Guest guest Posted January 16, 2003 Report Share Posted January 16, 2003 Source: Pain Volume 100, Issues 1-2, Pages 131-139 Date: November 2002 URL: http://www.sciencedirect.com/science/journal/03043959 A community-based survey of fibromyalgia-like pain complaints following the World Trade Center terrorist attacks ------------------------------------------------------------------------ --- G. Raphael (a,*), H. Natelsona (, Malvin N. Janal (a) and Sangeetha Nayak (a) a University of Medicine and Dentistry of New Jersey, New Jersey Medical School, 183 S. Orange Ave., BHSB F-1512, Newark, NJ 07103, USA b War-Related Illness and Injury Study Center, VA New Jersey Health Care System, 385 Tremount Ave., East Orange, NJ 07018, USA * Corresponding Author Contact Information Corresponding author. Tel.: +1-973-972-5462; fax: +1-973-972-8305; email: raphaekg@... Received 29 May 2002; accepted 12 July 2002. Available online 23 October 2002. Abstract A purported pathogenic mechanism for the development of fibromyalgia, a medically unexplained syndrome involving widespread pain, is stress and associated psychiatric disorder. The major stressor of recent World Trade Center terrorist attacks provides a natural experiment for evaluating this mechanism. This study sought to determine whether symptoms consistent with fibromyalgia increased post-September 11 and whether exposure to specific terrorism-related events or prior depression predicted symptom increase. In a large community sample of women in the New York/New Jersey metropolitan area (n=1312), a cohort initially surveyed for pain and psychiatric symptoms before September 11th were recontacted approximately 6 months after the attacks to assess current symptoms and specific terrorism-related exposures. `Fibromyalgia-like' (FM-L) four-quadrant pain reports consistent with a diagnosis of fibromyalgia were compared at baseline and follow-up. Result showed that FM-L rates did not increase significantly between baseline and post-attack follow-up. Event exposure did not relate to FM-L onset at follow-up, nor did depressive symptoms at baseline interact with event exposure. Depressive symptoms did not predict new onsets better than the extent of their comorbidity with FM-L at baseline. The failure to detect a significant increase in symptoms consistent with a diagnosis of fibromyalgia and the failure of new onsets of such symptoms to be accounted for by exposure to major stressors or prior depressive symptoms suggests that these hypothesized risk factors are unlikely to be of major importance in the pathogenesis of fibromyalgia. Author Keywords: Fibromyalgia; World Trade Center terrorist attacks; Community survey; Stress; Depression 1. Introduction Several recent studies (Galea and Schuster) have documented the impact of the terrorist attacks of September 11, 2001 on psychiatric and psychological symptoms in the community, most notably post-traumatic stress disorder (PTSD) and depression. Although popular media sources report that pain clinics have been flooded with patients after the attacks ( Goldstein and Thernstrom) and that these symptoms persist for months ( Young, 2002), no research to date has documented the impact of the attacks on pain in the general community. Major psychological trauma, such as that resulting from the terrorist attacks, is believed to play a role in the pathogenesis of some painful syndromes, especially those involving `medically unexplained pain' in such conditions as fibromyalgia, a syndrome of unknown etiology characterized by widespread non-articular musculoskeletal pain (Wolfe et al., 1990). Supporting this belief in stress as a fibromyalgia trigger are data from the experience of Gulf War veterans among whom the most commonly reported specific syndrome was fibromyalgia ( Anonymous, 1997). A second supportive body of research indicates that retrospectively assessed sexual and physical trauma is prevalent among women with fibromyalgia ( Boisset; ; and Walling). In addition, patients with fibromyalgia report elevated rates of PTSD ( Sherman et al., 2000). Nevertheless, the applicability of such data to post-attack changes in the prevalence of pain in the community is questionable. The only empirical report published to date on changes in pain following the attacks (Young et al., 2002) supports media proclamations of increased pain severity among chronic patients, but this increase may be specific to chronic pain patients seen in tertiary care settings. Such an increase may not be seen in individuals who did not seek care previously or who did not experience symptoms prior to the attacks. In addition, some data suggest that trauma can affect care-seeking behavior rather than painful symptoms per se ( and ). Other reports even raise doubts about the validity of general claims of increased medical care-seeking following September 11, as Veterans Affairs facilities' outpatient visits for general medical care did not significantly increase in the time period immediately surrounding the attacks ( Rosenheck, 2002). A second reason to question the role of major trauma in pain derives from the only prospective cohort study of adult pain as a function of court-documented early childhood trauma ( Raphael et al., 2001) in which no association was found between documented childhood trauma and pain in adulthood. Such findings raise questions about the role of trauma in the pathogenesis of pain in general and `medically unexplained pain syndromes' such as fibromyalgia specifically. Trauma is often ( Adler et al., 1989) assumed to have a most relevant role in the pathogenesis of these disorders. A recent review of the literature (Norris et al., 2002) concluded that somatic complaints are likely to increase following major disasters. However, such studies have been largely limited to casecontrol comparisons of exposed versus unexposed individuals or have relied on retrospective self-report of increased symptoms. Only a single previous study ( Bravo et al., 1990) was able to assess symptoms both prior to and after a