| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the University of Illinois at Chicago, Chicago, IL (R.R.T.); DePaul University, Chicago, IL (L.A.J.); and the Finch University of the Health Sciences/Chicago Medical School, Chicago, IL (S.C.J.).
Address reprint requests to: Renee R. Taylor, PhD, Department of Occupational Therapy (MC 811), University of Illinois at Chicago, 1919 W. Taylor St., Chicago, IL. E-mail: rtaylor{at}uic.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: A stratified random sample of 18,675 adults residing in diverse neighborhoods in Chicago completed a telephone-screening questionnaire. A control group without chronic fatigue (N = 74) and a group of individuals with chronic fatigue (N = 227) were identified and administered a semi-structured psychiatric interview. Stepwise logistic regression analyses predicting occurrence of current and lifetime psychiatric disorders according to chronic fatigue status and sociodemographics were conducted on this overall sample of 301 participants.
RESULTS: Chronic fatigue, low socioeconomic status, and unemployment were among significant predictors of overall Axis I psychiatric disorders. Chronic fatigue functioned as a predictor for mood and anxiety disorders (including posttraumatic stress disorder), but did not function as a predictor for somatoform disorders, substance abuse/dependence, and eating disorders. Low socioeconomic status and unemployment were significantly associated with current psychiatric disorder, and low socioeconomic status was also significantly associated with mood and anxiety disorders. Women were significantly more likely to experience mood disorder, and minorities (eg, African Americans, Latinos, and individuals of other ethnicity) were significantly more likely to report posttraumatic stress disorder.
CONCLUSIONS: Results support prior findings for increased rates of psychiatric disorder among individuals with chronic fatigue and highlight the roles of low socioeconomic status, unemployment, being a woman, and being classified as a minority in their association with certain psychiatric disorders.
Key Words: chronic fatigue, • sociodemographic, • community-based sample, • epidemiology.
Abbreviations: CFS = chronic fatigue syndrome;; SCID = The Structured Clinical Interview for the DSM-IV.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
In terms of specific diagnoses, a number of studies have found a significant positive association between chronic fatigue and depressive and anxiety disorders (6–8). While findings regarding a positive relationship between chronic fatigue and anxiety disorders have been relatively consistent, there has been some discrepancy in findings regarding the significance of the role of chronic fatigue in the diagnosis of depression (9, 10). Moreover, the relationship of other psychiatric diagnoses to chronic fatigue has not been given equivalent attention within the research community.
A primary explanation for the abundance of discrepant findings regarding psychiatric comorbidity in chronic fatigue involves the predominant use of nonrandom sampling procedures to select study participants (11). Most studies have relied on medical facility populations to conduct research in this area (12). Katon and Walker (13) illustrated that nonrandom sampling can lead to discrepant rates of psychiatric comorbidity through their observation that the relationship between fatigue and psychiatric comorbidity changes according to the type of population sampled. Comorbidity was strongest in tertiary care medical populations, less strong in primary care populations, and weakest in community-based populations.
Reliance on medical care samples is problematic not only because individuals differ in their level of access to health care and in their help-seeking behaviors, but also because medical care samples tend to be limited with respect to ethnic and socioeconomic diversity (3). The role of sociodemographic variables such as these is important to explore, because individuals of varying ethnic and socioeconomic backgrounds have been found to differ with respect to a number of issues including: health care practices (eg, nutrition, regular exercise, routine medical examinations); behavioral risk factors (eg, use of alcohol, drugs, and tobacco); access to adequate health care (eg, health insurance benefits and adequacy of care provided); level of psychosocial stress; amount of negative environmental exposures (eg, air pollution, lead, and other toxins); and level of hazard with respect to occupation (14–16).
In sum, little is known about the relationships between psychiatric disorder, chronic fatigue, and sociodemographic characteristics, such as parental status (children vs. no children), work status, socioeconomic status (education and occupation), sex, age, marital status, and ethnicity, within a diverse urban community-based population. Moreover, discrepancies in rates of psychiatric comorbidity between studies remain, and continued investigation of this issue using nonrandom medical facility samples biased by health care access and utilization patterns is frequently criticized (3). To reduce some of these prior limitations, the present study explored these relationships in a randomly selected, community-based sample of adults residing in diverse neighborhoods of Chicago. It was expected that individuals with chronic fatigue and those with fewer resources (ie, lower socioeconomic status and being unemployed) would demonstrate higher overall rates of current and lifetime psychiatric disorders as compared with control subjects.
