ABSTRACT
OBJECTIVE To characterize patients diagnosed with multiple chemical sensitivity (MCS), chronic fatigue syndrome (CFS), or fibromyalgia (FM), to compare their level of function with Canadian population average values, and to assess factors associated with function.
DESIGN Chart review and abstraction of clinical information.
SETTING The Environmental Health Clinic (EHC) at Women’s College Hospital in Toronto, Ont, which is a provincial referral centre for patients with illnesses with suspected environmental links, especially MCS, CFS, and FM.
PARTICIPANTS A total of 128 consecutive patients diagnosed with 1 or more of MCS, CFS, or FM, seen between January 2005 and March 2006 at the EHC.
MAIN OUTCOME MEASURES Demographic and socioeconomic characteristics, comorbid diagnoses, duration of illness, health services usage, life stresses, helpful therapeutic strategies, and functional impairment measured by the Short Form–36, compared with Canadian population average values. Factors significantly associated with function in bivariate analyses were included in multiple linear and logistic regression models.
RESULTS The patient population was predominantly female (86.7%), with a mean age of 44.6 years. Seventy-eight patients had discrete diagnoses of 1 of MCS, CFS, or FM, while the remainder had 2 or 3 overlapping diagnoses. Most (68.8%) had stopped work, and on average this had occurred 3 years after symptom onset. On every Short Form–36 subscale, patients had markedly lower functional scores than population average values, more so when they had 2 or 3 of these diagnoses. Having FM, younger age at onset, and lower socioeconomic status were most consistently associated with poor function.
CONCLUSION Patients seen at the EHC demonstrated marked functional impairment, consistent with their reported difficulties working and caring for their homes and families during what should be their peak productive years. Early comprehensive assessment, medical management, and social and financial support might avoid the deterioration of function associated with prolonged illness. Education and information resources are required for health care professionals and the public, along with further etiologic and prognostic research.
Chronic fatigue syndrome (CFS), fibromyalgia (FM), and multiple chemical sensitivity (MCS) are relatively common chronic conditions with the potential to substantially limit functioning and health-related quality of life.1–7 The Statistics Canada 2005 Canadian Community Health Survey found that 1.2 million Canadians, or 5% of those aged 12 or older, reported having been diagnosed with at least 1 of these 3 illnesses: 1.3% with CFS, 1.5% with FM, and 2.4% with MCS.8 The overall prevalence rose with age to 6.9% for women aged 45 to 64—a period in women’s lives when they would usually be employed and active contributors to society. All 3 of these conditions have symptoms referable to multiple body systems, and studies in the United States show high rates of overlap among them.5,7,9,10 However, they each have some distinct features, as outlined in the case criteria and descriptions provided in Boxes 1 to 311–16; the most notable are the profound physical and cognitive fatigue of CFS, the widespread musculoskeletal pain of FM, and the common triggering of neurologic symptoms in MCS on exposure to diverse chemicals at previously tolerable levels or at levels tolerated by most people. There is a paucity of Canadian literature describing these commonly encountered illnesses, particularly in the clinical setting.
Chronic fatigue syndrome case criteria and description*
-
New onset, unexplained, persistent fatigue that substantially reduces activity
-
Postexertional malaise or fatigue with pathologic slow recovery
-
Sleep dysfunction
-
Pain in muscles and joints
-
Neurologic or cognitive manifestations–at least 1 symptom in 2 categories:
-
- autonomic
-
- neuroendocrine
-
- immune manifestations
-
-
Illness persisting at least 6 mo (3 mo in children); onset is usually distinct, but can be gradual
Fibromyalgia case criteria and description*
-
More than 3 mo of widespread musculoskeletal pain (bilateral, upper and lower body, axial)
-
Report of pain by patient in 11 or more of 18 tender points when palpated at 4 kg of pressure (nail blanches) in addition to the following symptoms13:
-
- fatigue, particularly after exertion
-
- neurologic or cognitive complaints
-
- headaches
-
- loss of sleep rhythm
-
- heat or cold intolerance
-
- emotional numbness or anxiety
-
- cardiovascular complaints (eg, dizziness, hypotension, heart rhythm abnormalities)
-
- marked weight change
-
Multiple chemical sensitivity case criteria and description*
-
Symptoms reproducible with exposure
-
Condition is chronic
-
Low levels of exposure provoke symptoms (lower than previously or commonly tolerated)
-
Symptoms improve or resolve when incitants removed
-
Responses occur to multiple chemically unrelated substances
-
Symptoms involve multiple organ systems
-
University of Toronto discriminating symptoms15
-
- Having a stronger sense of smell than most people and 2 of the following:
-
– feeling spacey
-
– feeling dull or groggy
-
– having difficulty concentrating
-
-
This study aimed to describe a clinical patient population in a Canadian context, to document functional status with the Medical Outcomes Study’s 36-item short form (SF-36)17 compared with Canadian population average values, and to determine whether level of function was associated with demographic or diagnostic characteristics.