disaster. The unavailability of pre-disaster symptom assessments in all other studies limits interpretation of post-disaster symptom complaints. In addition, whether pain-related complaints may differ in their response to disaster when compared with other related somatic complaints such as fatigue has not been reported. Beliefs about depression as a risk factor for fibromyalgia relate to hypotheses about stress as a trigger for fibromyalgia. Prior depression may be a risk factor for post-attack pain, especially under conditions of high stress for two reasons: the high rates of comorbidity between major depression and fibromyalgia (Epstein; Goldenberg; Hudson; Katz and Offenbaecher) and the view ( Blumer; Bohr and Meyer), although controversial ( Turk and Salovey, 1984), that fibromyalgia is a somatic manifestation of depression. Depression is a risk factor for chronicity or poor outcome in other conditions that are not viewed as affective spectrum disorders, such as myocardial infarction (Bush; Frasure and Frasure). However, if history of depressive symptoms prior to the attacks interacts with exposure to terrorism-related events to increase risk of fibromyalgia-like symptoms, the reason for the high prevalence of depression in fibromyalgia would be clarified, as would its link with trauma. To better understand the role of major psychological trauma in painful symptoms, the current study capitalized on a recent community-based survey of fibromyalgia symptoms among women in the New York/New Jersey metropolitan area by recontacting a subset of participants after the terrorist attack and reinterviewing them to address the following aims: (1) to determine whether painful symptoms indicative of fibromyalgia changed following the September 11th attacks; (2) to determine whether treatment-seeking patterns for pain changed following the September 11th attacks; (3) to determine whether non-pain symptoms of fatigue associated with fibromyalgia showed similar patterns of change; and (4) to test whether exposure to specific attack-related events, a history of depressive symptoms, and their interaction predict the occurrence of post-attack fibromyalgia-like symptoms. 2. Methods 2.1. OVERVIEW A subsample of women in the New York/New Jersey metropolitan area who were participating in a population-based survey of pain prior to September 11, 2001 were recontacted approximately 56 months after the terrorist attacks, to assess gradient of exposure to the attacks and current pain-related symptoms. 2.2. BASELINE SURVEY In September of 2000, after receiving Institutional Review Board (IRB) approval at the University of Medicine and Dentistry of New Jersey (UMDNJ), a community-based study of familial and psychiatric factors among women with fibromyalgia was initiated. Telephone surveys were conducted by randomly selecting telephone numbers from a purchased list of households identified as having an adult female member. The list was compiled from national white page telephone directories and scored for sex from secondary sources including magazine subscription lists, voter registration lists, and driver's license information. Thus, with the notable exception of excluding unlisted numbers and households in which a woman had not been identified as female through secondary sources, this process approximates a random digit dialing procedure, but with improved cost efficiency. Up to 12 attempts were made to contact an adult woman at each number. Telephone numbers were restricted to exchanges in Manhattan or within 45 min commuting time to Newark, New Jersey. Prior to September 11, 2001, 11,963 surveys had been completed. These initial surveys required first making a live contact in 23,870 households and determining that 12,178 of these households contained study-eligible women. Among the women completing surveys, 9042 (75.6%) agreed to permit additional survey by the research team. 2.3. POST-ATTACK FOLLOW-UP SURVEY After receiving UMDNJ IRB approval to conduct a follow-up survey, 2026 participants were randomly selected from among those who completed the baseline survey. Interviewing commenced on February 6, 2002 and was completed by April 1, 2002. Half of all surveys were completed by February 20th and less than 10% were completed after March 14th. Thus, most surveys represent an elapsed time since the attack of 56 months. The sample selected for follow-up was stratified on the basis of presence or absence of symptoms consistent with fibromyalgia (see Measures below) and location in Manhattan or nearby New Jersey. Of 1839 households successfully contacted after a maximum of 12 dialing attempts, follow-up surveys were completed with 1312 women (71.4%). Women with whom follow-up surveys were completed differed from women on whom follow-up surveys were not completed in being older (mean=43.0, s.d.=1.2 vs. MEAN=38.6, s.d.=1.3, respectively, t=7.85, P<0.001), better educated (mean=15.2 years, s.d.=2.6 vs. MEAN=14.9, s.d.=2.74, respectively, t=2.45, P<0.05), more likely white/Caucasian (72.9 vs. 65.4%, respectively, chi^2=12.28, P<0.001), but less likely to identify as Hispanic (8.1 vs. 15.2%, respectively, chi^2=24.68, P<0.001). Completers did not differ significantly from non-completers on the percentage identifying as Black/African-American (19.6 vs. 20.2%, chi^2=0.09, P>0.10) or on any psychiatric or pain-related measure. 