Fibromyalgia at an educational facility: is there a link to indoor air quality?

Introduction

On December 27, 1996, a local health department director sent a Request for Assistance to the bureau chief of environmental epidemiology at the Florida State Department of Health. Faculty from an educational facility had requested an epidemiological analysis of a potential sick building on campus. It was reported that four faculty members had been diagnosed with multiple sclerosis and 10 faculty members had been diagnosed with fibromyalgia.

Within a month of that request, the educational facility and the regional staff person for the Department of Environmental Epidemiology developed an agreement: They would cooperate in an effort to describe the rate and distribution of the two illnesses and to determine, if possible, whether there were any environmental links.

The building, known as Building W, housed staff from the dental, nursing, and allied health sciences disciplines. It was located on the southeast corner of campus. Laboratories in the building used substances that necessitated increased ventilation rates. Ventilation deficiencies involving fume hoods, exhausts, room exchange rates, and room pressurization, had been documented by private indoor air consultants hired by the college as early as 1993. Concern had focused on the laboratories located on the first floor.

In March 1997, the author met with the OSHA representative from the educational facility to receive a description of the situation to date. The epidemiological work and the work to be done in response to indoor air complaints were separated and assigned to the different bureaus of the Florida Department of Health (DOH). It was recognized that information regarding health status is sensitive, and assurance of confidentiality also was established. The issue of health records confidentiality was addressed with the vice president of administration during a telephone conversation in the same month.

In April, a DOH team at the Bureau of Environmental Epidemiology discussed the merits of the case. Several strategies were suggested, and a survey instrument was drafted. Since the faculty wanted to include a list of symptoms common to those with fibromyalgia and multiple sclerosis, a comparison population was needed so that rates at which the symptoms occurred could be compared. The rates of physician-diagnosed illnesses could be compared with expected rates found in literature.

In May, an expert in neurological disease at the Centers for Disease Control and Prevention (CDC) reviewed the survey and the list of symptoms consistent with multiple sclerosis and fibromyalgia. He advised that a narrow focus was best and provided literature on multiple sclerosis for reference.

TABLE 1

Numbers of Questionnaires Returned

                   Participants Who
               Returned Questionnaires   Females   Males   Mean Age

Building W               55                 48        7       51
Building F               34                 23       10       47
Totals                   89                 71       17       49

An agreement was made with the director of the health sciences about the timing of a survey. It was agreed that during the fall semester the survey instrument and a cover letter explaining the confidential nature of the data would be sent to the homes of all full- and part-time faculty who had been housed in the building. The educational facility reviewed the questionnaire, and it was pretested in the summer by faculty not involved in the survey.

In June, modifications were made to the questionnaire on the basis of faculty suggestions. The personnel director provided address labels for the survey group. Postage was paid by the educational facility. In addition, the educational facility selected the comparison building, which housed campus administration (Building F), according to the selection criteria.

In July, pretesting of the questionnaire was completed. Elicited comments were incorporated. A meeting was held with a facility representative, the project coordinator from the DOH and, to prevent duplication of effort, the private consultant who had conducted the initial assessment of ventilation. The private consultant had administered an occupational health and comfort questionnaire using a stock format provided by the American Industrial Hygiene Journal. The faculty had objected to answering questions about their health when the results were not confidential, and the private consultant had been an engineer rather than a health professional.

In September, the final versions of the questionnaire and the cover letter were presented to the educational facility. It was agreed that the survey should be completed before the holidays.

Methods

A cross-sectional study was developed to compare rates of symptoms consistent with fibromyalgia and multiple sclerosis in two buildings on campus. The comparison building did not have laboratories. Other building selection criteria were age and sex distribution of the populations; the age of the building; and the age of heating, ventilation, and air conditioning (HVAC) system. If warranted, a follow-up case control study could constitute another phase of the project since this study was intended only as an initial step to confirm the cases and describe an event.

In addition to comparing rates between buildings, the study compared rates found in Building W with rates reported in literature, and the comparison was analyzed for significance. A case definition was established. It was agreed that cases of fibromyalgia or multiple sclerosis must be physician diagnosed. To reduce recall bias, the 1997 calendar year was chosen for study.

In October, the personnel director provided the list of names and addresses of all full- and part-time faculty who worked in buildings W and E The envelopes were stamped ahead of time by the educational facility. In all, 140 questionnaire packets were mailed by October 29, 1997. Three envelopes were undeliverable. Completed surveys were received as early as the first week in November. By the end of November, 62 completed surveys had been received, and the return rate had slowed considerably. A reminder letter and another questionnaire were sent to those who had yet to respond. This second set of envelopes was mailed on November 24, before the Thanksgiving break. Survey protocol established a deadline of 90 days after the initial mailing for acceptance of questionnaires.