METHODS
Setting and population
The Environmental Health Clinic (EHC) is the only government-funded, academically affiliated clinic in Ontario that is provincially mandated to provide assessment of patients with complex, chronic environmentally linked conditions. Patients are referred by physicians with either previous diagnoses or possible diagnoses of CFS, FM, or MCS, or suspicion of another environmentally related condition. With approval from the Research Ethics Board of Women’s College Hospital, data were extracted from charts of all consecutive patients seen between January 2005 and March 2006 who had been diagnosed by EHC physicians with at least 1 of CFS, FM, and MCS (N = 128).
Measures
Before being assessed, referred patients completed a 20-page intake questionnaire. The demographic questions, exposure history, and symptoms, including those necessary to meet diagnostic criteria for MCS, CFS, and FM, were derived from the University of Toronto Health Survey, an instrument reproducible and validated for MCS.18 Patient postal codes were linked to 2001 census dissemination areas to obtain median income and average value of dwellings.
Data extracted included the date of onset of the main symptoms, the date the patient last felt consistently well, and the date he or she stopped work, if applicable (permitting calculation of patient age for each); health care usage; helpful therapies; and sources of support. Stressful life events, including loss or illness of loved ones, loss of job, separation, divorce, addiction in self or someone close, and physical, emotional, or sexual abuse were summed before and after symptom onset.
The EHC physicians used consensus case definitions in diagnosing patients: for CFS and FM, the Canadian clinical working case definitions,12,13 which built on earlier consensus definitions11,14; and for MCS, a combination of the 1999 consensus criteria16 and 4 symptoms found to best discriminate among patients expected to have a high prevalence of the condition.15 Additional clinical diagnoses were also recorded by EHC physicians as indicated by history, physical examination, laboratory investigations, or consultant reports. Given the small sample size, diagnostic subcategories were combined, with more specific diagnoses subsumed under a limited set of labels.
All patients completed the SF-36, which assesses functional status on 8 descriptively named subscales: physical functioning, role limitations due to physical health problems, bodily pain, general health perceptions, energy and fatigue, social functioning, role limitations due to personal or emotional health problems, and emotional well-being.17 The derivation of these subscales has been well described by Ware and Sherbourne.17 Each has been shown to have high reliability and validity. They have been used in general population surveys of Canadians,19,20 in a range of patient populations, including those with CFS,1,2,7,21 FM,22–24 or MCS,6 as well as in a comparative study of all 3 groups.7 Scoring followed published guidelines,17 resulting in scale ranges from 0 to 100, with higher values representing better function.
Statistical analyses
Raw SF-36 score values were transformed into z scores based on the same Canadian age- and sex-specific population means and standard deviations.25,26 Two sub-scales, physical and emotional role limitation, exhibited nonnormal distributions, with a high proportion of values clustered at zero. The z scores for these 2 subscales were dichotomized into 2 groups (greater than or equal to the median and less than the median) for subsequent analysis.
Descriptive statistics were calculated for independent variables by CFS, FM, or MCS diagnosis or combination of diagnoses. Associations between SF-36 subscale z scores and patient characteristics were first examined in a series of bivariate analyses, including 1-way ANOVA (analysis of variance) tests, χ2 tests for dichotomous and categorical variables, and Spearman correlation and Satterthwaite t tests for continuous variables. Starting with all relevant variables significant at P < .10 in bivariate analyses, multivariate regression analyses were conducted, with reduction to variables significant at P < .05 in final regression models.