Even less is known about the role of sociodemographic variables in the relationship between chronic fatigue and specific types of psychiatric disorders within an urban context. For example, is being a woman related to increased rates of mood disorders among individuals with chronic fatigue, or is this finding an artifact of increased health care utilization patterns among women (4)? A secondary objective of this study was to examine the differing roles of chronic fatigue and sociodemographic characteristics in relation to specific psychiatric disorders. This may help clarify previous discrepancies in findings regarding the role of chronic fatigue in specific psychiatric disorders. Consistent with the literature on this topic, an association between chronic fatigue and mood and anxiety disorders was predicted. In addition to chronic fatigue status, being a woman was also predicted to be associated with having an increased rate of mood disorder in this study.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
In Stage 1, we called a random sample of residential/working telephone numbers and screened respondents for chronic fatigue symptomatology. Procedures developed by Kish (18) were used to select one adult from each household to receive the telephone-screening questionnaire. The randomized telephone numbers were obtained from Survey Sampling Incorporated (details of the procedure are presented elsewhere, in (19)). The sample was stratified to represent a population living on the Southwest Side of Chicago and 8 additional community areas within Chicago, each containing residents from diverse ethnic and socioeconomic backgrounds. Based on estimates from the Geographic Information System software, the demographic breakdown of the combination of all areas sampled was approximately 43% African American, 24% Latin American (predominantly Mexican American and Puerto Rican American), and 33% White. In terms of income, 30% were estimated to be below the United States census poverty level and 15% were estimated to earn over $30,000 a year.
Individuals who indicated on the survey that they were experiencing severe fatigue, extreme tiredness, or exhaustion that was present for a period of 6 months or longer and the concurrent occurrence of four or more physical and cognitive symptoms (20), were defined as the chronic fatigue group. The control group was composed of individuals selected randomly from those who screened negative for chronic fatigue.
Those in the chronic fatigue and control groups were then invited to participate in Stage 2. In this stage, respondents received a semi-structured psychiatric interview to evaluate for the presence of DSM-IV, Axis I psychiatric disorders.
Participants
In Stage 1, we called 28,673 residential/working telephone numbers, and 18,675 adults (65.1%) were screened by phone for chronic fatigue symptomatology. This is a rather conservative response rate, as it includes households where an answering machine was reached. If we only included those residential/working numbers in which we reached an eligible household (N = 24,953) and we did not count answering machines, the completion rate would be 74.8%.
Of the 408 participants who screened positive based on the telephone survey, 227 (55.6%) agreed to participate in a semi-structured psychiatric interview. There were no significant differences on sociodemographic characteristics or fatigue severity scores between those 227 screened positive individuals who agreed to participate in the psychiatric interview, and the 181 screened positive individuals who declined to participate. The control group was composed of individuals selected randomly from the 18,260 screened negative participants. Of the 199 screened negative participants, 74 (37.2%) agreed to complete the psychiatric interview. There were no significant differences on sociodemographic characteristics or fatigue severity scores between the 74 screened negatives who agreed to participate and the 125 who declined. Participants declining participation in the psychiatric interview did so for a variety of reasons, the major reasons being unwillingness to dedicate the time required for the interview and reluctance to disclose information of a personal and psychological nature. Together, the chronic fatigue and control groups comprised a total sample of 301 participants.
Measures
The CFS Screening Questionnaire (21) was used in Stage 1 to select participants with chronic fatigue symptomatology and controls. This questionnaire has been found to be a valid and reliable measure that correctly discriminates individuals with various forms of chronic fatigue from those without it (21). It assessed whether each respondent was currently experiencing unexplained, severe fatigue for 6 or more months, and it included a checklist of diagnostic items from all three of the currently used definitions of CFS (20, 22, 23). These items included somatic complaints, such as sore throat, painful lymph nodes, and multi-joint pain. Sociodemographic characteristics assessed by this questionnaire included information on parental status (children vs. no children), work status, socioeconomic status, sex, age, marital status, and ethnicity. Hollingshead’s (24) scoring scale was used to classify socioeconomic status, and explicit definitions were provided to incorporate the variables of occupation and education. Continuous scores from this scale ranging from 5 to 54 were used to define socioeconomic status according to the following categories: low (5–14), low-middle (15–24), middle (25–34), middle–high (35–44), and high (45–54). The screening questionnaire took approximately 15 to 20 minutes to administer by phone.