2.4. MEASURES AT BASELINE AND FOLLOW-UP SURVEYS 2.4.1. Pain/fibromyalgia symptoms In addition to collecting demographic information, participants were asked whether they experienced self-defined `widespread pain', that is, pain in lots of different areas of their body. They were also asked if they had pain in their muscles, bones or joints lasting at least 1 week. The time reference for pain questions was the 3 months prior to the relevant survey, and answers were scored either yes or no. For those individuals who indicated pain lasting at least 1 week, questions regarding pain distribution were asked. These questions were derived from an instrument validated to efficiently screen for fibromyalgia in the general community (White et al., 1999a). In the validation study, individuals with pain lasting at least 1 week who also reported axialskeletal pain and limb pain on the right and left side were scored as screening positive for fibromyalgia. In that study ( White et al., 1999a), when comparing muscle palpation examinations conducted to assess American College of Rheumatology specifications for fibromyalgia ( Wolfe et al., 1990), the predictive value of a positive screen was 73.0% for women. Those individuals receiving a positive screening diagnosis for fibromyalgia are hereafter referred to as having fibromyalgia-like symptoms (FM-L) characterized by self-report of four-quadrant pain. Participants were also asked to rate pain severity and interference with activities due to pain. For pain severity, participants were asked to rate severity of their worst pain in the past 3 months on a 100-point scale, where 0=`no pain at all' and 100=`the worst pain they could imagine.' For interference, participants were asked to rate the degree to which their worst pain in the past 3 months interfered with their ability to do things, where 0=`no interference at all' and 100=`pain interfered completely.' Participants who reported at least 1 week of pain were asked whether they saw a doctor for their pain and whether they took any medication or treatment for their pain. The time frame for care-seeking questions was the past 6 months. 2.4.2. Fatigue Participants were asked whether they were currently having a problem with a lot of tiredness or fatigue. If they said that they were, they were next asked whether, compared to their activities before they were tired or fatigued, the fatigue made `no decrease,' `a mild decrease,' `a moderate decrease' or `a substantial decrease' in their normal activities. Chronic Fatigue Syndrome (Fukuda et al., 1994) requires a duration of at least 6 months of fatigue leading to substantial activity decrease. However, since the timing of the follow-up survey was less than 6 months after the attacks for a subset of participants interviewed in February, the duration criterion at both baseline and follow-up surveys was reduced to `36 months' at both time points. Participants who indicated that their fatigue caused a `substantial decrease' in their normal activities and had at least `36 months duration' were considered to have `chronic fatigue-like' symptoms (CF-L). 2.5. MEASURES AT BASELINE SURVEY ONLY 2.5.1. Depression For study purposes relevant at the time of conducting the baseline survey, participants were screened for lifetime history of major depressive disorder. In earlier research (Dohrenwend et al., 1999), an optimally weighted combination of psychiatric symptom items were shown to bear good approximation (i.e. sensitivity 76.587.0% and specificity of 82.487.0% for different subgroups) to a lifetime diagnosis of major depressive disorder derived through an intensive, semi-structured psychiatric interview administered by a trained mental health professional (SCID; Structured Clinical Interview for DSM-III-R) ( Spitzer et al., 1988). Items included stem questions from the Diagnostic Interview Survey ( Robins and Helzer, 1994) for depressed mood and anhedonia, a 100-point depression severity score for both depressed mood and anhedonia, and a question assessing whether the respondent ever felt that she was `on the verge of a nervous breakdown or emotional crisis.' Those individuals receiving a positive screening diagnosis for lifetime major depressive disorder were considered to have lifetime `major depressive-disorder-like' symptoms (MDD-L). 2.6. MEASURES AT FOLLOW-UP SURVEY ONLY 2.6.1. Terrorism-related events At the follow-up survey, respondents were asked a series of questions designed to assess exposure to specific events related to the September 11 attacks. These were adapted from F.N. Norris's exposure module available on the NIH's Office of Behavioral and Social Science Research's web site (http://obssr.od.nih.gov/Activities/911/attack.htm). From this questionnaire, we determined whether participants had exposure to any of the following major terrorism-related events of September 11, with percentages indicating the percent of the overall sample who indicated experiencing the event: respondent was in the direct vicinity of the World Trade Center (WTC) at the time of the attack (i.e. below 14th Street in Manhattan) (3.7%), immediate family member was in the direct vicinity of the WTC (1.4%), a close friend (for whom the respondent provided at least some emotional support after the attacks) was in the direct vicinity of the WTC (12.8%) , death of immediate family member (0.6%), death of close friend (not acquaintance or coworker) (8%), evacuated from work or other place because of safety fears (18.8%), stranded away from home or family (12.5%), respondent or person on whom the respondent relies for income was laid off from their job as a direct result of the WTC disaster (5.1%). In exploratory analyses related to results presented below, comparable analyses were conducted using specific events and number of events experienced. Given that conclusions were entirely consistent when using these strategies compared to a more parsimonious presentation in which individuals were scored as either experiencing one or more major terrorist events or no major event, the latter scoring procedure is presented in all results below. 