4

FACTOR ANALYSIS AND ITS USE IN STUDIES OF SYMPTOMS IN GULF WAR VETERANS

Because of the large number of symptoms reported by 1990–1991 Gulf War veterans that are potentially associated with chronic multisymptom illness (CMI), and the absence of definitive diagnostic tests for the condition, statistical analysis of reported symptoms has been prominently used to evaluate chronic multisymptom illness (CMI)–defining symptoms. The two most frequently used statistical methods have been factor analysis and cluster analysis. Researchers began using statistical analyses to evaluate whether symptoms found in Gulf War veterans might constitute a unique syndrome. This chapter provides a brief discussion of the types of analyses and summaries of the studies that identified symptom factors or symptom clusters. In some cases, the researchers used their findings to inform the development of a case definition, but factor analysis alone cannot create a case definition. Although the focus of this chapter is on the studies that used factor and cluster analyses, there is overlap with the symptom studies reviewed in Chapter 3 and the case-definition studies discussed in Chapter 5. For a more detailed description of the statistical analyses used in the studies discussed below, see Appendix A.

FACTOR ANALYSIS

Factor analysis is a statistical method for conducting structural analyses of datasets. Large numbers of quantitative observations or responses can be resolved into “distinct patterns of occurrence” (Forbes et al., 2004). The patterns that are derived in a factor-analytic model are referred to as factors (Kline, 2000). Each factor explains a portion of the variance in such a way that the first factor explains the greatest percentage of the variance and each successive factor accounts for decreasing percentages of the variance. “Factor scores” estimate people’s relative levels (number or severity) of symptoms associated with each factor. A factor score combines a person’s responses to items associated with the factor and corresponding weights that represent the strength of associations between individual items and the factor. A factor score is generated for each person and each factor; thus, if a four-factor solution is posited, each person will have four factor scores. In some factor-analytic methods, the estimated factors are allowed to correlate with each other. When that approach is used, the intercorrelations are referred to as factor correlations. The relative strength of the relationship between any individual item and a factor is expressed as its factor loading. Factor loadings are the weights used in calculating people’s factor scores. Factor analysis can be exploratory or confirmatory. In a confirmatory factor



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 67
4 FACTOR ANALYSIS AND ITS USE IN STUDIES OF SYMPTOMS IN GULF WAR VETERANS Because of the large number of symptoms reported by 1990–1991 Gulf War veterans that are potentially associated with chronic multisymptom illness (CMI), and the absence of definitive diagnostic tests for the condition, statistical analysis of reported symptoms has been used to evaluate CMI-defining symptoms. The two most frequently used statistical methods have been factor analysis and cluster analysis. Researchers began using statistical analyses to evaluate whether symptoms found in Gulf War veterans might constitute a unique syndrome. This chapter provides a brief discussion of the types of analyses and summaries of the studies that identified symptom factors or symptom clusters. In some cases, the researchers used their findings to inform the development of a case definition, but factor analysis alone cannot create a case definition. Although the focus of this chapter is on the studies that used factor and cluster analyses, there is overlap with the symptom studies reviewed in Chapter 3 and the case- definition studies discussed in Chapter 5. For a more detailed description of the statistical analyses used in the studies discussed below, see Appendix A. FACTOR ANALYSIS Factor analysis is a statistical method for conducting structural analyses of datasets. Large numbers of quantitative observations or responses can be resolved into “distinct patterns of occurrence” (Forbes et al., 2004). The patterns that are derived in a factor-analytic model are referred to as factors (Kline, 2000). Each factor explains a portion of the variance in such a way that the first factor explains the greatest percentage of the variance and each successive factor accounts for decreasing percentages of the variance. “Factor scores” estimate people’s relative levels (number or severity) of symptoms associated with each factor. A factor score combines a person’s responses to items associated with the factor and corresponding weights that represent the strength of associations between individual items and the factor. A factor score is generated for each person and each factor; thus, if a four-factor solution is posited, each person will have four factor scores. In some factor-analytic methods, the estimated factors are allowed to correlate with each other. When that approach is used, the intercorrelations are referred to as factor correlations. The relative strength of the relationship between any individual item and a factor is expressed as its factor loading. Factor loadings are the weights used in calculating people’s factor scores. Factor analysis can be exploratory or confirmatory. In a confirmatory factor analysis factor loadings are posited a priori, and the resulting hypothesis is submitted to statistical testing (Kline, 2000). Confirmatory factor analysis is appropriate only when there is 67

OCR for page 67
68 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS some body of knowledge or theory regarding the factor structure. In addition, in confirmatory factor analysis, items typically load on only a single factor. Some of the studies use another exploratory method known as principal component analysis (see Appendix A). The data for factor-analytic studies can be people’s responses to a set of items or a list of symptoms with respect to their presence or absence and their severity. The items can be dichotomous (such as yes–no questions) or have more than two possible responses (such as never–sometimes–always questions). In the case of CMI, responses to both kinds of items have been used in factor analysis. Investigators use the results of a factor analysis to posit a plausible factor structure; that is, they estimate patterns regarding, for example, how different symptoms are related to each other and to the derived factors. It is inappropriate (and misleading) to refer to factors as “emerging” from a factor analysis (Pett et al., 2003). Such terminology inaccurately implies that a “true” set of factors underlie the data and that the factor set needs only to be unearthed. Factor- analytic results are seldom unequivocal, and they are influenced by a series of analytic decisions made by the researcher. An editor’s note accompanying the Kang et al. (2002) study (discussed later in this chapter) on factor analysis states: “Factor analysis is not completely objective; for example, there are no definite rules for selecting the appropriate number of factors . . . or rules for selecting from among the many possible methods of rotation. It is an empirical method.” Factor Analysis for Data Reduction Factor analysis has been used in studies of Gulf War veterans, initially to see whether a unique “Gulf War syndrome” could be identified and later to inform case definitions of CMI. In attempting to reduce the amount of data that would be gathered (the large and varied number of symptoms), researchers used factor analysis so that a structure that would include substantially fewer factors than symptoms could be proposed. Typically in CMI studies, a survey that includes many individual items or symptoms is administered. For example, a survey administered by Knoke et al. (2000) included 98 individual items or symptoms, but the factor-analytic results suggested a data structure of five factors. In that example, one factor included 27 symptoms. Factor scores estimate people’s relative levels of given factors. By reducing large sets of symptom data into their structural components, factor analysis can simplify comparisons of symptoms that are potentially related to Gulf War deployment. The application of the data-reduction capabilities of factor analysis has been attractive for the study of Gulf War veterans’ symptoms, but, as will be discussed below, studies’ findings have been inconsistent. Different studies have identified different numbers of factors and assigned different names to common groups of symptoms. Some lack of consistency in findings of factor-analytic studies is expected because of differences in methods and questionnaires and because of random variation. Furthermore, when factor analyses are conducted without an a priori hypothesis, the subjective labeling of factors can be controversial, so factor labels should be critically reviewed (Ismail and Lewis, 2006). The factors posited in a given study will depend on the statistical approach used in the factor analysis, including decisions about how factors are “extracted” and “rotated.” In addition, researchers must subjectively assign a name that they believe represents the items of a factor. To compare studies, it is necessary to understand which

