8

Results for Chronic Multisymptom Illness

This chapter describes chronic multisymptom illness (CMI), sometimes also referred to as Gulf War illness (GWI), among veterans of the first Gulf War and presents the results of the case-control studies and structured literature search.¹ The chapter also includes conclusions on possible relationships between CMI and the deployment-related environmental and occupational exposures examined. Criteria to prioritize results of the statistical analyses presented in this chapter are an adjusted odds ratio (OR) of at least 1.10 and the exclusion of 1.0 in the 95% confidence interval (CI). In addition, the chapter presents results of stratified analyses and a cumulative exposure analysis. Although traumatic brain injury (TBI) was specified as an outcome in the Statement of Task, the committee instead treated it as a covariate and stratified each exposure–outcome pair by TBI status, as TBI is not caused by these military exposures. For these results, the committee prioritized reporting those in which the adjusted ORs between strata were different, at least one stratum’s OR was above 1.0 and the lower bound of the 95% CI was also above 1.0, and the 95% CIs did not overlap between strata. Furthermore, the chapter presents results of cumulative exposure analyses based on additive counts of individual binary exposures, given evidence of a trend of increased risk with increasing exposure. Full results of all data analyses are in Appendix G.

___________________

¹ In the literature review, this report follows the language (CMI or GWI) used in the original study.

Due to time constraints, the committee began its structured literature search before obtaining data on exposures from the Individual Longitudinal Exposure Record (ILER); therefore, it is not directly based on each of the exposure categories the committee had available for its original data analysis (see Chapter 4 for more information). For example, ILER data contained exposures for dust and particulate matter (PM), exhaust, and incinerator emissions, while the literature search was conducted for PM because it is a component of these exposures and has a robust literature base. Chapter 4 describes how the committee integrated evidence from its data analyses with the literature search results to draw its conclusions. However, the committee cautions again that the results presented are based on a sample of post-9/11 veterans who served in Southwest Asia or Afghanistan and received care at Veterans Health Administration (VHA), so the samples used for its analyses have limited generalizability. All conclusions should be interpreted as specific to this population. Certain military exposures are specifically of concern in relation to CMI. For example, the Research Advisory Committee on Gulf War Veterans’ Illnesses 2008 report Gulf War Illness and the Health of Gulf War Veterans concluded,

Evidence strongly and consistently indicates that two Gulf War neurotoxic exposures are causally associated with Gulf War illness: 1) use of pyridostigmine bromide (PB) pills, given to protect troops from effects of nerve agents, and 2) pesticide use during deployment. Evidence includes the consistent association of Gulf War illness with PB and pesticides across studies of Gulf War veterans, identified dose-response effects, and research findings in other populations and in animal models. (RAC-GWVI, 2008, p. 1)

The report further did not rule out an association between GWI and exposures including nerve agents, close range to oil well fires, receipt of several vaccines, and effects of aggregated exposures (RAC-GWVI, 2008). However, it was not feasible to investigate the relationship between these exposures and CMI because data on these exposures were not available in ILER.

CHRONIC MULTISYMPTOM ILLNESS

At the conclusion of the first Gulf War (1990–1991), service members who deployed to Southwest Asia reported a complex and unexplained constellation of symptoms that has since been defined as GWI. GWI is sometimes referred to as “CMI,” including by the Department of Veterans Affairs (VA) and Department of Defense (DoD) and among service members and veterans of more recent conflicts who did not serve during the 1990–1991

Gulf War (VA, 2014). Several case definitions have been developed and used over the years. Each requires multiple chronic symptoms across the domains of fatigue, pain, sleep, and musculoskeletal, gastrointestinal, and respiratory symptoms (IOM, 2014). These symptoms have also been reported among more recent cohorts of veterans who deployed to Southwest Asia.

Due to the lack of standardized definition for GWI/CMI, determining prevalence can be difficult. Two predominant case definitions exist, and both are based entirely on assessment of self-reported symptoms. The Centers for Disease Control and Prevention (CDC) definition requires one or more symptoms lasting longer than 6 months across at least two of three domains: pain, fatigue, or mood and cognition (IOM, 2014). The Kansas criteria require symptoms in at least three of six domains: fatigue and sleep; pain; neurologic, cognitive, or mood; gastrointestinal; respiratory; and skin. The Kansas criteria also account for severity of symptoms: at least one must be moderate to severe (IOM, 2014). One issue with the Kansas criteria is that they exclude some psychiatric conditions that could interfere with the ability to report on some of the symptom domains. A study of veterans who met the Kansas criteria found that those who were so excluded scored higher on the VA Frailty Index (measured in a patient’s electronic health record [EHR] by VHA and used as a proxy for current health status) than those without an exclusionary condition (Chao, 2023).