Linear regression was performed for normally distributed scores, with regression coefficients and 95% confidence intervals (CIs) reported; logistic regression was performed for physical and emotional role limitation scores, with odds ratios (ORs) and associated 95% CIs reported, corresponding to scores greater than or equal to the median.
RESULTS
Characterization of the patient population
Considerable overlap was evident in the distribution of MCS, CFS, and FM diagnoses (Table 1). All clinic patients had markedly low scores relative to Canadian population average values matched for age and sex categories. The lowest scores were among those with all 3 conditions (Figure 1).
Comparison of EHC patients and Canadian population averages (women, aged 35 to 54) on the 8 scales of the SF-36
BP–bodily pain, CFS–chronic fatigue syndrome, E/F–energy and fatigue, EHC–Environmental Health Clinic, EWB–emotional well-being, FM–fibromyalgia, GH–general health perceptions, MCS–multiple chemical sensitivity, PF–physical functioning, RL-E–emotional role limitation, RL-P–physical role limitation, SF–social functioning, SF-36–Short Form–36.
Characteristics of the EHC patient population: A) Sex, relationship status, income, health coverage, additional diagnoses, and strategies for dealing with health problems; B) age, life stresses, health services usage, paid work hours, and average value of dwellings.
Common additional diagnoses included depression (34.4%), irritable bowel syndrome (27.3%), food sensitivities (27.3%), sleep disorders (25.8%), and nutritional deficiencies (24.2%) (Table 1).
The mean age of the study sample was 44.6 years (SD 11.2), and patients were predominantly female (86.7%). On average, people stopped feeling well at age 36.9 (SD 12.6), usually before they identified the specific symptoms that brought them to the clinic. Among those who had stopped work (68.0%), a mean of 3 years had elapsed between symptom onset and their becoming unable to work. Of the 53 individuals who did not work for pay and did not receive disability benefits, 30 were married or cohabitating, likely supported by their partners’ incomes. The remaining 23 might have been living on savings or local social assistance.
The mean number of visits to a family physician in the 12 months before completion of the intake questionnaire was 10.7 (SD 8.7; n = 103); the mean number of visits to other physicians was 13.7 (SD 18.2; n = 123). These numbers are far higher than comparable 2005 Canadian Community Health Survey values of 2.9 (SD 4.3) and 0.79 (SD 2.0), respectively, for Canadian adults aged 30 to 60 or 3.4 (SD 4.6) and 0.98 (SD 2.2), respectively, for Canadian women aged 30 to 60.26
Factors associated with functional status
In bivariate analyses, any diagnosis of FM, either alone or with another condition, was associated with significantly lower scores on all subscales (P < .05) except physical role limitations. A diagnosis of CFS was associated with lower scores on physical function, emotional well-being, bodily pain, and general health (P < .05). In multivariate linear regression analysis, having FM remained strongly associated with lower function on all subscale models except general health. In logistic regression it was associated with better function on the emotional role limitations subscale and was not associated with physical role limitations (Table 2). A diagnosis of CFS remained significantly associated with lower physical function (R = −0.67, 95% CI −1.17 to −0.17) and emotional well-being (R = −1.04, 95% CI −1.56 to −0.52), while a diagnosis of MCS was significantly associated with improved function on the bodily pain subscale (R = 0.19, 95% CI 0.07 to 0.83), relative to the rest of the patient population (Table 2).
Factors associated with EHC patients’ functional status in linear and logistic regression: Negative coefficients and ORs signify that factors are associated with reduced function; blank cells represent results that were not statistically significant (P < .05).
Men generally reported lower functional status than women, significant for energy and fatigue (R = −0.72, 95% CI −1.23 to −0.13) and social functioning (R = −1.00, 95% CI −1.60 to −0.39) despite the small number of patients (n = 17). However, their functioning with respect to physical role limitations was significantly higher (OR 0.20, 95% CI 0.06 to 0.59) (Table 2). Neither reported lifetime stress nor abuse was associated with current function. While 42 of the 128 patients (32.8%) indicated that they had no support for dealing with their health problems or in their day-to-day lives, no significant difference on any SF-36 subscale was associated with lack of support. Unmarried people scored significantly better than married people on the emotional well-being sub-scale (R = 0.53, 95% CI 0.03 to 1.03) (Table 2).