The Structured Clinical Interview for the DSM-IV (SCID) (25), administered in Stage 2, is a valid and reliable semi-structured interview guide that approximates a traditional psychiatric interview. It has been used reliably by a number of researchers investigating rates of psychiatric comorbidity in individuals with chronic fatigue syndrome (22, 26, 27). The SCID begins with a semi-structured interview portion designed to yield a tentative diagnosis. This tentative diagnosis is then systematically assessed during the structured portion of the interview via embedded questions that conform to exact, Axis I criteria set forth by the DSM-IV. The SCID is designed to yield current and lifetime (at least one diagnosis at any time during an individual’s lifetime) diagnoses. In the present study, the SCID was administered by phone during a single session lasting 45 minutes to 1 hour. Other researchers have reliably used telephones to collect psychiatric data within the general population (28), and have successfully administered the SCID by telephone to individuals with chronic fatigue syndrome (27). Because valid administration of the SCID required the interviewer to possess strong clinical and diagnostic skills (29), Master’s level psychology clinicians were trained extensively in SCID administration, viewed SCID training tapes, and were monitored while conducting mock as well as actual study interviews. Following training, clinicians were supervised weekly by a licensed clinical psychologist. A Spanish version of the SCID was administered to Spanish-speaking participants by bilingual, Master’s level psychology clinicians (29).
Data Analysis
Forward stepwise logistic regression analyses were performed using the SPSS for Windows 9.0 REGRESSION (BINARY LOGISTIC) procedure. The forward stepwise procedure was chosen as the variable selection method because it examines the independent variables at each step for entry or removal from the regression equation based on specific entry criteria. Wald criteria were used to judge the statistical significance of predictors. These regression analyses were employed to determine whether chronic fatigue status (chronic fatigue vs. control) and any of the sociodemographic variables predicted differences in rates of the following psychiatric diagnoses, which served as the dependent variables in each of the analyses: a) all current Axis I psychiatric disorders; b) all lifetime Axis I psychiatric disorders; c) current mood disorders; d) lifetime mood disorders; e) current anxiety disorders; f) lifetime anxiety disorders; g) current posttraumatic stress disorder; h) lifetime posttraumatic stress disorder; i) somatoform disorders (current only); j) current substance abuse/dependence; k) lifetime substance abuse/dependence; l) current eating disorders; and m) lifetime eating disorders. Chronic fatigue status and all sociodemographic variables presented in Table 1 were loaded into the same model for each regression analysis.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Current and Lifetime Psychiatric Diagnoses
Consistent with what was predicted, results of the first stepwise logistic regression analysis indicated that chronic fatigue status and socioeconomic status were significant predictors of current psychiatric diagnosis. Participants with chronic fatigue exhibited significantly higher rates of current psychiatric diagnosis than controls (66% vs. 27%) [b = 1.57, OR = 4.79, 95% CI = 2.46 to 9.32, p < .01]. In addition, participants of low (71%) [b = 1.24, OR = 3.47, 95% CI = 1.10 to 10.97, p < .05] and low-middle (75%) [b = 1.55, OR = 4.73, 95% CI = 1.35 to 16.51, p < .05] socioeconomic status exhibited significantly higher rates of current psychiatric diagnosis than individuals of high (30%) socioeconomic status. The remaining social-demographic variables did not make significant contributions to the model.
Results of the second logistic regression analysis indicated that chronic fatigue status [b = 1.69, OR = 5.40, 95% CI = 2.91 to 10.02, p < .01] and work status [b = 1.41, OR = 4.10, 95% CI = 1.34 to 12.56, p < .05] were the only significant predictors of lifetime psychiatric diagnosis. Consistent with what was predicted, participants with chronic fatigue demonstrated significantly higher rates of lifetime psychiatric diagnosis than controls (83% vs. 50%). In addition, unemployed individuals demonstrated significantly higher rates of lifetime psychiatric diagnosis than individuals working full time (92% vs. 69%).
Specific Psychiatric Diagnoses1
To test relationships between chronic fatigue status, sociodemographic characteristics, and specific current and lifetime psychiatric disorders, a series of additional stepwise logistic regression analyses were conducted according to the procedures described previously. Consistent with the literature on this topic, the predicted association between chronic fatigue and mood and anxiety disorders was found. As predicted, being a woman was also found to be associated with having an increased rate of mood disorder. For current mood disorders, findings indicated significantly higher diagnostic rates in the chronic fatigue group as compared with the control group (32% vs. 7%). In addition, a significantly greater number of individuals of low socioeconomic status reported a current mood disorder than those of high socioeconomic status (40% vs. 11%). A significantly greater number of women reported a current mood disorder than men (31% vs. 14%). With respect to lifetime mood disorders, participants with chronic fatigue demonstrated significantly higher rates of mood disorder than controls (60% vs. 22%) and women demonstrated significantly higher rates than men (55% vs. 39%).