2.6.2. Analytic notes Data were analyzed using SPSS ver 10.1 and weighted to account for unequal probabilities of selection for each person sampled and for non-response. For comparisons of symptom prevalence in the baseline versus follow-up periods, McNemar tests were used to test for changes in dichotomous measures. Paired sample t-tests were used for continuous measures. To identify risk factors for the development of FM-L after the attacks, multiple logistic regression was used in which demographic variables and baseline symptoms were entered at the first step, potential risk factors were entered at the second step, and interactions of risk factors were entered at the final step. 3. Results Table 1 shows the change in specific symptoms, care-seeking behavior, and syndrome-like diagnoses between baseline and follow-up. While rates of specific pain-related symptoms increased after September 11, rates of FM-L did not increase significantly. Rates of seeking care by a doctor for pain decreased, while rates of taking other types of treatment for pain did not change significantly. In contrast to changes in rates of pain symptoms, rates of fatigue decreased significantly and rates of CF-L remained unchanged. Pain severity and disability associated with pain were next examined among women who reported at least 1 week of pain in their muscles, bones or joints at both baseline and post-attack follow-up (n=308). At baseline, mean pain severity (mean=57.1, s.d.=23.7) was not significantly lower than post-attack follow-up pain severity (mean=58.9, s.d.=24.2) (t=1.15, P>0.10). The extent to which pain was reported to interfere with ability to do things increased between baseline (mean=44.3, s.d.=31.35) and follow-up (mean=49.4, s.d.=31.67) (t=2.75, P<0.01). Logistic regression analysis was next used to predict odds of FM-L at post-attack follow-up. At the first step (see Table 2), demographic factors which had shown an association (P<0.10) with follow-up FM-L status were entered into the analysis: age, education, income, race (coded white vs other). Status of baseline FM-L (present/absent) was also entered. At the next step, MDD-L and exposure to one or more attack-related events were entered into the model. At the final step, the interaction of MDD-L and event exposure was added. As shown in Table 2, both baseline FM-L and age were independent predictors of FM-L at post-attack follow-up. At step 2, we find that a history of MDD-L is a significant predictor of post-attack FM-L but exposure to one or more terrorism-related events is not. At the third and final step, the interaction between prior MDD-L and event exposure does not approach significance and does not support the hypothesis that the combined effect of stress and depression triggers FM-L illness. To determine whether post-attack onsets or offsets rather than symptom change in general could be a function of stress, depression or their interaction, we next divided the sample according to the presence or absence of FM-L at baseline. As shown in Table 3A, none of the factors in the model predicted the offset of FM-L for those women who had FM-L at baseline. Comparing Table 3A and B, we find that the full-sample relationship between prior MDD-L and post-attack FM-L represents a near doubling of the odds of onset of post-attack FM-L due to MDD-L in those who had no FM-L at baseline. Specifically, among these new onset cases of FM-L, 50% had MDD-L at baseline. In contrast, of those who continued to have a negative FM-L screening diagnosis at post-attack assessment, 39.1% had MDD-L at baseline. Since it was not clear whether the association between new onsets of FM-L and MDD-L was a function of the general comorbidity between MDD-L and FM-L at any point in time, logistic regression was used to predict pre-attack lifetime MDD-L as a function of having no FM-L at either assessment, stable FM-L at both assessments, or FM-L at follow-up assessment only. After controlling for demographic factors in the first step, a dummy variable representing presence of FM-L at followup only was entered. At the last step, dummy variables representing FM-L stability and FM-L at either baseline or follow-up assessment were added. As shown in Table 4, although there was a significant association between new onset of FM-L at follow-up (step 2), this relationship was entirely accounted for by the association between MDD-L and FM-L at either assessment period (step 3). 4. Discussion In general, pain symptoms increased in a 56-month period after the World Trade Center terrorist attacks. The increase in four-quadrant pain consistent with a diagnosis of fibromyalgia (FM-L) was not significant. In contrast to media reports, we found a significant decrease in participants' report of physician care-seeking for pain following the attacks. The general increase in pain symptoms other than FM-L was not paralleled by an increase in another somatic complaint, fatigue. Specifically, symptoms of fatigue decreased significantly following the attacks, and frequency of chronic fatigue-like illness did not change. Analyses did not support the hypothesis of an interaction between specific terrorism-related stressful exposures and prior major depression-like symptoms in increasing odds of FM-L. Although prior major depression appeared to be an independent risk factor for post-attack FM-L onset, additional analyses determined that the power of MDD-L to predict new onset between baseline and follow-up was not any greater than the extent of comorbidity between these two symptom patterns at baseline. Others have interpreted the ability of prior psychological symptoms to predict subsequent fibromyalgia-like pain in the community as a sign that unexplained pain may be a somatization of distress (Magni and McBeth). However, unless it can be shown that prior psychological symptoms predict pain over and above their comorbidity at a single time point, such a conclusion may be unwarranted. In the absence of such a finding, depression as a predictor of new pain onset may be more parsimoniously interpreted as depression that has developed as a reaction to pain symptoms which exacerbate and remit over time. In the rare study identifying first-onset cases of pain symptoms, a clearer causal inference between prior psychological status and pain may be possible. One such study to date (Von Korff et al., 1993) suggests that the role of prior depression as a risk factor for subsequent regional pain may differ as a function of pain site, finding that depression predicted new onset of headache and chest pain but not back pain. In contrast, a recent British birth cohort study ( Power et