The younger patients were when their symptoms began, the greater the reported reduction in their physical functioning. Longer duration of symptoms, as reflected by months passed since patients last felt well, was associated with greater emotional role limitation (Table 2). As SF-36 scores included in this analysis have been standardized to age- and sex-specific population average values, it appears that younger patients with CFS, FM, or MCS experience greater functional impairment.
Higher numbers of visits to family physicians reported in the previous 12 months were associated with lower scores on general health, and visits to other physicians were associated with lower physical function scores (Table 2). Patients who had not identified anything on the previsit questionnaire as being helpful for their health conditions (7.0%) had lower scores on all SF-36 subscales, with significant differences for social functioning (P = .003). Those who reported that avoiding observed environmental symptom triggers was helpful (28.9%) had higher odds for better emotional role function (Table 2). No other therapies, including use of naturopathic or homeopathic remedies, supplements, or dietary changes, were associated with significant differences in SF-36 scores. It must be stressed that patients seen at the EHC represent those with greater duration and severity of illness, and that some individuals who identified helpful management strategies might have been less likely to be included in our population because they did not present to the clinic for care.
Higher reported weekly hours of paid work were associated with better scores on physical function in multivariate analysis (Table 2). Patients in census areas with higher socioeconomic status, measured using the proxy of average value of dwellings, had better scores for physical function, bodily pain, and general health (Table 2).
DISCUSSION
The substantial functional impairment we observed among CFS, FM, and MCS patients is consistent with findings in other countries.1–4,6,7 The SF-36 profile for CFS differs from that for depression, in that the last subscale (emotional well-being) is relatively high.1,2 Although not necessarily comparable because of difference in methods, EHC patients appear to be more impaired than other persons of the same sex and age range with these disorders in the recent Canadian Community Health Survey, of whom a quarter reported “dependency.”8 Because the EHC is the only Ontario government clinic of its kind accepting referrals from family physicians and specialists, there are long wait times for assessment. This biases patient selection toward those with greater duration and severity of illness and lack of sufficient response to earlier management attempts, including symptom trigger avoidance and rest. The high number of visits to physicians that we observed is likely related to the complex nature of these conditions, the associated comorbidities, and the low reported functional status. Given the limited perceived effectiveness for these patients of current management approaches, clinical consultation might also represent a search for clearer answers and the need for ongoing medical support. Physician visits might also be influenced by reported limitations in social support and illness-related family relationship difficulties.
As EHC patients completed the SF-36 at their first visits, they might have been unaware that frequent exposure to triggering substances could mask associations with chronic symptoms.27 Information provided by the EHC about how to do elimination and reexposure challenges, a strategy consistently reported to be helpful,28–30 might have subsequently assisted EHC patients in identifying and avoiding more triggering substances. Similarly, before EHC consultation, patients can be unaware of the tendency for people with chronic low energy to be more active on days they feel somewhat better, then to be bedridden for several days.12 Instead, pacing of rest and gentle exercise have been found to be helpful.12,31 Nevertheless, reliance on recall for measurement of many of the historical variables might limit their validity. The small sample sizes in some of the diagnostic categories likely reduce their explanatory power.
To be diagnosed with FM, patients must experience widespread musculoskeletal pain. For that reason, poorer function related to bodily pain would be expected in this subgroup; but why our patients experienced poorer function on 4 other subscales and somewhat better function on the emotional role limitation subscale is less clear. Further, half our sample had MCS; this could be because MCS is more widely recognized than CFS and FM as being environmentally linked and, because referring physicians perceive EHC’s expertise to be in environmentally linked conditions, they are more likely to refer patients with MCS to the clinic.
The unclear origin and course of each disorder, the subjective nature of diagnostic criteria, and the absence of consistently abnormal physical (except with FM) and laboratory findings have generated difficulties for patients with CFS, FM, or MCS in securing disability support awards in Canada. This has recently led to commission of a literature review and report on environmental sensitivities (including MCS) by the Canadian Human Rights Commission,32 which recognizes CFS and FM as conditions overlapping with MCS. Low rates of employment and lack of access to disability support payments can lead to strains on household resources and greater declines in function, as suggested by the better function among those in wealthier residential census tracts.
