians, Psychiatrists and General
Practitioners. London: RCP, 1996 [PMC free article] [PubMed]
5.Ismail K, Kent K, Brugha T, et al. The mental health of UK
Gulf War veterans: phase 2 of a two phase cohort study.
BMJ
2002;325:
576-81 [DOI] [PMC free article] [PubMed] [Google Scholar]