al., 2001) found that psychological distress 10 years earlier was a predictor of new low back pain onset. In both studies, questions remain about the certainty with which the absence of a lifetime back pain diagnosis can be made at baseline. Resolution of such questions is important, in order to rule out the possibility that psychological symptoms at baseline are still reactive to prior pain experience. Given that participants were interviewed at only two points in time, it is possible that increases in selected pain symptoms over a 6-month period are reflective of the natural history of such complaints in an aging population rather than indicating that such symptoms are uniquely elevated during a period of endemic stress. Prevalence of four-quadrant pain and fibromyalgia have been shown to be higher in older cohorts (White and Wolfe) and incidence has been shown to increase over time within a single cohort ( Burton; MacFarlane and Waxman). Terrorism-related events failed to increase odds of presenting with fibromyalgia-like symptoms, even when participants had a pre-attack history of depression-like symptoms. Since the entire nation was, to some extent, traumatized by the events of September 11, specific events may have had somewhat reduced salience and exposure to all possible terrorism-related events was not assessed. Similarly, the magnitude of the potential synergism between events and depression in increasing odds of post-attack fibromyalgia-like symptoms may have been attenuated by the unique context in which the attacks had a broad and pervasive impact on individuals who were not formally or directly exposed to terrorism-related events. It is also possible that terrorism-related events might have a greater impact in environments in which multiple, unpredictable events occur over long periods, as is true for large cities in Israel. Nevertheless, in other analyses not detailed here, we determined that both the number of events and presence of at least one terrorism-related event was related significantly to other follow-up measures not relevant to this report, i.e. self-reported depressed mood, anhedonia, symptoms of PTSD, and global self-reported stress severity. The failure of terrorism-related events to bear a significant relationship to FM-L symptoms suggests that FM-L symptoms are less related to external stressors than are clearly psychological symptoms. Given that the initial sampling frame for the first wave of data collection was restricted to women, the post-attack follow-up sample also failed to recruit men. Since women differ from men on both general symptom expression and pain processing (Barsky; Fillingim and Rollman), these findings may not apply to men in the community. Nevertheless, women are at increased risk for both depression ( Weissman and Weissman) and fibromyalgia ( White and Wolfe), factors which would be thought to increase risk of pain in the face of major stressors. A limitation to this study is its reliance on self-reported symptoms of four-quadrant pain as an approximation of standardized assessment for fibromyalgia. While such self-reports are reasonably efficient screening tools for fibromyalgia (White et al., 1999a), differences exist ( White et al., 1999b) among individuals who meet full American College of Rheumatology (ACR) criteria for fibromyalgia ( Wolfe et al., 1990) (i.e. four-quadrant pain combined with 11 or more tender points on palpation), and those who also have four-quadrant pain but who do not meet the tender point criterion. Specifically, pain severity, post-activity fatigue and weakness are more common among those meeting full ACR criteria than among those with four-quadrant pain alone ( White et al., 1999b). In addition to a false positive rate in women of about 25%, ( White et al., 1999a) preliminary data from our laboratory ( Janal et al., 2002) also indicate that the highly structured four-quadrant pain screening questions miss 1015% of true cases of fibromyalgia. Nevertheless, there is no reason to believe that the predictive value of the screening instrument changes from pre- to post-attack assessments. Thus, while the screening questions are unlikely to predict absolute rates of fibromyalgia, the relative value of the screen is likely to be constant over time. Therefore, the change in FM-L rates over time are likely to reflect similar change in rates of true fibromyalgia. Another potential limitation to this study is its reliance on dichotomous measures of terrorist event exposure, MDD-like symptoms, and FM-like symptoms. While this methodology was largely dictated by measurement constraints associated with the conduct of a large community survey, more intensive assessments of all these factors, yielding continuously scored and potentially more sensitive measures, might have yielded larger effects. Similarly, and especially given the secondary focus on fatigue, we limited our assessment of CF-like symptoms to only three questions. Use of a standardized but longer fatigue screening scale such as that developed by et al. (1999) might have yielded different conclusions about changes in chronic fatigue-like symptoms between assessment periods. The failure to demonstrate increases in fibromyalgia-like symptoms 56 months following a major disaster does not rule out the possibility of more immediate or fleeting elevations in such symptoms. Similarly, our failure to demonstrate increased care-seeking does not rule out the possibility that temporary increases may have occurred or that a subset of careseekers had greatly increased the frequency of their treatment visits within a shorter or longer-term period. What these data do indicate is that predicted increases in fibromyalgia-like symptoms and care-seeking were not detectable in a large community sample of women, nearly 6 months after the terrorist attacks. Were such differences to have been detected after 6 months, they would have more significance than immediate symptom increases, from both a public health and, potentially, etiological perspective. Given