OCR for page 67
FACTOR ANALYSIS AND ITS USE 69 symptoms were associated with each factor, rather than depending on the label that researchers assigned to each factor. Factor Analysis for Case Definitions Although factor-analytic studies have facilitated and clarified comparisons of symptom prevalence and severity in deployed and nondeployed people, they have been less useful in specifying a case definition of CMI. The results of factor analyses do not differentiate among groups of people and cannot create a case definition. That fact has been obscured because investigators often operationalize a case definition by dichotomizing factor scores obtained from a factor-analytic model. However, people do not have factors. As explained above, everyone will have a score on every factor, but dichotomization of factor scores to define a “case” is a postprocessing decision made by investigators and not a direct result of the factor-analytic model. Factor analysis can be used to evaluate whether the structure of symptom data is different in different populations (such as deployed and nondeployed), but this is not the same question as whether populations have higher symptom levels or greater symptom severity. The unique question that can be asked in the context of a factor-analytic study is whether the factor structure varies among compared populations. That question is most appropriately posited as a formal statistical test, in which the probability of observing the differences between the factor structures in the samples is estimated under the null hypothesis that the factor structures are the same in the two populations. For example, Ismail and colleagues (1999) compared three UK military cohorts: veterans of the Bosnia conflict, those deployed to the Gulf War, and Gulf War veterans not deployed to the Gulf. In addition to applying exploratory factor analysis, the investigators used a particular application of confirmatory factor analyses for which they generated statistics to estimate the goodness of fit of the factor-analytic model. They also tested a series of three models with different constraints: (1) factor correlations are equal in Gulf War–deployed and era veterans, (2) correlations between factors and factor loadings are equal in the two groups, and (3) all parameters are equal. Those models provided a direct, thorough, and hypothesis-based test of whether the factor structure differed in the two groups. That the constrained models did not fit significantly better than the unconstrained model indicates congruence in factor structure in the Gulf War–deployed and Gulf War–era veterans. However, most of the studies of factor-structure differences have failed to test the hypothesis directly, and none has used structural equation models. Instead, investigators have commonly relied on hypotheses related to factor scores or on descriptive comparisons of factor scores, factor loadings, and factor correlations. There are two ways of comparing factor scores. In one, factor scores are generated for all members of reference and comparison groups. The scores are derived on the basis of a single factor-analytic model. The scores of deployed and nondeployed persons are compared to ascertain whether the presence or severity of the symptoms that define a factor differ by deployment status. No comparison of factor structure is made, because a single modeled structure is used to generate all scores. The other approach is to conduct separate factor analyses in the reference and comparison groups, derive factor scores for everyone on the basis of both factor-analytic models, and then compare the resulting scores; this is not an accepted method of comparing factor structures. The studies that have used factor analysis to investigate symptoms in Gulf War veterans have used several analytic strategies. Some studies used statistical testing of the hypothesis that

OCR for page 67
70 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS the factor structures of deployed and nondeployed veteran populations are significantly different (that is, testing the null hypothesis that the structures are the same) (Ismail et al., 1999), as discussed below and in Chapter 3. Other studies relied exclusively on descriptive statistical techniques, such as correlations among factors, factor scores, or factor loadings (Doebbeling et al., 2000; Kang et al., 2002). Finally, a number of studies used “visual inspection” to discern differences between the factor structures of deployed and nondeployed groups (Knoke et al., 2000; Nisenbaum et al., 2004; Shapiro et al., 2002). FACTOR-ANALYSIS STUDIES1 This section discusses studies that used factors to determine whether veterans’ symptoms might constitute a new syndrome or are a variant of a known syndrome. The cohorts are described fully in Chapter 3, and the descriptions below are limited to the methods and results associated with the factor analyses. The studies are presented by cohort and in the same order as in Chapter 3 and are also listed in Table 4.1. Many researchers used the data collected from their studies to inform or develop case definitions, which are discussed in greater detail in Chapter 5. Department of Veterans Affairs The nationally representative Department of Veterans Affairs (VA) study searched for potential new syndromes through factor analysis (Kang et al., 2002). Data were from a sample of 15,000 deployed and 15,000 nondeployed active-duty, reserve, National Guard, and retired service members in all four branches. Through questionnaures, the authors inquired about 47 symptoms on a three-point ordinal scale. On the basis of judgments of factor interpretability, they chose a five-factor solution for the nondeployed and a six-factor solution for the deployed sample. By inspection, the investigators judged that the first five factors were “very similar” in the two groups. The six factors were fatigue and depression,2 musculoskeletal and rheumatologic,3 gastrointestinal,4 pulmonary,5 upper respiratory,6 and neurologic.7 However, the last factor extracted contained symptoms consistent with neurologic impairment in the Gulf War group but not the non–Gulf War group. It should be noted that each successive factor that is extracted in a factor analysis accounts for less of the variance than the previous one. In the deployed sample, the sixth factor, labeled neurologic impairment, accounted for only 3% of the total variance, compared with 79% for the first factor. The neurologic factor was not extracted in the nondeployed group but accounted for 4% of the variance. In the neurologic factor, four symptoms—loss of balance or dizziness, speech difficulty, blurred vision, and tremors or shaking—loaded for the deployed but not for the nondeployed group. The authors indicated that 1 The descriptions of the factor-analytic studies have been summarized from previously published Institute of Medicine reports (IOM, 2006, 2010). 2 Awakening tired and worn out; concentration and memory problems; excessive fatigue; fatigue more than 24 hours after exertion; feeling anxious, irritable, or upset; feeling depressed or blue; sleep difficulty; and sleepiness during daytime. 3 Back pain or spasms, generalized muscle aches, joint aches, numbness in hands or feet, swelling in joints, and swelling in extremities. 4 Constipation, diarrhea, nausea; reflux, heartburn, or indigestion; stomach or abdominal pain; and vomiting. 5 Coughing, irregular heartbeat, shortness of breath, tightness in chest, and wheezing. 6 Coughing, runny nose, sore throat, swollen glands, and trouble swallowing. 7 Blurred vision, concentration or memory problems, irregular heartbeat, loss of balance or dizziness, speech difficulty, sudden loss of strength, tremors or shaking, excessive fatigue, and fatigue more than 24 hours after exertion.