In terms of current anxiety disorders (not including posttraumatic stress disorder), participants with chronic fatigue demonstrated significantly higher diagnostic rates than controls (29% vs. 10%), and a significantly greater number of individuals of low (49%) and middle-low (53%) socioeconomic status reported current anxiety disorders than individuals of high socioeconomic status (18%). Results were similar for lifetime anxiety disorders (not including posttraumatic stress disorder). Individuals with chronic fatigue demonstrated significantly higher rates of lifetime anxiety disorders than controls (36% vs. 15%), and individuals of low (62%) and middle-low (60%) socioeconomic status exhibited significantly higher rates than individuals of high socioeconomic status (18%). In addition, women demonstrated significantly higher rates of lifetime anxiety disorders than men (49% vs. 33%). Rates of posttraumatic stress disorder were examined separately, and findings indicated that participants with chronic fatigue demonstrated significantly higher rates of current posttraumatic stress disorder than controls (18% vs. 5%). In addition, participants identifying themselves as Latin American (25%) or Multiracial/Other (24%) exhibited significantly higher rates than participants identifying themselves as White (7%). With respect to lifetime posttraumatic stress disorder, individuals with chronic fatigue demonstrated significantly higher rates than controls (25% vs. 7%), and African American (23%), Latin American (32%), and other (29%) ethnicities demonstrated significantly higher rates than Whites (10%).
Distinct from findings for mood and anxiety disorders, chronic fatigue status did not emerge as a significant predictor for somatoform disorders, substance abuse/dependence, or eating disorders. Sex emerged as the only significant predictor of lifetime substance abuse/dependence. Men (51%) demonstrated significantly higher rates than women (51% vs. 30%). There were no other significant sociodemographic predictors of any of these disorders.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
In comparison with other studies, percentage rates of current and lifetime psychiatric diagnoses in the present study are in the high range, suggesting that the experience of chronic fatigue appears to be closely associated with the occurrence of psychiatric disorder. Prior investigations have demonstrated that certain subgroups of individuals with chronic fatigue and psychiatric comorbidity display neuroimmunological and neurochemical abnormalities on various tests of brain functioning (31–35). These abnormalities are reported to affect psychiatric functioning and are thought to be associated with overlapping cognitive abnormalities observed in many individuals with chronic fatigue (36). Furthermore, previous research suggests that some individuals with chronic fatigue not only experience multiple physical, interpersonal, occupational, and financial losses as a result of their illness, but also report a prior history of stressful life events, trauma, and early disruptions in object relations, depending on the individual (37–39).
Rates of psychiatric comorbidity were not only high in the chronic fatigue group, but they were also high in the control group. The rate of 50% lifetime psychiatric comorbidity among controls in this study is markedly higher than general community estimates for lifetime psychiatric disorder in individuals without fatigue, which range from 23.9% to 26.2% (40). These rates may be, in part, attributable to some self-selection issues operative in the identification of controls in this study. Another explanation for the high rates of psychiatric comorbidity in both groups may involve unidentified risk factors associated with living in urban Chicago. One study analyzing predictive mental health needs assessment models identified both individual (age, sex, ethnicity, marital status, and educational status) and neighborhood (neighborhood social rank, lifestyle, and urbanization) risks for mental illness (41). It is also possible that the extensive time and involvement required for participation in this study caused individuals with fewer psychiatric complaints, and/or reluctance to disclose psychiatric symptomatology, to decline participation, leaving more symptomatic individuals more willing to participate.
In addition to chronic fatigue status, findings indicated that low socioeconomic status significantly predicted current psychiatric disorder within the overall sample. These results are consistent with findings from other studies (42–44). In part, results may be attributable to the effects of increased psychosocial stress associated with the circumstance of poverty, and to a synergistic relationship between limited educational status and decreased opportunities for employment (42–44). These findings must be interpreted with care, however, because they can not be generalized to the overall population; the sample employed in this study was biased toward containing more individuals with chronic fatigue than control subjects.