that women in the community with FM-like symptoms may differ from women seeking treatment for FM, our findings may diverge from other studies which conclude that stress and psychiatric factors (Bansevicius; ; Sherman; Van and Winfield) may play a role in FM. Stress may play a different role in exacerbation of FM-like symptoms in the community than in the development or exacerbation of a chronic pain syndrome for which a subset of sufferers ultimately seek treatment. Finally, the failure of prior depressive symptomatology to predict subsequent fibromyalgia-like pain symptoms beyond the extent attributable to their comorbidity at any time point, raises doubts about the extent to which fibromyalgia-like symptoms are triggered by stress and depression. Future research may yet support an alternative hypothesis that the comorbidity of depression and fibromyalgia is a function of the stress of living with medically unexplained pain. Acknowledgements This study was supported by NIH Grant # R01 DE13486 (S1). The authors thank Dr Bruce Dohrenwend for his helpful discussions regarding terrorism-related event assessment. Tables Table 1. Changes in pain, pain-related care-seeking, and fatigue prior to and following the September 11 terrorist attacks (weighted n=1312) ------------------------------------------------------------------------ --------------------------------- Symptom/syndrome or care-seeking behavior % 'Yes' at % 'Yes' at % 'Yes' at % 'No' at P value baseline follow-up baseline baseline but 'No' at but 'Yes' follow-up follow-up ------------------------------------------------------------------------ --------------------------------- 'Widespread' pain 25.4 28.9 14.9 18.9 0.012 Pain in muscles, bones, or joints 36.9 40.3 21.9 26.0 0.028 Fibromyalgia-like syndrome (FM-L) 10.5 11.8 6.6 7.9 0.218 Saw doctor for pain 23.1 16.5 16.4 9.2 0.000 Taken any medication or treatment for pain 26.4 24.1 16.7 14.2 0.108 Problem with a lot of tiredness or fatigue 40.3 37.2 23.0 19.0 0.033 Chronic fatigue-like (CF-L) 6.1 5.7 3.9 3.5 0.675 ------------------------------------------------------------------------ --------------------------------- Table 2. Logistic regression predicting fibromyalgia-like (FM-L) syndrome at post-attack follow-up from demographic factors, prior depression, the combined effect of attack-related events and Baseline Major Depression-like syndrome (MDD-L) (weighted n=1312)^a ------------------------------------------------------------------------ --------------------------------- Variables in model Step 1 Step 2 Step 3 ------------------------ ------------------------ ------------------------ Adjusted OR 95% CI Adjusted OR 95% CI Adjusted OR 95% CI ------------------------------------------------------------------------ --------------------------------- Age (decades) 1.27** 1.09,1.50 1.32** 1.11,1.55 1.31** 1.11,1.55 Education (years) 0.93 0.86,1.01 0.93 0.85,1.00 0.93 0.85,1.00 Income (US $20,000 blocks) 0.95 0.84,1.09 0.97 0.85,1.11 0.97 0.85,1.11 Race (white=1 vs. other=0) 0.79 0.51,1.21 0.78 0.51,1.21 0.79 0.51,1.21 Baseline FM-L 8.03*** 5.22,12.37 7.39*** 4.78,11.45 7.42*** 4.79,11.49 MDD-L + 1.58* 1.07,2.32 1.41 0.83,2.38 Terrorist event + 1.18 0.80,1.75 1.03 0.58,1.83 Interaction of MDD-L and 1.28 0.59,2.78 terrorist event ------------------------------------------------------------------------ --------------------------------- ^a * P<0.05, ** P<0.01, *** P<0.001. Table 3. Logistic regression predicting fibromyalgia-like (FM-L) syndrome at post-attack follow-up from demographic factors, prior depression, and the combined effect of attack-related events and Baseline Major Depression-like syndrome (MDD-L), for (A) participants with baseline FM-L (weighted N=579) and ( for participants without baseline FM-L (weighted N=733)^a ------------------------------------------------------------------------ --------------------------------- Variables in model Step 1 Step 2 Step 3 ------------------------ ------------------------ ------------------------ Adjusted OR 95% CI Adjusted OR 95% CI Adjusted OR 95% CI ------------------------------------------------------------------------ --------------------------------- (A) Age (decades) 0.94 0.68,1.31 0.95 0.68,1.33 0.95 0.68,1.34 Education (years) 1.02 0.88,1.18 1.02 0.88,1.18 1.02 0.88,1.18 Income (US $20,000 blocks) 0.79 0.61,1.02 0.77 0.60,1.01 0.77 0.59,1.01 Race (white=1 vs. other=0) 1.41 0.62,3.24 1.46 0.63,3.37 1.49 0.64,3.48 MDD-L + 0.85 0.39,1.85 0.67 0.24,1.90 Terrorist event + 1.37 0.64,2.93 1.00 0.30,3.31 Interaction of MDD-L and 1.70 0.36,7.91 Terrorist event ( Age (decades) 1.39*** 1.16,1.68 1.45*** 1.19,1.76 1.44*** 1.19,1.75 Education (years) 0.90* 0.82,1.00 0.89* 0.81,0.99 0.89 0.81,0.99 Income (US$ 20,000 blocks) 1.04 0.89,1.22 1.06 0.90,1.25 1.06 0.90,1.25 Race (white=1 vs other=0) 0.62 0.37,1.03 0.62 0.37,1.04 0.62* 0.37,1.03 MDD-L + 1.89** 1.20,2.97 1.78 0.97,3.26 Terrorist event + 1.12 0.71,1.78 1.04 0.53,2.04 Interaction of MDD-L and 1.15 0.46,2.85 Terrorist event ------------------------------------------------------------------------ --------------------------------- ^a * P<0.05, ** P<0.01, *** P<0.001. Table 4. Logistic regression predicting major depression-like (MDD-L) symptoms at pre-attack baseline from demographic factors, FM-L at either baseline or follow-up, FM-L change between baseline and follow-up, and onset of FM-L at follow-up (weighted n=1312)^a ------------------------------------------------------------------------ --------------------------------- Variables in model Step 1 Step 2 Step 3 ------------------------ ------------------------ ------------------------ Adjusted OR 95% CI Adjusted OR 95% CI Adjusted OR 95% CI ------------------------------------------------------------------------ --------------------------------- (A) Age (decades) 0.095 0.86,1.05 0.94 0.85,1.03 0.91 0.82,1.01 Education (years) 1.04 0.99,1.10 1.05 0.99,1.10 1.05 1.00,1.11 Income (US $20,000 blocks) 0.84*** 0.77,0.91 0.83*** 0.77,0.91 0.84*** 0.77,0.92 Race (white=1 vs. other=0) 1.06 0.80,1.40 1.08 0.81,1.43 