OCR for page 67
FACTOR ANALYSIS AND ITS USE 71 a group of 277 deployed veterans (2.4%) and a group of 43 nondeployed veterans (0.45%) had all four of those symptoms. The authors interpreted their findings as suggesting a possible neurologic syndrome related to Gulf War deployment that would require objective supporting clinical evidence. It is possible, however, that this is an overinterpretation of the data inasmuch as the factor accounts for a small amount of the total variance and the nonextracted sixth factor in the nondeployed group accounted for more variance than it did in the deployed group. The Iowa Study The Iowa study (Iowa Persian Gulf Study Group, 1997) grouped symptoms into categories suggestive of existing syndromes or disorders, such as fibromyalgia or depression. Its finding of a considerably higher prevalence of symptom groups suggestive of fibromyalgia, depression, and cognitive dysfunction in Gulf War veterans motivated the first applications of factor analysis to grouping and classifying veterans’ symptoms. Several years later, the same team of Iowa investigators performed a factor analysis on the Iowa cohort (Doebbeling et al., 2000). They studied the frequency and severity of 137 self-reported symptoms in 1,896 Gulf War veterans and 1,799 era veterans. Doebbeling et al. (2000) applied factor analysis in a sample of veterans who had been deployed and in a sample of nondeployed era controls. The deployed sample was divided into a training sample and a validation sample to evaluate the reproducibility of the factor solution for the group. Comparisons were made by correlating both factor loadings and factor scores in the deployed vs nondeployed samples. The authors identified three symptom factors in deployed veterans in the derivative sample that accounted for 35% of the variance: somatic distress (joint stiffness, myalgia, polyarthralgia, numbness or tingling, headache, and nausea), psychologic distress (feeling nervous, worrying, feeling distant or cut off; depression; and anxiety), and panic (anxiety attacks; a racing, skipping, or pounding heart; attacks of chest pain or pressure; and attacks of sweating). The researchers also conducted factor analysis in the nondeployed group and found the same three factors, which accounted for 29% of the variance. The authors concluded that their analyses did not support the existence of a new syndrome. Oregon and Washington Veteran Studies Investigators studied clusters of unexplained symptoms in a study of Portland area veterans by creating a new case definition of unexplained illness. Cases were identified on the basis of meeting a threshold number and combination of symptoms (cognitive and psychologic, and musculoskeletal) and on the basis of the duration of fatigue. Veterans whose symptom clusters remained unexplained at clinical examination (after exclusion of established diagnoses) were defined as constituting cases. Controls were those who at the time of clinical examination had no history of case-defining symptoms during or after their service in the Gulf War (Bourdette et al., 2001; Storzbach et al., 2000). The researchers undertook a factor analysis and then re-examined 48 symptoms in a second factor analysis. Three factors—cognitive and psychologic, mixed somatic, and musculoskeletal—were retained for followup factor analysis and accounted for 34.2% of the common variance. The authors used their three-factor solution to test the validity of their a priori case definition, which was composed of 35 symptoms encompassing musculoskeletal pain, cognitive and psychologic changes, gastrointestinal complaints, skin or mucous membrane lesions, and unexplained fatigue (discussed in greater detail in Chapter 5).

OCR for page 67
72 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS There were two major findings when the researchers compared the three-factor solution with the a priori case definition of Gulf War unexplained illnesses. First, their three factors did not include any symptoms related to the gastrointestinal system, the skin, or mucous membranes. Second, three of the symptoms—numbness in fingers or toes, clumsiness, and dizziness—were not included in the case definition. A limitation of the case-control design eliminated the possibility of examining differences between deployed and nondeployed veterans in that the study population by definition comprised only Gulf War veterans. United Kingdom Veteran Studies University of Manchester Veteran Study Cherry et al. (2001) extracted seven distinct factors on the basis of data collected in a large, population-based study of British Gulf War–era service members who answered 95 symptom questions. Deployed veterans—two random samples of Gulf War veterans (main and validation cohorts)—were compared with a stratified sample of service members who had not been deployed. The seven factors, which accounted for 48% of the variance, could be found in all three groups separately: psychologic (24 symptoms), peripheral (10 symptoms), neurologic (13 symptoms), respiratory (11 symptoms), gastrointestinal (six symptoms), concentration (10 symptoms), and appetite (five symptoms). Deployed veterans’ mean factor scores8 were significantly higher for five factors: psychologic, peripheral, respiratory, gastrointestinal, and concentration. No difference was found in the neurologic factor scores, and appetite factor scores were significantly lower in the nondeployed cohort. None of the factors was exclusive to Gulf War veterans, so the investigators concluded that their findings did not support the existence of a new syndrome (Cherry et al., 2001). Guy’s, King’s, and St. Thomas’s Schools of Medicine Studies Ismail et al. (1999) applied factor analysis to a representative sample of 7,379 UK veterans who served in the Bosnia conflict, who were deployed to the Gulf War, and who were not deployed. The researchers extracted three factors, which they labeled as mood and cognition (headache, irritability or outbursts of anger, sleeping difficulties, feeling jumpy or easily startled, unrefreshing sleep, fatigue, feeling distant or cut off from others, forgetfulness, loss of concentration, avoiding doing things or situations, and distressing dreams); respiratory system (unable to breathe deeply enough, faster breathing than normal, feeling short of breath at rest, and wheezing); and peripheral nervous system (tingling in fingers and arms, tingling in legs and arms, and numbness or tingling in fingers or toes). The pattern of symptom reporting by Gulf War veterans differed little from that by Bosnia and nondeployed era comparison groups, although the Gulf War cohort reported a higher frequency of symptoms and greater symptom severity. In addition to applying exploratory factor analysis, Ismail et al. (1999) used a particular application of confirmatory factor analyses (see Appendix A for more information). They tested a series of three models with different constraints and concluded that the factor structure did not differ significantly between the Gulf War–deployed and the Gulf War–era veterans. 8 Mean factor scores were computed by adding the sum of mean symptom scores (0–21) for each symptom that loaded onto the factor and dividing by the number of symptoms.