No definitive etiology of GWI or CMI has been proposed; however, studies have linked exposure to pesticides; chemicals, such as PB, sarin, and cyclosarin; or burning oil wells with GWI in veterans deployed to the Gulf War (Steele et al., 2012; White et al., 2016).

One survey of veterans deployed to the Gulf War found 43.9% reported symptoms consistent with CMI (Dursa et al., 2016). A literature review of multiple studies found prevalence among Gulf War veterans of 25–32% (White et al., 2016). One survey of Gulf War veterans found 39.9% and 84.2% met the Kansas or CDC criteria, respectively (Gifford et al., 2021). A study using data from VA’s Cooperative Studies Program 2006/Million Veteran Program 029 Cohort, a repository of EHR system data linked with biobank and genetic data, found the prevalence of GWI using the Kansas definition, including exclusionary conditions, to be 21.5%; the Kansas definition, without accounting for exclusionary conditions, to be 67.1%; and the CDC definition to be 81.1% (Duong et al., 2022). DoD reports that of the over 700,000 troops deployed to the first Gulf War, 175,000–250,000 (roughly 25% to 36%) had GWI (DoD, 2018).

Among veterans deployed to post-9/11 conflicts in Southwest Asia or Afghanistan, estimates of CMI prevalence assessed using International Classification of Diseases, Ninth Revision (ICD-9) codes of related diagnoses vary greatly. One study that measured CMI as one or more ICD-9 diagnosis of chronic fatigue syndrome, fibromyalgia, or irritable bowel

syndrome determined a prevalence of 8.2% among female Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn (OEF/OIF/OND)² veterans (Mohanty et al., 2015). Another study of OEF/OIF/OND veterans, which used the same CMI measure, found a prevalence of 8.4% in female and 4.25% in male veterans (Mohanty et al., 2018). Using the CDC definition of CMI, a longitudinal cohort study determined that 1 year after deployment, 49.5% of OEF/OIF veterans met criteria for mild/moderate CMI and 10.8% met the criteria for severe CMI (McAndrew et al., 2016).

Analysis Results

There were 105,125 members (9.2%) of the cohort who met the criteria for CMI. Figure 8-1 shows that exposure to dust and PM is associated with a risk-conferring relationship with CMI with an adjusted OR of 1.10 (95% CI: 1.08–1.13). Estimated ORs from exposure to exhaust, incinerator emissions, and solvents are also elevated in the direction of a risk-conferring relationship but did not meet the committee’s prioritization criteria. A sensitivity analysis showed no difference between requiring three diagnoses as a criterion for CMI (as the committee did), only one (a more relaxed cutoff), and four (a more stringent cutoff), with no meaningful difference in estimated ORs for the exposures. Additionally, due to concerns that long COVID symptoms overlap with many diagnoses comprising CMI, the committee conducted sensitivity analyses that only considered CMI cases before 2020; these produced similar OR estimates to those of the primary analysis (that did not restrict to pre-2020 cases). Evidence among people with TBI showed that the association between exposure to burn pits and CMI is 27.4% higher than among people without TBI. TBI strata also had evidence of higher odds for exposure to radiation and CMI, though it did not meet the prioritization criteria. Finally, increasing the number of exposures (compared to zero) is generally associated with elevated risk of CMI.

Literature Search Results

Burn Pits and CMI

While the literature search yielded zero results on a possible relationship between exposure to burn pits and the risk of CMI, it did produce one study that examined exposure to burning trash or feces and CMI (assessed using the CDC criteria) among post-9/11 Iraq and Afghanistan veterans.

___________________

² OEF was the military operation in Afghanistan from October 7, 2001, to December 28, 2014. OIF was the military operation in Iraq from March 19, 2003, to August 31, 2010. OND was the military operation in Iraq from September 1, 2010, to December 15, 2011.

After accounting for demographics (age, gender, education), combat exposure, posttraumatic stress symptoms, and smoking status, exposure to a cluster of smoke inhalation exposures (smoke/air pollution, fumes/exhaust from heaters or generators, and burning trash or feces) was not a significant predictor of CMI (DeBeer et al., 2017).