1.08 0.81,1.43 FM-L onset at follow-up 1.66* 1.07,2.59 0.74 0.39,1.43 FM-L change between baseline 1.08 0.51,2.29 and follow-up FM-L at either baseline or 2.33** 1.31,4.13 follow-up ------------------------------------------------------------------------ --------------------------------- ^a * P<0.05, ** P<0.01, *** P<0.001. References et al., 1997. L.A. , L.A. Bradley, G.S. Alarcon, M. Triana-, R.W. , M.Y. and K.R. Alberts , Perceived physical and emotional trauma as precipitating events in fibromyalgia. Associations with health care seeking and disability status but not pain severity. Arthritis Rheum 40 (1997), pp. 453-460. Adler et al., 1989. R.H. Adler, S. Zlot, C. Hurny and C. Minder , Engel's `psychogenic pain and the pain-prone patient': a retrospective, controlled clinical study. Psychosom Med 51 (1989), pp. 87-101. et al., 1998. R.W. , L.A. Bradley, G.S. Alarcon, M. Triana-, L.A. , K.R. Alberts et al., Sexual and physical abuse in women with fibromyalgia: association with outpatient health care utilization and pain medication usage. Arthritis Care Res 11 (1998), pp. 102-115. Anonymous, 1997. Anonymous , Self-reported illness and health status among Gulf War veterans. A population-based study. The Iowa Persian Gulf Study Group. J Am Med Assoc 277 (1997), pp. 238-245. Bansevicius et al., 2001. D. Bansevicius, R.H. Westgaard and T. Stiles , EMG activity and pain development in fibromyalgia patients exposed to mental stress of long duration. Scand J Rheum 30 (2001), pp. 92-98. Barsky et al., 2001. A.J. Barsky, H.M. Peekna and J.F. Borus , Somatic symptom reporting in women and men. J Gen Intern Med 16 (2001), pp. 266-275. Blumer and Heilbronn, 1982. D. Blumer and M. Heilbronn , Chronic pain as a variant of depressive disease: the pain-prone disorder. J Nerv Ment Dis 170 (1982), pp. 381-406. Bohr, 1996. T. Bohr , Problems with myofascial pain syndrome and fibromyalgia syndrome. Neurology 46 (1996), pp. 593-597. Boisset-Pioro et al., 1995. M.H. Boisset-Pioro, J.M. Esdaile and M.A. Fitzcharles , Sexual and physical abuse in women with fibromyalgia syndrome. Arthritis Rheum 38 (1995), pp. 235-241. Bravo et al., 1990. M. Bravo, M. Rubio-Stipec, G.J. Canino, M.A. Woodbury and J.C. Ribera , The psychological sequelae of disaster stress prospectively and retrospectively evaluated. Am J Community Psychol 18 (1990), pp. 661-680. Burton et al., 1996. A.K. Burton, R.D. e, T.D. McClune and K.M. Tillotson , The natural history of low back pain in adolescents. Spine 21 (1996), pp. 2323-2328. Bush et al., 2001. D.E. Bush, R.C. Ziegelstein, M. Tayback, D. Richter, S. s, H. Zahalsky and J.A. Fauerbach , Even minimal symptoms of depression increase mortality risk after acute myocardial infarction. Am J Cardiol 88 (2001), pp. 337-341. et al., 2001. M.C. , A.J. Zautra and J.W. Reich , Vulnerability to stress among women in chronic pain from fibromyalgia and osteoarthritis. Ann Behav Med 23 (2001), pp. 215-226. Dohrenwend et al., 1999. B.P. Dohrenwend, K.G. Raphael, J.J. Marbach and R.M. Gallagher , Why is depression comorbid with chronic myofascial face pain? A family study test of alternative hypotheses. Pain 83 (1999), pp. 183-192. Epstein et al., 1999. S.A. Epstein, G. Kay, D. Clauw, R. Heaton, D. Klein, L. Krupp, J. Kuck, V. , D. Masur, M. Wagner, R. Waid and S. Zisook , Psychiatric disorders in patients with fibromyalgia. A multicenter investigation. Psychosomatics 40 (1999), pp. 57-63. Fillingim, 2000. R.B. Fillingim , Sex, gender, and pain: women and men really are different.Curr Rev Pain 4 (2000), pp. 24-30. Frasure-, 1991. N. Frasure- , In-hospital symptoms of psychological stress as predictors of long-term outcome after acute myocardial infarction in men. Am J Cardiol 67 (1991), pp. 121-127. Frasure- et al., 1993. N. Frasure-, F. Lesperance and M. Talajic , Depression following myocardial infarction. Impact on 6-month survival. J Am Med Assoc 270 (1993), pp. 1819-1825. Fukuda et al., 1994. K. Fukuda, S.E. Straus, I. Hickie, M.C. Sharpe, J.G. Dobbins and A. Komaroff , The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med 121 (1994), pp. 953-959. Galea et al., 2002. S. Galea, J. Ahern, H. Resnick, D. Kilpatrick, M. Bucuvalas, J. Gold and D. Vlahov , Psychological sequelae of the September 11 terrorist attacks in New York city. N Engl J Med 346 (2002), pp. 982-987. Goldenberg, 1989. D.L. Goldenberg , Psychological symptoms and psychiatric diagnosis in patients with fibromyalgia. J Rheumatol Suppl 19 (1989), pp. 127-130. Goldstein, 2001. Goldstein A. Terrorism tied to jump in pain problems. Washington Post. Washington, DC, A01. 2001. Hudson et al., 1992. J.I. Hudson, D.L. Goldenberg, H.G. Pope, Jr, P.E. Keck, Jr and L. Schlesinger , Comorbidity of fibromyalgia with medical and psychiatric disorders. Am J Med 92 (1992), pp. 363-367. Janal et al., 2002. M.N. Janal, K.G. Raphael and S. Nayak , Prevalence and comorbidity of fibromyalgia and myofascial temporomandibular disorder in a community sample. J Pain 3 (2002), p. 5. et al., 1999. L.A. , J.A. Richman, A.W. Rademaker, K.M. Jordan, A.V. Plioplys, R.R. , W. McCready, C.F. Huang and S. Plioplys , A community-based study of chronic fatigue syndrome. Arch Intern Med 159 (1999), pp. 2129-2137. Katz and Kravitz, 1996. R.S. Katz and H.M. Kravitz , Fibromyalgia, depression, and alcoholism: a family history study. J Rheumatol 23 (1996), pp. 149-154. MacFarlane et al., 1996. G.J. MacFarlane, E. , A.C. Papageorgiou, J. Schollum, P.R. Croft and A.J. Silman , The natural history of chronic pain in the community: a better prognosis than in the clinic?. J Rheumatol 23 (1996), pp. 1617-1620. Magni et al., 1994. G. Magni, C. Moreschi, S. Rigatti-Luchini and H. Merskey , Prospective study on the relationship between depressive symptoms and chronic musculoskeletal pain. Pain 56 (1994), pp. 289-297. McBeth et al., 2001. J. McBeth, G.J. Macfarlane, S. and A.J. Silman , Features of somatization predict the onset of chronic widespread pain: results of a large population-based study. Arthritis Rheum 44 (2001), pp. 940-946. Meyer-Lindenberg