OCR for page 67
FACTOR ANALYSIS AND ITS USE 73 The UK authors interpreted their results as arguing against the existence of a unique Gulf War syndrome. Strengths of the study were its two comparison groups and its evaluation of the fit of the three-factor solution in the Bosnia and nondeployed cohort samples. The response rate of 65% may have introduced selection bias. Using two previously studied cohorts (Fukuda et al., 1998; Ismail et al., 1999), Nisenbaum et al. (2004) conducted a factor analysis of symptom data on 3,454 UK Gulf War veterans, 1,979 people deployed to Bosnia for UN peacekeeping operations, and 2,577 nondeployed era veterans. The researchers also compared results with those on a sample of 1,163 US Gulf War veterans, but these comparisons were limited by the fact that the US sample responded to a different survey. On the basis of visual inspection of results, the investigators observed considerable overlap in factor structure and some differences. They judged the findings not to “represent a unique illness or ‘Gulf War syndrome.’” Australian Cohort In a population-based study of all Australian Gulf War veterans, Forbes et al. (2004) applied factor analysis to findings from a 62-item symptom questionnaire that included measures of severity (“none,” “mild,” “moderate,” and “severe”). They found three factors that accounted for 47.1% of the variance: psychophysiologic distress (23 symptoms), cognitive distress (20 symptoms), and arthroneuromuscular distress (6 symptoms). Those were broadly similar to factors extracted in previous analyses and were the same as factors that were based on data collected from a sample of nondeployed Australian veterans. However, although the prevalence was similar among deployed and nondeployed veterans, factor scores were higher among the deployed than among the nondeployed; the authors noted that this indicated a greater severity of symptoms. They concluded that there was no evidence of a unique pattern of self-reported symptoms in deployed veterans. Seabee Studies Haley et al. (1997) studied a battalion of 249 naval reservists called to active duty for the Gulf War. More than half the battalion had left the military by the time of the study; 41% of the battalion participated in the study. Of those participating, 70% reported having had a serious health problem since returning from the Gulf War. The study was the first to examine groupings of symptoms in Gulf War veterans with factor analysis. Through standardized symptom questionnaires and a two-stage exploratory factor analysis, the investigators defined what they considered to be either six syndromes or six variants of a single syndrome, which they labeled impaired cognition, confusion–ataxia, arthromyoneuropathy, phobia–apraxia, fever–adenopathy, and weakness–incontinence. One-fourth of the veterans in the study (63) were classified as having one of the six syndromes. The study was limited by its lack of a comparison group; the authors were unable to comment on the uniqueness of the factors in relation to other groups of veterans. Haley et al. (2001) attempted to replicate their factor-analysis findings in a validation cohort, which was separate from their original cohort of Seabees. The validation cohort of 335 consisted of veterans who were living in north Texas and who had registered with a VA clinic in Dallas or were recruited by advertising. In the 2001 study, a more detailed questionnaire was used than in the earlier Seabee cohort in an effort to replicate the 1997 findings. Haley et al. (2001) undertook a series of analyses to test whether the factor structure that they found in the earlier cohort could be

OCR for page 67
74 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS replicated in the larger and more representative cohort. In their confirmatory factor analysis, they imposed an additional constraint by allowing only four symptoms per factor for each of five models and compared the results of the five models with their earlier findings by using structure- equation models. The five models had either 12 or 16 measured variables, which loaded onto three first-order factors and zero or one higher-order factor. In two models, the four additional variables (or symptom factors) were allowed to load onto the primary or higher-order factors. The three primary syndrome factors were impaired cognition, confusion–ataxia, and central pain (termed arthromyoneuropathy in the original study); and the four additional variables or secondary symptom factors were chronic watery diarrhea, chronic fatigue involving excessive muscle weakness, chronic fever and night sweats, and middle and terminal insomnia. The higher- order factor was the presence of an underlying single Gulf War syndrome posited to explain all variance and covariance among the three first-order factors. Overall, 29% of participants had one or more of the three first-order factors, defined by dichotomizing the syndrome factor scale at 1.5, as in the original study. Haley et al. (2001) interpreted the results as confirming a three- factor solution, originally extracted in the Seabee cohort (Model 1). They also concluded that the three syndrome factors probably represented a higher-order syndrome, such as a single Gulf War syndrome, and that some additional symptoms (the four secondary symptom factors) appeared in all three syndrome variants. The researchers suggested that the confusion–ataxia syndrome may represent a more severe form of a single Gulf War syndrome of which impaired cognition and central pain variants (the other two syndrome factors) were less severe forms. Knoke et al. (2000) applied factor analysis to data from a population of active-duty Seabees in response to the factor analysis conducted by Haley et al. (1997). The study population was drawn from US Navy construction-battalion personnel (Seabees) who were on active duty in 1990 and remained on active duty in 1994, when the study was conducted. The instrument contained 98 symptom questions. Among the 524 Gulf War veterans and 935 nondeployed Seabees, Knoke et al. (2000) performed three factor analyses: the first on the deployed Seabees, the second on the nondeployed Seabees, and the third on both. The three factor analyses accounted for 80%, 89%, and 93% of the total variance, and each extracted five factors. The factors were labeled insecurity or minor depression (27 symptoms), somatization (13 symptoms), depression (10 symptoms), obsessive-compulsive (7 symptoms), and malaise (7 symptoms). Knoke et al. (2000) derived standardized factor scores that were based on each solution and then evaluated whether the scores differed. Scores among the three analyses were similar for insecurity or minor depression; higher in Gulf War veterans for somatization, depression, and obsessive–compulsive; and higher in nondeployed Seabees for malaise.9 Somatization, depression, and obsessive–compulsive affected an excess of about 20% of Gulf War veterans. This indirect method of testing the differences in factor structure led them to conclude that deployed and nondeployed veterans report “more of the same symptoms and illnesses” and that “identifying a new syndrome such as the putative Gulf War syndrome is a difficult task and is unlikely to be accomplished by factor analysis, or any other statistical methodology, performed on a small, selected group of Gulf War Veterans.” The authors also conducted a discriminant analysis to test the ability of the factors to discriminate between Gulf War–deployed and nondeployed veterans: the probability of misclassification was 7.4% in nondeployed veterans and 76.5% in Gulf War–deployed veterans. The findings were similar to those of Doebbeling et 9 Factor scores used to compare the groups were computed from the regression coefficients of the Gulf War veteran factor analysis, standardized for both groups by subtracting the median and dividing by the semi-interquartile range of the score for the Gulf War veteran group.