The data were entered into a records file in Epi Info 6 software. Assumptions that were made upon data entry were marked on the questionnaire. In some cases, participants were contacted for clarification.

Two questions on the survey instrument asked the participant to name others who might have fibromyalgia or multiple sclerosis. The cohorts listed by the participants were cross-checked against the list of those asked to participate. This process showed that some cohorts had not been included in the survey. In January and February, an effort was made to contact these individuals. Five persons were listed, and four were contacted by telephone. A mailing to two individuals resulted in returned questionnaires. A third individual did not participate in the survey because she did not work at the educational facility in 1997 at all. The fourth individual responded in mid-March, which was too late for data to be entered into the records.

Results

A total of 89 questionnaires (64 percent) were completed and returned for analysis. Of these, approximately 62 percent were returned from employees in Building W and 38 percent from employees in Building F (Table 1). The vast majority of participants were female (81 percent) and Caucasian or white (91 percent). The mean age of participants was 49 years.

Comparison Between Buildings

Fibromyalgia

There were eight cases of physician-diagnosed fibromyalgia (Table 2). Seven of the cases occurred in respondents from Building W. One case occurred in a respondent who worked in Building E All eight respondents who had physician-diagnosed fibromyalgia were female; six were white, one was black, and one did not report race.

The odds ratio was calculated at 5.02. Fisher's Exact test was performed since the cells numbered fewer than five. An exact lower 95 percent confidence limit was calculated at 0.59, and an upper 95 percent limit was calculated at 233.05. The difference between the rates found in Building F and Building W was not significant.

All respondents who reported physician-diagnosed fibromyalgia were contacted for a date of diagnosis. Seven out of the eight had been diagnosed with fibromyalgia 10 years or more after being hired by the educational facility. The eighth respondent with fibromyalgia had been diagnosed six years after being hired. The residential history associated with each case was reviewed. Two of the respondents with fibromyalgia reported having lived outside Florida: one in Kentucky before 1990 and one in New York before 1982. The average age of respondents with fibromyalgia was 51 years, which was only two years older than the average age of the participants.

TABLE 2

Responses to Survey Question About Physician-Diagnosed Fibromyalgia

                        Case      No Case      Total

Building W                7           46           53
Building F                1           33           34
Totals                    8           79           87

The rooms occupied by the fibromyalgia patients during the fall of 1997 were marked on the floor plans of Building W. This process showed an uneven distribution of cases with all seven cases working on the second floor at some point during the week. The second floor housed the nursing classrooms, offices and laboratories, and a suite of administrative offices. Only one of the respondents with fibromyalgia reported working on the first floor in the dental clinic and classroom; this person also had an office space on the second floor. All respondents reporting fibromyalgia worked more than 30 hours per week.

Multiple Sclerosis

One case of multiple sclerosis was reported. This person had been diagnosed with multiple sclerosis two years after being hired by the educational facility and worked in Building W. Another respondent with a possible case of multiple sclerosis commented that the attending physician described a syndrome of multiple sclerosis-like symptoms in January 1995. This person also worked in Building W and had been hired by the educational facility in 1989. No physician records were reviewed in this phase of the survey.

Rheumatoid Disease

Of the 84 respondents who answered this question, nine reported physician-diagnosed rheumatoid disease (Table 3). Eight of the respondents with rheumatoid disease were female; one was male. Seven worked in Building W; two worked in Building F. Five respondents with rheumatoid disease also reported physician diagnoses of fibromyalgia. A statistical analysis of these data produced an odds ratio of 2.60. As with fibromyalgia, a Fisher's Exact test was conducted. This test showed an exact lower 95 percent confidence limit of 0.45 and an upper 95 percent confidence limit of 27.06.

Neurological Disorders 

Out of the 81 participants who answered the question, five participants reported physician-diagnosed neurological disorders (Table 4). Four of the respondents with neurological disorders were female; one was male. Four worked in Building W; one worked in Building F. None of the respondents with neurological disorders also had physician-diagnosed fibromyalgia or multiple sclerosis. Statistical analysis produced an odds ratio of 3.07. A Fisher's Exact test revealed an exact lower 95 percent confidence limit of 0.28 and an upper 95 percent confidence limit of 155.68.