A Canadian national research agenda with sustained funding is needed for comprehensive study of these 3 common conditions. Etiologic studies should build upon the recent identification of genetic polymorphisms associated with diagnoses of CFS,33–36 FM,37 and MCS,38,39 which might render patients with these conditions more susceptible to adverse effects from exposure to ubiquitous toxins or infectious agents. International collaboration would be worthwhile, given the promising research advances being made in the United Kingdom on differential gene expression linked to viral or pesticide exposures, which can perturb mitochondrial, immune, and neurologic functions.40,41 Prospective cohort studies, initiated earlier in the clinical course, could aid further understanding of prognosis and guide clinical management approaches.
Conclusion
Substantial functional impairments relative to Canadian same age and sex population averages were observed in EHC patients diagnosed with CFS, FM, or MCS, consistent with findings in other countries and with patients’ reported difficulties working and caring for their homes and families. The association of poorer function with younger age at symptom onset, lower socioeconomic status, having more than 1 of these diagnoses or FM alone suggests that early comprehensive assessment and medical management, as well as nondiscriminatory recognition by social service agencies with attendant financial support32 might avoid the deterioration of function associated with prolonged illness. Provincial referral centres like the EHC can serve as foci for identification of best practices, collaborative educational, and research efforts, as well as support to front-line health professionals. Expanded health professional training and continuing education in Canada, along with expanded public and professional information resources, such as those recently launched by the US Centers for Disease Control and Prevention,42 are imperative to reduce the large gaps in response to the burdens observed in these patients and their families.
Acknowledgment
We thank Nancy Bradshaw for her assistance with the logistics of chart access and Drs Riina Bray and Alison Bested for helpful reviews of the manuscript.
Notes
EDITOR’S KEY POINTS
-
Multiple chemical sensitivity, chronic fatigue syndrome, and fibromyalgia are relatively common chronic conditions–about 5% of Canadians have been diagnosed with at least 1 of these illnesses–but little Canadian research has explored them.
-
In this study, diagnosis of fibromyalgia was associated with decreases in physical functioning, emotional well-being, vitality (energy and fatigue), and social functioning and with increased bodily pain; chronic fatigue syndrome was associated with decreases in physical functioning and emotional well-being; and multiple chemical sensitivity was associated with less bodily pain than reported by the rest of the study population. Although the study population was pre-dominantly female, men generally reported lower functional status than women did.
-
This study was conducted among patients of the Environmental Health Clinic in Toronto, Ont, which might have biased patient selection toward those with longer duration and greater severity of symptoms.
POINTS DE REPÈRE DU RÉDACTEUR
-
Le HCM, le SFC et la FM sont des Canadiens fréquents – environ 5 % des Canadiens ont au moins un de ces diagnostics – mais peu d’études canadiennes y ont été consacrées.
-
Dans cette étude, le diagnostic de FM s’accompagnait d’une baisse de l’état fonctionnel physique, de la sensation de bien-être, de la vitalité et du fonctionnement social, et d’une augmentation des douleurs; le SFC, d’une baisse de l’activité physique et du bien-être émotionnel; et l’HCM, de moins de douleurs corporelles que ce qui est rapporté pour le reste de la population. Bien que les sujets de l’étude étaient surtout des femmes, les hommes rapportaient généralement une état fonctionnel inférieur à celui des femmes.
-
Cette étude portait sur des patients de l’Environmental Health Clinic de Toronto, Ont, ce qui pourrait avoir biaisé la sélection des patients en faveur de symptômes plus sévères et plus durables.
Footnotes
-
Contributors
All authors contributed to study conception and design. Ms Lavergne undertook the extraction of data from clinic charts, analysis of data, and drafting of the article. Dr Cole oversaw data analysis. Drs Kerr and Marshall acquired the data, assisted in the interpretation of results, and contributed to drafting the article. All authors provided critical review of the draft and approved the final version to be published.
-
Competing interests
None declared
-
This article has been peer reviewed.
- Copyright© the College of Family Physicians of Canada
References
In this issue
Jump to section
Related Articles
Cited By...
- Observations recentes concernant 3 troubles de sante mal reconnus: Encephalomyelite myalgique/syndrome de fatigue chronique, fibromyalgie et manifestations dintolerance au milieu/de sensibilite chimique multiple
- Recent insights into 3 underrecognized conditions: Myalgic encephalomyelitis-chronic fatigue syndrome, fibromyalgia, and environmental sensitivities-multiple chemical sensitivity