OCR for page 67
FACTOR ANALYSIS AND ITS USE 75 al. (2000), Fukuda et al. (1998), and Ismail et al. (1999). They concluded that there was no evidence of a unique spectrum of neurologic injury. Iannacchione et al. (2011) conducted a validation study of the Haley et al. (1997) case definition in a larger population of Gulf War veterans. The study is discussed in Chapter 5. Pennsylvania Air National Guard Study In response to a request from the Department of Defense, VA, and Pennsylvania, Fukuda et al. (1998) used factor analysis and other methods to assess the health status of Gulf War Air Force veterans. The objective was to assess the prevalence and causes of an unexplained illness in members of one Air National Guard unit compared with three comparison Air Force units. The investigators aimed to organize symptoms into a case definition and to carry out clinical evaluations of members of an Air National Guard unit (the index unit). They administered a 35- item symptom inventory that included symptom severity (mild, moderate, or severe) and duration (less than 6 months or 6 months or longer) and divided the 3,255 participants who had answered all symptom questions into two subsamples of 1,631 and 1,624. The authors conducted a principal components analysis of the first subsample, extracting 10 components with eigenvalues10 greater than 1.0; three of the components accounted for 39.1% of the total variance. (See Appendix A for a discussion of the relationship between principal components analysis and factor analysis.) When the three components were examined in a confirmatory factor analysis in the second subsample, two were confirmed. The first, labeled mood–cognition– fatigue, consisted of these symptoms: feeling depressed, feeling anxious, feeling moody, difficulty in remembering or concentrating, trouble in finding words, difficulty in sleeping, and fatigue. The second, labeled musculoskeletal, consisted of these symptoms: joint stiffness, joint pain, and muscle pain. Fukuda et al. (1998) used 10 symptoms associated with the two factors from their confirmatory factor analysis to develop a preliminary case definition. Department of Veterans Affairs Gulf War Health Registry Hallman et al. (2003) examined patterns of reported symptoms in participants in the VA Gulf War Health Registry. The study sample consisted of a state-based random sample of 2,011 veteran registry members who resided in Delaware, Illinois, New Jersey, New York, North Carolina, Ohio, or Pennsylvania and who were not participating in other studies. Questionnaires included 48 symptoms, which were rated on a three-point ordinal scale, and were returned by 1,161 veterans (58% of the sample). The investigators divided the participants into two groups and conducted five factor analyses in each group to examine replicability. They identified four factors that accounted for 50.2% of the variance. The factors were mood–memory–fatigue (depression, anxiety, sudden mood changes, problems in concentrating and remembering, unexplained weakness, sleep problems, and unexplained fatigue); musculoskeletal (pain or numbness in joints or muscles); gastrointestinal (abdominal pain and gas, diarrhea, nausea, and vomiting); and throat and breathing (difficulty in swallowing, swollen glands, nose or sinus problems, coughing, difficulty in breathing, and difficulty in tasting). Like Cherry et al. (2001), Hallman et al. (2003) conducted a cluster analysis (see below) to examine consistency between the two statistical methods. The principal limitation of the study is the lack of a nondeployed 10 In the context of factor analysis, an eigenvalue is a measure of variance. It indicates the amount of variation in the dataset that is accounted for by each factor. The eigenvalue for a given factor is the sum of the squared loadings of each variable on that factor (Ismail et al., 1999).

OCR for page 67
76 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS control group, which limits its ability to identify factors that may have been peculiar to exposure in the Gulf War. By starting with presumably the most symptomatic subset of Gulf War veterans (those who had left the service and registered with the Gulf War Health Registry), the authors might have had the chance of identifying clusters unique to Gulf War veterans if they existed. However, the four factors that they identified were largely similar to factors identified by other Gulf War investigators (Fukuda et al., 1998). CLUSTER-ANALYSIS STUDIES Another data-reduction technique used by Gulf War investigators is cluster analysis. This technique has been used in three cohorts to determine how groups of patients who have particular symptoms may be related to one another (Cherry et al., 2001; Everitt et al., 2002; Hallman et al., 2003). Cluster-analysis methods are discussed in Appendix A. In brief, cluster analysis posits a set number of clusters (groups of people) and then finds a solution that assigns people to clusters in such a way as to minimize the distances between people within clusters on the basis of their symptoms. There are several methods of cluster analysis, but all studies discussed use k-means cluster analysis methods (see Appendix A). United Kingdom Veteran Studies Several groups of researchers have examined symptoms in UK veterans. Two conducted cluster analyses. University of Manchester Veteran Study Cherry et al. (2001) sequentially partitioned members of three cohorts (Gulf War deployed, a second group of Gulf War deployed for validation purposes, and nondeployed) by using scores on 95 symptoms reported on a visual analogue scale. The authors stated that they chose the number of clusters to fit “by eye,” choosing the largest number of clusters (six) when the clusters appeared to be similar among the three cohorts. In doing so, the authors precluded finding a Gulf War–specific cluster. Rather than showing the symptom means for each of the six clusters, they showed means by cluster of seven factor scores, which they had derived from a factor analysis of the same 95 symptoms, thus complicating the interpretation of the cluster analysis. Cluster 1 was composed primarily of well people and had a smaller proportion of Gulf War veterans (36.4%) than of nondeployed veterans (48.5%). Clusters 2 and 3 had similar prevalences in deployed and nondeployed groups. The final three clusters accounted for 23.8% of Gulf War veterans but only 9.8% of nondeployed veterans and included clusters with high scores on respiratory and gastrointestinal illnesses (cluster 4), on psychologic ill health (cluster 5), and both overall and especially on neurologic symptoms (cluster 6). There was an excess of 14% of Gulf War veterans in the three least healthy clusters. Guy’s, King’s, and St. Thomas’s Schools of Medicine Studies Everitt et al. (2002) randomly sampled 500 participants from among three cohorts: Gulf War veterans, Bosnia veterans, and nondeployed Gulf War–era controls. They regrouped the original 50 “nonspecific symptoms common in the general population,” which were then recategorized into 10 groups by body system with the same four-point severity score. They also used a technique, known as the gap statistic, that can be used to suggest the number of clusters that describe the data best (Tibshirani et al., 2001). The researchers identified five clusters by