TABLE 3

Responses to Survey Question About Physician-Diagnosed Rheumatoid
Disease

                         Case            No Case            Total

Building W                 7                43                50
Building F                 2                32                34
Total                      9                75                84
TABLE 4

Responses to Survey Question About Physician-Diagnosed Neurological
Disorders

                         Case            No Case            Total

Building W                4                43                47
Building F                1                33                34
Total                     5                76                81

Symptoms

The questionnaire asked for information about symptoms that had lasted more than 24 hours per episode in the past year. A list of 14 symptoms, derived from the diagnoses of multiple sclerosis and fibromyalgia, was used. Comparison by building showed that for each of the symptoms except dizziness, participants who worked in Building W reported more episodes than those who worked in Building F (Table 5). The symptoms were sensory loss, visual loss, weakness, skin sensations, pain all over, pain in parts, fatigue, sleep difficulty, feelings of not being rested, muscle coordination, tremors, headaches, and loss of balance.

Comparison with Expected Rates

Multiple Sclerosis

The survey found no clusters or excess numbers of multiple sclerosis cases. The fact that only a single case of physician-diagnosed multiple sclerosis was found with this survey prevents comparison between buildings. Populations at risk for multiple sclerosis have been described as having the following demographic characteristics: white, female, 50 to 59 years of age, and living at the 37th parallel or higher (1). The population at this educational facility was described as predominantly white females, 40 to 49 years of age, living at the 30th parallel. The expected rate calculated less than one.

Fibromyalgia

Prevalence rates vary by type of setting (2). The rate that best fits the population at the educational facility is discussed as follows in David A. Nye's Physician's Guide to Fibromyalgia Syndrome:

Fibromyalgia syndrome (FMS) is an underdiagnosed disorder of unknown etiology affecting over 5% of the patients in a general medical practice ... and an estimated 2-4% of the general population ..., women more often than men. Patients complain that they ache all over. A large number of other symptoms are often present, particularly fatigue, morning stiffness, sleep disturbance, paresthesias, and headaches (3).

In this study, an overall nine percent rate of fibromyalgia was established; eight respondents answered "yes" to the question about physician diagnosis of this syndrome. A total of 87 persons answered the question. For Building W alone, the rate of fibromyalgia was calculated at 13 percent, but the number of participants dwindled to 53. An indirect comparison of these two populations was calculated with a standard morbidity ratio of 13 to 5 or 2.6 to 1. A test of statistical significance was not possible without knowing more about the population mentioned by Nye.

Rheumatoid Disease

Fibromyalgia is considered a rheumatoid disease. Of the nine survey participants who reported rheumatoid disease, five also reported diagnoses of fibromyalgia. The fibromyalgia prevalence rate doubles in patients at a rheumatology clinic (3).

The prevalence of physician-diagnosed rheumatoid disease in this survey was calculated as nine out of 84 responses, or 10.7 percent. In 1990, the rate of rheumatoid diseases in Florida was calculated as 19.1 percent (4).

Other Neurological Disease

The literature did not provide an expected rate for the classification "other neurological disease." Further analysis of neurological symptoms would have to include a more specific diagnosis.

Discussion

Phase I of the survey did not include information from occupants of the building for years before 1997. Case findings were thus limited. In the process of case finding for multiple sclerosis, the author spoke to a faculty member with multiple sclerosis who had not worked at the educational facility at all in 1997. It was reported over the telephone that two others who had occupied Building W also had been diagnosed with multiple sclerosis in 1995 or 1996. An additional cohort who had worked in Building F was referred to by a survey participant only with a common first name. If these cases all had been confirmed physician-diagnosed cases, the total would have been five. Four of these cases would have been in Building W, which may account for the initial report of four cases. If another phase of the survey is proposed, it could include case findings for years before 1997. A future retrospective study would, however, have difficulty locating all previous building occupants no longer employed by the educational facility. It would also be difficult for the participants to recall information with enough accuracy to be reliable.

TABLE 5

Cases of Episodes Lasting More Than 24 Hours in Past Year

Symptom                             Building W         Building F

Sensory loss                         3 (5.9%)                0
Visual loss                          7 (13.7%)           2 (6.3%)
Weakness                            16 (30.8%)           5 (15.6%)
Skin sensation                      14 (26.9%)           1 (3.1%)
Pain all over                        8 (15.4%)               0
Pain in parts                       18 (33.3%)           4 (12.5%)
Fatigue                             23 (44.2%)            8 (25%)
Sleep difficult                     24 (46.2%)            8 (25%)
Not rested                          23 (44.2%)           9 (28.1%)
Muscle coordination lost             4 (7.8%)                0
Dizziness                            6 (11.8%)           5 (15.6%)
Tremor                                1 (2%)                 0
Headaches                           18 (34.6%)           7 (21.2%)
Loss of balance                      8 (15.4%)           1 (3.1%)

The 13 percent rate of fibromyalgia in Building W was elevated compared with the rate in Building F. It was also elevated compared with the five percent rate expected in a general medical practice. The rate in Building W is not elevated, however, when compared with a 19 percent rate described in Norwegian females 20 to 49 years of age (5). More of the respondents from Building W were females (85 percent) than males (15 percent). It has been noted in literature that fibromyalgia affects females seven times more often than males (6).