PM and CMI

The search yielded zero results on the possible relationship between exposure to PM and the risk of developing CMI. However, as described in the previous section, DeBeer and colleagues (2017) examined the relationship between smoke/air pollution and fumes/exhaust from heaters or generators, both of which are composed of PM, and CMI. Also, as described in the previous section, smoke inhalation including these exposures was not a significant predictor of CMI.

Metals and CMI

The search yielded one study on depleted uranium exposure and CMI. This nested case-control study of Gulf War veterans found no difference in the likelihood of developing GWI across different urinary uranium levels. This suggests that depleted uranium exposure is not significantly associated with developing GWI (Parrish and Haley, 2021).

Mold and CMI

The search yielded one cross-sectional study that investigated the presence of fungal antigens in blood sera among Gulf War veterans at a Pennsylvania veterans’ hospital. It found that patients with GWI had high IgG titers against Chaetomium, a genus of mold, compared to two groups of controls—healthy civilian volunteers who had not been in the Persian Gulf and healthy civilian volunteers who had been in the Middle East but did not have GWI (Cherwonogrodzky et al., 2024). A key limitation of this study is that using serum measurements cannot definitively establish whether a patient had a Chaetomium infection and not an immune response to a different exposure.

Fuels, Radiation, or Solvents and CMI

The committee’s search yielded zero results on the possible relationships between exposure to fuels, radiation, or solvents and the risk of CMI.

Conclusion

Conclusion 8-1: Based on its analysis of the available data, the committee finds there is a possible risk-conferring relationship between exposure to dust and particulate matter (PM) and chronic multisymptom illness (CMI). The committee also finds there is some evidence of a potential risk-conferring relationship between exposure to exhaust, incinerator emissions, or solvents and CMI. The committee does not find a possible risk-conferring relationship between exposure to burn pits, fuels, metals, mold, or radiation and CMI.

Based on the literature review, there is insufficient evidence of a possible risk-conferring relationship between exposure to burn pits, PM, metals, or mold and CMI. There is no identified literature on the relationship between exposure to fuels, radiation, or solvents and CMI.

Synthesizing the committee’s data analysis and literature review, the committee concludes there is a possible risk-conferring relationship between exposure to dust and PM and CMI. The committee further concludes there is inadequate or insufficient evidence of a possible risk-conferring relationship between exposure to burn pits, exhaust, fuels, incinerator emissions, metals, mold, radiation, or solvents and CMI.

SUMMARY

Through its data analysis and literature review, the committee examined possible relationships between CMI and its exposures of interest: burn pits, dust and PM, exhaust, fuels, incinerator emissions, metals, mold, radiation, and solvents. Synthesizing the results of these two streams of evidence, one possible risk-conferring relationship with CMI exists: for dust and PM. Evidence was inadequate or insufficient of a possible risk-conferring relationship between burn pits, exhaust, fuels, incinerator emissions, metals, mold, radiation, or solvents and CMI. Based on the sample population, conclusions are specific to post-9/11 veterans who served in Southwest Asia and Afghanistan and received health care at VHA. Furthermore, owing to lack of available data, it was not possible to investigate certain military exposures specifically of concern in relation to CMI.

REFERENCES

RAC-GWVI (Research Advisory Committee on Gulf War Veterans’ Illnesses). 2008. Gulf War illness and the health of Gulf War veterans. Washington, DC: Research Advisory Committee on Gulf War Veterans’ Illnesses, Department of Veterans Affairs.

Steele, L., A. Sastre, M. M. Gerkovich, and M. R. Cook. 2012. Complex factors in the etiology of Gulf War illness: Wartime exposures and risk factors in veteran subgroups. Environmental Health Perspectives 120(1):112–118.

VA (Department of Veterans Affairs). 2014. VA/DoD Clinical Practice Guideline for the Management of Chronic Multisymptom Illness. https://www.healthquality.va.gov/guidelines/MR/cmi/VADoDCMICPG.pdf (accessed September 2, 2024).

White, R. F., L. Steele, J. P. O’Callaghan, K. Sullivan, J. H. Binns, B. A. Golomb, F. E. Bloom, J. A. Bunker, F. Crawford, J. C. Graves, A. Hardie, N. Klimas, M. Knox, W. J. Meggs, J. Melling, M. A. Philbert, and R. Grashow. 2016. Recent research on Gulf War Illness and other health problems in veterans of the 1991 Gulf War: Effects of toxicant exposures during deployment. Cortex 74:449–475.

This page intentionally left blank.