OCR for page 67
FACTOR ANALYSIS AND ITS USE 77 using cluster analysis. Inspection of the five-cluster solution shows clusters that display increasing severity of symptoms rather than distinct patterns of co-occurrences. Cluster 1 had low scores on all symptoms, cluster 2 had the highest scores on musculoskeletal symptoms and high scores on neuropsychologic, cluster 3 had high scores on neuropsychologic and higher scores on the remaining nine symptom groups, cluster 4 had high scores only on musculoskeletal symptoms, and cluster 5 had high scores on all 10 symptom groups, especially musculoskeletal and neuropsychologic. Many more Gulf War veterans fell into clusters 2, 3, and 5 than Bosnia or nondeployed veterans. With the gap statistic, two clusters were identified: one with low scores on each symptom group and one with higher mean scores on the musculoskeletal and neuropsychologic groups. This analysis also assessed the relationships between cluster membership and other variables; only cohort membership was significantly associated with cluster membership. Some 72% of Gulf War veterans, 87% of Bosnia veterans, and 94% of era- deployed veterans were classified in cluster 1. The authors interpreted their findings to mean that there was no convincing evidence of a Gulf War syndrome. Because groups of 10 symptoms rather than individual symptoms were used for the analysis, the contribution of individual symptoms to cluster formation is unknown. Department of Veterans Affairs Gulf War Health Registry Hallman et al. (2003) conducted cluster analysis in their examination of 1,161 veterans who were participating in the VA Gulf War Health Registry. The researchers used the mean factor scores from their factor analysis to group respondents on the basis of severity of symptoms. Examining the two randomly divided subsamples five times each but using cluster analysis, they identified two stable clusters. Cluster 1, making up 60.4% of the sample, consisted of veterans who reported no or mild symptoms in each of the four factors. Cluster 2, the remaining 39.6% of participants, consisted of veterans who had moderate to severe factor scores in the mood–memory–fatigue and musculoskeletal factors and mild to moderate scores in the gastrointestinal and throat and breathing factors. People classified in cluster 2 reported twice as many symptoms (37.2% vs 17.8%) as and reported more severe problems, were in poorer health, and had a greater reduction in mean activity than people in cluster 1. SUMMARY AND CONCLUSIONS The studies described in this chapter, despite their methodologic differences, have findings that are similar, that is, similar groups of symptoms were identified as falling roughly into factors associated with fatigue, pain, and neurocognitive symptoms (see Table 4.1). Less commonly reported are factors that involve gastrointestinal and respiratory symptoms. Taken together, the studies’ findings do not support a unique syndrome, although they do highlight more frequent and severe symptoms in Gulf War–deployed than in nondeployed. Well-conducted factor analyses should have high participation rates and include representative samples. Some of the studies fall short on those two criteria. For example, several studies included members of only one branch of service (Fukuda et al., 1998; Haley et al., 1997; Knoke et al., 2000), collected small samples (Haley et al., 1997), or drew samples only from symptomatic groups of veterans (Haley et al., 1997; Hallman et al., 2003). Another problem is the lack of a comparison group in some of the studies, which limits investigators’ ability to compare factor structure in deployed and other groups (such as nondeployed and deployed elsewhere) (Bourdette et al., 2001; Haley et al., 1997). Although results of those studies are

OCR for page 67
78 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS valuable and add rich detail to the epidemiologic literature surrounding Gulf War veterans, other studies are more representative, and their results therefore more generalizable (Cherry et al., 2001; Doebbeling et al., 2000; Ismail et al., 1999; Kang et al., 2002). Findings of the more representative studies were quite similar, broadly describing neurologic, psychologic, cognitive, fatigue, and musculoskeletal symptoms. One exception to the congruence of those findings is the extraction by Kang et al. (2002) of factors that represented symptoms labeled as gastrointestinal, pulmonary, and upper respiratory. Three studies compared factors in representative deployed and nondeployed groups (Cherry et al., 2001; Doebbeling et al., 2000; Ismail et al., 1999). The factors extracted in those studies were remarkably similar between the deployed and nondeployed groups, and the findings did not suggest a unique complex of symptoms in the deployed group. In each of the studies, as in many of the less generalizable studies, symptoms were more frequent and more severe (in papers that reported severity) in the deployed than in the nondeployed groups. The three studies that used cluster analysis had broadly similar findings. The two studies that included nondeployed comparison groups (Cherry et al., 2001; Everitt et al., 2002) failed to identify a cluster of people that presented with a unique syndrome. Each study identified a highly symptomatic cluster of people that contained a higher proportion of Gulf War veterans than of non–Gulf War veterans, ranging from 14% in Cherry et al. (2001) to 22.2% in Everitt et al. (2002). The cluster analyses were consistent with the findings of the most representative factor- analysis studies: although they did not support a unique symptom complex in Gulf War veterans, they found that these veterans were more symptomatic than their nondeployed counterparts. Factor analysis and cluster analysis may be useful methods for making sense of the large number of symptoms potentially associated with CMI. However, the findings obtained with these methods must be validated against other observed variables. The choice of variables to include in a model is critical, and omission of key symptoms will result in models that do not capture the most salient features of CMI. In most of the studies, the percentage of variance explained is not great. The heterogeneity of the survey questions makes it difficult to account for a lot of the variance with only a few factors. Moreover, the validity of factor analysis or cluster analysis depends on the quality of data. Methodologic flaws in such studies can bias their results (Ismail and Lewis, 2006). The committee notes that neither factor analysis nor cluster analysis alone can directly produce a case definition; such definitions are the product of postprocessing of factor- analytic model results (for example, dichotomization of factor scores to operationalize a case definition). Given the historical dependence on factor-analytic methods in CMI studies and the methodologic flaws associated with many of them, the following are suggested practices for future factor-analytic studies. More details and definitions are available in Appendix A. For interested readers, excellent published resources provide systematic and extensive descriptions of factor-analytic methods; see, for example, Brown et al. (2012), Kline (2000), Norman and Streiner (2003), Pett et al. (2003), Rummel (1970), and Stewart (1981). • Describe the factor-analytic process in sufficient detail to allow replication, including the wording of each item, the methods used to decide the number of factors, the method of factor extraction, the method of factor rotation, and any additional postprocessing conducted by the authors. • Because factor analysis is a family of methods rather than a single method, select a factor-analytic approach that is aligned with the research question.

OCR for page 67
FACTOR ANALYSIS AND ITS USE 79 • In deciding how many factors to extract in an exploratory factor analysis, consider parallel analysis—a simulation-based approach in which factors are retained only if their eigenvalues exceed what would be obtained in random samples with no underlying factors. • Factor analysis is not ideal for application to dichotomous (such as yes–no) data, so symptom survey items should include multiple response categories for collecting symptom data. • Account for the measurement level of the data when conducting factor analysis; for example, use a polychoric correlation matrix, rather than a Pearson correlation matrix, to account for ordinal-level data. See Appendix A for further discussion of these matrices. • Select a factor rotation method that is consistent with expectations regarding the relationships among factors (for example, choose an oblique rotation when factors are expected to correlate with each other). • Explicitly test whether factor-analytic results are reproducible (for example, compare with a “holdout” sample and replicate in an independent sample). • Apply a confirmatory factor model only when there is an a priori hypothesis regarding data structure. • To evaluate whether the factor structure of symptom data differs in different populations (for example, in deployed vs nondeployed), posit and test the question as a formal statistical hypothesis.