To identify other non-laboratory-associated environmental links, a literature search was conducted. Current medical and scientific literature consistently cites no definite etiology for fibromyalgia. The American College of Rheumatology endorsed the term fibromyalgia in 1990 to replace other terms for the same condition (2).

Although the etiology of fibromyalgia is poorly understood, the Physician's Guide offers some suggestions:

It has been suggested that the pain of FMS is related to micro-trauma in deconditioned muscles and that exercise works by conditioning these muscles... FMS may be due to nonrestorative deep sleep... Although no specific inheritance pattern has been identified, an increased incidence in relatives of affected patients has been noted... Development of the syndrome may require a predisposing factor, possibly inherited, as well as a precipitating factor such as trauma, infection, stress, or sleep disruption. The immunologic abnormalities suggest an infectious etiology, but if FMS were infectious we would expect to see an increased incidence in spouses of an affected patient, and this is not the case (3).

Evaluations of the potential role of infections in precipitating fibromyalgia have been reported recently in the literature. A high prevalence (16 percent) of fibromyalgia syndrome among patients infected with hepatitis C was described in 1997 (7). Similarly, in a group of postpoliomyletis patients, 14 percent were diagnosed with fibromyalgia (8). Two retrospective studies suggest that emotional trauma plays a role in precipitating fibromyalgia. The first, from Canada, found "a statistical association between fibromyalgia and the frequency and severity of sexual abuse earlier in life" (9). The second study supported the association but stated that "sexual abuse does not appear to be a specific factor in the etiology of fibromyalgia" (10).

Conclusion

This study found a 13 percent rate of fibromyalgia among faculty housed in Building W. The distribution of the fibromyalgia cases did not indicate any exposure from the first-floor laboratories. Current literature also did not support the idea that an environmental laboratory contaminant had triggered the cases. The educational facility was not advised to pursue further epidemiological analysis.

Acknowledgements: The author gratefully acknowledges the valuable help of Raul Quimbo, Charles Helmick, Joyce Massie, Ron Maston, and Tom Belcuore.

Corresponding Author. Emily J. Wilson, R.S., M.P.H., Florida Department of Health, 1000 NE 16th Ave., Box 19, Gainesville, FL 32609. E-mail: <[e-mail]>.

REFERENCES

1. Baum, H.M., and B.B. Rothschild (1981), "The Incidence and Prevalence of Reported Multiple Sclerosis," Annals of Neurology, 10(5):420-428.

2. Wolfe, F. (1990), "Fibromyalgia" Rheumatic Disease Clinics of North America, 16(3):681-697.

3. Nye, D.A. (1997), "A Physician's Guide to Fibromyalgia Syndrome" <http://www.ncf.carleton.ca/fibromyalgia/fm-md.text>.

4. Helmick, C., S.P. Heyse R.C. Lawrence, E. Lloyd, and R.A. Pollard (1995), "Arthritis and Other Rheumatic Conditions: Who Is Affected Now, Who Will Be Affected Later?" Arthritis Care and Research, 8(4):203-211.

5. Forseth, K.O., and J.T. Gran (1993) "The Occurrence of Fibromyalgia-Like Syndromes in a General Female Population," Clinical Rheumatology, 12(1):23-27.

6. Wolfe, F. (1995), "The Prevalence and Characteristics of Fibromyalgia in the General Population," Arthritis and Rheumatism, 38(1):19-28.

7. Buskila, D., N. Hilzenrat, L. Neumann, A. Shnaider, E. Sikuler, and D. Zilberman (1997), "Fibromyalgia in Hepatitis C Virus Infection: Another Infectious Disease Relationship," Archives of Internal Medicine, 157 (21):2497-2500.

8. Trojan, D.A., and N.R. Cashman (1995), "Fibromyalgia Is Common in a Postpoliomyelitis Clinic "Archives of Neurology, 52:620-624.

9. Boisset-Pioro, M.H., J.M. Esdaile, and M.A. Fitzcharles (1995), "Sexual and Physical Abuse in Women with Fibromyalgia Syndrome" Arthritis and Rheumatism, 38(2):235-241.

10. Taylor, M.L., M.E. Csuka, and D.R. Trotter (1995), "The Prevalence of Sexual Abuse in Women with Fibromyalgia," Arthritis and Rheumatism, 38(2):229-234. buse in Women