|
Unwin et al., 1999
UK
|
2735 UK veterans deployed to Gulf War vs 2393 deployed to Bosnia vs 2422 deployed elsewhere
|
Four of 29 related to pesticides or solvents: “personal pesticides,” “other paints or solvents,” “handled prisoners of war,” “pesticides on clothing or bedding”
|
Symptom questionnaires, exposure questionnaire
|
All solvent or pesticide exposures associated with chronic multisymptom syndrome, posttraumatic stress reaction; physical functioning in all three cohorts “Concentration” factor not associated with any relevant pesticide or solvent exposure in multivariate regression analysis; “neurologic” factor associated with pesticide handling in multivariate regression analysis (p< 0.001), but no clear dose-response relation; “neurologic” factor not associated with other pesticide, solvent exposures in multivariate analysis
|
Self-reported symptoms and exposures; lack of adjustment for interrelationships between multiple exposures; use of p value of 0.05 despite multiple comparisons
|
|
Cherry et al, 2001a
UK
(see also Table 7.1)
|
4795 UK veterans deployed to Gulf War (and validation cohort of 4750) vs 4793 UK veterans not deployed to Gulf War
|
Four of 14 related to pesticides or solvents: “days exposed to handling pesticides,” “days living in quarters sprayed with insecticides,” “days respraying vehicles,” “days applying insecticide to skin”
|
Symptom questionnaire, exposure questionnaire; surveys completed 7 years or more after war
|
Self-reported symptoms and exposures
|
|
Iowa Persian Gulf Study Group, 1997
US
|
1896 deployed veterans from Iowa as home of record vs 1799 nondeployed veterans from Iowa as home of record
|
Two of over 20 items related to solvents or pesticides: “solvents/petrochemicals,” “pesticides”
|
Symptom questionnaires, exposure questionnaire
|
Symptoms of cognitive dysfunction associated with higher prevalence of exposure to “solvents/petrochemicals” (prevalence difference of 6.6%, p<0.001), “pesticides” (prevalence difference of 14.2%, p<0.001)
|
Self-reported symptoms and exposures; low proportion of minority-group subjects; internal validation of responses not assessed; no adjustment for multiple comparisons; multiple associations between variety of exposures and variety of outcomes
|