OCR for page 67
80 TABLE 4.1 Factor Analyses of Gulf War Veteran Cohorts Reference Population Variables and Data Method Rotationa No. Factors Factors Identified Unique Factors in Isolated; GWVs? % Variance Explained Kang et al., Active and Ordinal; 47 Iterative Oblique 6; % not reported Fatigue or depression, Factors similar, but 4 2002 retired, symptoms coded as principal- neurologic, neurologic symptoms n = 19,383 0 = none, 1 = mild, factor analysis musculoskeletal– loaded on neurologic or 2 = severe. rheumatologic, factor for deployed but gastrointestinal, not nondeployed. pulmonary, and upper respiratory Doebbeling Active and Ordinal and Unknown Orthogonal 3; 35% in Somatic distress, Correlation between et al., 2000 reserve, dichotomous; 78 and oblique deployed, 30% in psychologic distress, derivative and n = 3,695 symptoms rated 0 nondeployed and panic validation samples; (not present) to 4 same factors in (extremely nondeployed. bothersome). Prevalence not stated. Bourdette et Active and Dichotomous; 69 PCA Orthogonal 3; 34.2% Cognitive– NA al., 2001 reserve, symptoms; response psychologic, n = 443 scale not reported. mixed somatic, and musculoskeletal Cherry et Active and Interval; 95 PCA Orthogonal 7; 48% Psychologic, All present; mean factor al., 2001 retired, symptoms; visual separately in 3 peripheral, scores higher in GWVs n = 11,914 analogue symptom cohorts: Gulf neurologic, for psychologic, scores from 1 to 21. War–deployed respiratory, peripheral, respiratory, (main), Gulf gastrointestinal, gastrointestinal, War–deployed concentration, and concentration; lower for (holdout or appetite appetite. validation sample), and not deployed

OCR for page 67
Reference Population Variables and Data Method Rotationa No. Factors Factors Identified Unique Factors in Isolated; GWVs? % Variance Explained Ismail et al., Active, Ordinal and Principal Orthagonal 3; ~20% Mood–cognition, No, but 3-factor 1999 n = 3,214 dichotomous; 50–52 factors respiratory system, solution fit less well in nonspecific and peripheral nervous Bosnian cohort than in symptoms rated as system GW-deployed and less absent, mild, well in nondeployed moderate, or severe; than in Bosnian cohort. cases with missing Prevalence not responses were mentioned. excluded. Nissenbaum UK GW Dichotomous; 50 Exploratory Othogonal 3; % not reported Respiratory, mood– Gastrointestinal et al., 2004 veterns, n = symptoms; scored factor cognition, and symptom factor 3,454; UK yes or no; symptoms analysis; peripheral nervous appeared in UK GW Bosnia- differed between US confirmatory veteran data; deployed, n = and UK cohorts. factor analysis musculoskeletal factor 1,979; US appeared in US GW GW veterans, veteran data. n = 1,163 Forbes et Active and Ordinal; 63 Unknown Orthogonal 3; 47.1% Psychophysiologic No. Prevalence similar al., 2004 retired, symptoms ranked by and oblique distress, cognitive but severity higher in n = 2,781 severity (none, mild, distress, and GWVs. moderate, or severe); arthroneuromuscular reduced to 62 distress because of low prevalence (seizures in preceding month); 28 items recoded from 4 to 3 categories; 25 to 2 categories. 81

OCR for page 67
82 Reference Population Variables and Data Method Rotationa No. Factors Factors Identified Unique Factors in Isolated; GWVs? % Variance Explained Haley et al., Active and Interval. Principal-axis Orthogonal 6; 71% Impaired cognition, NA 1997 retired (Navy), factor and oblique confusion–ataxia, and n = 249 analysisa arthromyoneuropathy, phobia–apraxia, fever– adenopathy, and weakness– incontinence Haley et al., Active, Continuous. Principal Orthogonal Forced into 5 Impaired cognition, Compared fit of factor 2001 reserve, factors (in and oblique models with 3 confusion–ataxia, and analysis with that found retired, developmental syndrome central pain in Haley et al. (1997) n = 335 sample) factors; 29% by using structural estimating equations. Some models also fitted higher-order factor “Gulf War syndrome” and loaded 4 additional symptoms (chronic fatigue involving excessive muscle weakness, chronic fever and night sweats, middle and terminal insomnia, and chronic watery diarrhea) onto higher-order factor. Knoke et Active Ordinal and Principal-axis Orthogonal 5; 80-93% Insecurity, Somatization, al., 2000 (Navy), dichotomous; 98 factor somatization, depression, obsessive– n = 1,459 symptoms. analysisb depression; compulsive. 3 times as obsessive–compulsive; common in GWVs vs and malaise NDVs.

OCR for page 67
Reference Population Variables and Data Method Rotationa No. Factors Factors Identified Unique Factors in Isolated; GWVs? % Variance Explained Fukuda et Active and Ordinal; 35 PCA followed Oblique 3; 39.1% Fatigue, 45% of deployed met al., 1998 reserve (Air symptoms, including by mood–cognition, and factor score–based case Force), severity (mild, confirmatory musculoskeletal pain definition of CMI vs n = 3,255 moderate, or severe) factor analysis 15% of nondeployed. and duration (<6 months or ≥6 months). Hallman et Retired, Ordinal; 48 Principal-axis Oblique 4; 50.2% Mood–memory– al., 2003 participants in symptoms; present factor fatigue, VA Gulf War or recurring, mild, analysisb musculoskeletal, Health moderate, or severe. gastrointestinal, and Registry who Data split into throat–breathing felt their halves. illness was service- related, n = 981 NOTE: CMI = chronic multisymptom illness; GW = Gulf War; GWV = Gulf War veteran; NDV = nondeployed veteran; PCA = principal components analysis. a Factor solutions are rotated to maximize high loadings and minimize low loadings. Orthogonal rotation algorithms extract factors so that they do not correlate with one another. Oblique rotation algorithms allow factors to correlate with one another. b Principal-axis factor analysis is a common kind of factor analysis. 83