and Gallhofer, 1998. A. Meyer-Lindenberg and B. Gallhofer , Somatized depression as a subgroup of fibromyalgia syndrome. Z Rheumatol 57 Suppl 2 (1998), pp. 92-93. Norris et al., 2002. F.H. Norris, M.J. Friedman, P.J. , C.M. Byrne, E. and K. Kaniasty , 60,000 disaster victims speak. Psychiatry (2002) (in press). Offenbaecher et al., 1998. M. Offenbaecher, K. Glatzeder and M. Ackenheil, Self-reported depression, familial history of depression and fibromyalgia (FM) and psychological distress in patients with FM. Z Rheumatol 57 Suppl 2 (1998), pp. 94-96. Power et al., 2001. C. Power, J. , C. Hertzman, G. Schierhout and L. Li , Predictors of low back pain onset in a prospective British study. Am J Public Health 91 (2001), pp. 1671-1678. Raphael et al., 2001. K.G. Raphael, C.S. Widom and G. Lange , Childhood victimization and pain in adulthood: a prospective investigation. Pain 92 (2001), pp. 283-293. Robins and Helzer, 1994. L. Robins and J.E. Helzer , The half-life of a structured interview the Nimh Diagnostic Interview Schedule (Dis). Int J Methods Psychiatr Res 4 (1994), pp. 95-102. Rollman and Lautenbacher, 2001. G.B. Rollman and S. Lautenbacher , Sex differences in musculoskeletal pain. Clin J Pain 17 (2001), pp. 20-24. Rosenheck, 2002. R. Rosenheck , Reactions to the events of September 11. N Engl J Med 346 (2002), p. 6-29. Schuster et al., 2001. M.A. Schuster, B.D. Stein, L.H. Jaycox, R.L. , G.N. Marshall, M.N. Elliott, A.J. Zhou, D.E. Kanouse, J.L. on and S.H. Berry , A national survey of stress reactions after the September 11, 2001, terrorist attacks. N Engl J Med 345 (2001). Sherman et al., 2000. J.J. Sherman, D.C. Turk and A. Okifuji , Prevalence and impact of posttraumatic stress disorder-like symptoms on patients with fibromyalgia syndrome. Clin J Pain 16 (2000), pp. 127-134. Spitzer et al., 1988. R.L. Spitzer, J.B.W. , M. Gibbons and M. First Instruction manual for the Structured Clinical Interview for DSM-III-R, New York State Psychiatric Institute Biometrics Research, New York, NY (1988). et al., 1995. M.L. , D.R. Trotter and M.E. Csuka , The prevalence of sexual abuse in women with fibromyalgia. Arthritis Rheum 38 (1995), pp. 229-234. Thernstrom, 2001. Thernstrom M. Pain, the disease. New York Times. New York, December 16, 2001. Turk and Salovey, 1984. D.C. Turk and P. Salovey , `Chronic pain as a variant of depressive disease'. A critical reappraisal. J Nerv Ment Dis 172 (1984), pp. 398-404. Van Houdenhove et al., 2001. B. Van Houdenhove, E. Neerinckx, R. Lysens, H. Vertommen, L. Van Houdenhove, P. Onghena, R. Westhovens and M.B. D'Hooghe , Victimization in chronic fatigue syndrome and fibromyalgia in tertiary care: a controlled study on prevalence and characteristics. Psychosomatics 42 (2001), pp. 21-28. Von Korff et al., 1993. M. Von Korff, L. Le Resche and S.F. Dworkin , First onset of common pain symptoms: a prospective study of depression as a risk factor. Pain 55 (1993), pp. 251-258. et al., 1997. E.A. , D. Keegan, G. Gardner, M. Sullivan, D. Bernstein and W.J. Katon , Psychosocial factors in fibromyalgia compared with rheumatoid arthritis. Psychosom Med 59 (1997), pp. 572-577. Walling et al., 1994. M.K. Walling, R.C. Reiter, M.W. O'Hara, A.K. Milburn, G. Lilly and S.D. , Abuse history and chronic pain in women. Obstet Gynecol 84 (1994), pp. 193-199. Waxman et al., 2000. R. Waxman, A. Tennant and P. Helliwell , A prospective follow-up study of low back pain in the community. Spine 25 (2000), pp. 2085-2090. Weissman et al., 1996. M.M. Weissman, R.C. Bland, G.J. Canino, C. Faravelli, S. Greenwald, H.G. Hwu et al., Cross-national epidemiology of major depression and bipolar disorder. J Am Med Assoc 276 (1996), pp. 293-299. Weissman et al., 1984. M.M. Weissman, P.J. Leaf, C.E. Holzer, 3rd, J.K. Myers and G.L. Tischler , The epidemiology of depression. An update on sex differences in rates. J Affect Disord 7 (1984), pp. 179-188. White et al., 1999a. K.P. White, M. Harth, M. Speechley and T. Ostbye , Testing an instrument to screen for fibromyalgia syndrome in general population studies: the London Fibromyalgia Epidemiology Study Screening Questionnaire. J Rheumatol 26 (1999), pp. 880-884. White et al., 1999b. K.P. White, M. Speechley, M. Harth and T. Ostbye , The London Fibromyalgia Epidemiology Study: comparing the demographic and clinical characteristics in 100 random community cases of fibromyalgia versus controls. J Rheumatol 26 (1999), pp. 1577-1585. White et al., 1999c. K.P. White, M. Speechley, M. Harth and T. Ostbye , The London Fibromyalgia Epidemiology Study: the prevalence of fibromyalgia syndrome in London, Ontario. J Rheumatol 26 (1999), pp. 1570-1576. Winfield, 2000. J.B. Winfield , Psychological determinants of fibromyalgia and related syndromes. Curr Rev Pain 4 (2000), pp. 276-286. Wolfe et al., 1995. F. Wolfe, K. Ross, J. , I.J. and L. Hebert , The prevalence and characteristics of fibromyalgia in the general population. Arthritis Rheum 38 (1995), pp. 19-28. Wolfe et al., 1990. F. Wolfe, H.A. Smythe, M.B. Yunus, R.M. , C. Bombardier, D.L. Goldenberg, P. Tugwell, S.M. , M. Abeles and P. , The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 33 (1990), pp. 160-172. Young, 2002. Young MA. A nation in pain. New York Times. New York, 2002, A24. Young et al., 2002. M.A. Young, B.J. O'Young, S.A. Stiens, H. Hoffberg, D. Cassius, C. Narrow et al., Terrorism's effect on chronic pain perception: an analysis of a multi-center cohort. Pract Pain Manage Mar/Apr (2002), pp. 912. -------- © 2002 Elsevier B.V./ScienceDirect --------------------------------------------- Too much mail? Try a digest version. See http://www.co-cure.org/digest.htm Send posts to mailto:CO-CURE@... Join or leave the list at http://www.co-cure.org/sub.htm --------------------------------------------- --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.443 / Virus Database: 248 - Release Date: 1/10/2003 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.443 / Virus Database: 248 - Release Date: 1/10/2003 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.