OCR for page 67
84 CHRONIC MULTISYMPTOM ILLNESS IN GULF WAR VETERANS REFERENCES Bourdette, D. N., L. A. McCauley, A. Barkhuizen, W. Johnston, M. Wynn, S. K. Joos, D. Storzbach, T. Shuell, and D. Sticker. 2001. Symptom factor analysis, clinical findings, and functional status in a population-based case control study of Gulf War unexplained illness. Journal of Occupational and Environmental Medicine 43(12):1026-1040. Brown, B. L., S. G. Hendrix, D. W. Hedges, and T. B. Smith. 2012. Multivariate Analysis for the Biobeahvioral and Social Sciences: A Graphical Approach. Hoboken, NJ: John Wiley and Sons, Inc. Cherry, N., F. Creed, A. Silman, G. Dunn, D. Baxter, J. Smedley, S. Taylor, and G. J. Macfarlane. 2001. Health and exposures of united kingdom Gulf War veterans. Part II: The relation of health to exposure. Journal of Occupational and Environmental Medicine 58(5):299-306. Doebbeling, B. N., W. R. Clarke, D. Watson, J. C. Torner, R. F. Woolson, M. D. Voelker, D. H. Barrett, and D. A. Schwartz. 2000. Is there a Persian Gulf War syndrome? Evidence from a large population- based survey of veterans and nondeployed controls. American Journal of Medicine 108(9):695-704. Everitt, B., K. Ismail, A. S. David, and S. Wessely. 2002. Searching for a Gulf War syndrome using cluster analysis. Psychological Medicine 32(8):1371-1378. Forbes, A. B., D. P. McKenzie, A. J. Mackinnon, H. L. Kelsall, A. C. McFarlane, J. F. Ikin, D. C. Glass, and M. R. Sim. 2004. The health of Australian veterans of the 1991 Gulf War: Factor analysis of self- reported symptoms. Journal of Occupational and Environmental Medicine 61(12):1014-1020. Fukuda, K., R. Nisenbaum, G. Stewart, W. W. Thompson, L. Robin, R. M. Washko, D. L. Noah, D. H. Barrett, B. Randall, B. L. Herwaldt, A. C. Mawle, and W. C. Reeves. 1998. Chronic multisymptom illness affecting Air Force veterans of the Gulf War. Journal of the American Medical Association 280(11):981-988. Haley, R. W., T. L. Kurt, and J. Hom. 1997. Is there a Gulf War syndrome? Searching for syndromes by factor analysis of symptoms. Journal of the American Medical Association 277(3):215-222. [Erratum appears in Journal of the American Medical Association 1997; 278(5):388.] Haley, R. W., G. D. Luk, and F. Petty. 2001. Use of structural equation modeling to test the construct validity of a case definition of Gulf War syndrome: Invariance over developmental and validation samples, service branches and publicity. Psychiatry Research 102(2):175-200. Hallman, W. K., H. M. Kipen, M. Diefenbach, K. Boyd, H. Kang, H. Leventhal, and D. Wartenberg. 2003. Symptom patterns among Gulf War registry veterans. American Journal of Public Health 93(4):624-630. Iannacchione, V. G., J. A. Dever, C. M. Bann, K. A. Considine, D. Creel, C. P. Carson, H. Best, and R. W. Haley. 2011. Validation of a research case definition of Gulf War illness in the 1991 US military population. Neuroepidemiology 37(2):129-140. IOM (Institute of Medicine). 2006. Gulf War and Health, Volume 4: Health Effects of Serving in the Gulf War. Washington, DC: The National Academies Press. IOM. 2010. Gulf War and Health, Volume 8: Update of Health Effects of Serving in the Gulf War. Washington, DC: The National Academies Press. Iowa Persian Gulf Study Group. 1997. Self-reported illness and health status among Gulf War veterans: A population-based study. Journal of the American Medical Association 277(3):238-245. Ismail, K., and G. Lewis. 2006. Multi-symptom illnesses, unexplained illness and Gulf War syndrome. Philosophical Transactions of the Royal Society B: Biological Sciences 361(1468):543-551. Ismail, K., B. Everitt, N. Blatchley, L. Hull, C. Unwin, A. David, and S. Wessely. 1999. Is there a Gulf War syndrome? Lancet 353(9148):179-182.

OCR for page 67
FACTOR ANALYSIS AND ITS USE 85 Kang, H. K., C. M. Mahan, K. Y. Lee, F. M. Murphy, S. J. Simmens, H. A. Young, and P. H. Levine. 2002. Evidence for a deployment-related Gulf War syndrome by factor analysis. Archives of Environmental Health 57(1):61-68. Kline, P. 2000. The Handbook of Psychological Testing, Second Edition. New York: Routledge. Knoke, J. D., T. C. Smith, G. C. Gray, K. S. Kaiser, and A. W. Hawksworth. 2000. Factor analysis of self-reported symptoms: Does it identify a Gulf War syndrome? American Journal of Epidemiology 152(4):379-388. Nisenbaum, R., K. Ismail, S. Wessely, C. Unwin, L. Hull, and W. C. Reeves. 2004. Dichotomous factor analysis of symptoms reported by UK and US veterans of the 1991 Gulf War. Population Health Metrics 2. Norman, G. R., and D. L. Streiner. 2003. PDQ Statistics, Third Edition. BC Decker Inc. http://www.amazon.com/Pdq-Statistics-Series-Third-Edition/dp/1550092073 (accessed October 6, 2013). Pett, M. A., N. R. Lackey, and J. J. Sullivan. 2003. Making Sense of Factor Analysis: The Use of Factor Analysis for Instrument Development in Health Care Research. Thousand Oaks, CA: Sage Publications. Rummel, R. J. 1970. Applied Factor Analysis. Evanston, IL: Northwestern University Press. Shapiro, S. E., M. R. Lasarev, and L. McCauley. 2002. Factor analysis of Gulf War illness: What does it add to our understanding of possible health effects of deployment? American Journal of Epidemiology 156(6):578-585. Stewart, D. W. 1981. The application and misapplication of factor analysis in marketing research. Journal of Marketing Research 18:51-62. Storzbach, D., K. A. Campbell, L. M. Binder, L. McCauley, W. K. Anger, D. S. Rohlman, and C. A. Kovera. 2000. Psychological differences between veterans with and without Gulf War unexplained symptoms. Portland Environmental Hazards Research Center. Psychosomatic Medicine 62(5):726- 735. Tibshirani, R., G. Walther, and T. Hastie. 2001. Estimating the number of clusters in a data set via the gap statistic. Journal of the Royal Statistical Society. Series B (Statistical Methodology) 63(2):411-423.

OCR for page 67