In terms of specific diagnoses, individuals with chronic fatigue demonstrated significantly higher rates of current (32%) and lifetime (60%) mood disorders when compared with controls (7% for current, 22% for lifetime), and significantly higher rates of current (29%) and lifetime (36%) anxiety disorders when compared with controls (10% for current, 15% for lifetime). These findings are comparable with the results of previous investigations that have found close associations between chronic fatigue, mood disorders, and anxiety disorders (45). Sex emerged as a significant predictor of current mood disorders, with women (31% for current, 55% for lifetime) receiving higher diagnostic rates than men (14% for current, 39% for lifetime) in the overall sample. Previous research has demonstrated a higher prevalence of depressive disorders in women as compared with men (46), and may, in part, be explained by neurochemical differences between men and women (47) as well as social and behavioral differences (48). A recent study found that negative life events and chronic life difficulties were more prevalent in women, and these variables were associated with an increased prevalence of depression in women (48).
Socioeconomic status emerged as a significant predictor for current mood disorders, and for current and lifetime anxiety disorders, with lower socioeconomic status being significantly associated with the occurrence of these disorders as compared with high socioeconomic status in the overall sample. A number of variables might have contributed to these findings, including the possibility that individuals of lower socioeconomic status tend to have more limited access to health insurance and mental health services, thereby leaving these disorders untreated and hence more likely to persist or reoccur over time. Moreover, prior research suggests that individuals of lower socioeconomic status tend to experience more personal and environmental distress, while at the same time having fewer resources (41–44, 49) .
Additional analyses of current and lifetime rates of posttraumatic stress disorder revealed significantly higher diagnostic rates of both current (18%) and lifetime (25%) posttraumatic stress disorder in individuals with chronic fatigue as compared with control subjects (5% for current, 7% for lifetime). Other studies have found an association between significant life stressors and chronic health conditions, such as infections, fatigue, and chronic fatigue syndrome (37, 38). In addition, individuals identifying themselves as African American, Latino American, or other ethnic groups demonstrated significantly higher rates of posttraumatic stress disorder than individuals identifying themselves as White. Other investigations have found increased risk factors associated with traumatic events among African Americans and Latino Americans in the United States (14–16).
The methodology used in this study is somewhat unique. Two groups of individuals with and without chronic fatigue were identified from a larger sample of 18,675 individuals. Individuals with chronic fatigue were not matched with controls, and the relatively smaller number of control subjects limits the interpretation of results. For example, the smaller number of control subjects may have limited power in the analyses, particularly in the logistic regression models. Another limitation of this study is its uniform reliance on self-report for participant selection and measurement of psychiatric diagnosis. One problem inherent in using self-report for participant selection involves the increased risk for heterogeneity within the resulting sample. Although participants in the study self-reported as experiencing chronic fatigue of unexplained origin, a current medical evaluation is the only way to assure that the fatigue could not have been explained by an undetected disease process. In addition to this issue, conclusions regarding the role of physiological processes in differing outcomes related to psychiatric comorbidity could not be corroborated by pathophysiological evidence taken from medical examination. In addition, the ability to generalize findings from this study and conclusions involving the general population are limited in two ways. First, all findings involving sociodemographic predictors of psychiatric disorders in the overall sample must be interpreted within the context of a sample biased toward containing more individuals with chronic fatigue than control subjects. Second, because the data are cross-sectional and analyses were correlational, temporal conclusions regarding cause-effect relationships between fatigue, sociodemographic variables, and psychiatric disorders could not be made. A final limitation of this study involves potential bias imposed by response rates, which tended not to be high.
In summary, this study confirms previous findings that individuals with chronic fatigue are more likely to receive current and lifetime psychiatric diagnoses than control subjects, and that chronic fatigue is a significant predictor for mood and anxiety disorders. The finding for associations between chronic fatigue, mood, and anxiety disorders, including posttraumatic stress disorder, may have implications for the future study of common biopsychosocial pathways in the cause and treatment of these conditions. In addition, the present investigation reveals significant relationships between low socioeconomic status, unemployment, sex, ethnicity, and certain Axis I psychiatric diagnoses. Health care program planning would benefit from consideration of these variables in targeting the occurrence and perpetuation of psychiatric disorders and chronic fatigue in diverse urban communities.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication April 12, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Dinos, B. Khoshaba, D. Ashby, P. D White, J. Nazroo, S. Wessely, and K. S Bhui A systematic review of chronic fatigue, its syndromes and ethnicity: prevalence, severity, co-morbidity and coping Int. J. Epidemiol., December 1, 2009; 38(6): 1554 - 1570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. M. Nater, J.-M. S. Lin, E. M. Maloney, J. F. Jones, H. Tian, R. S. Boneva, C. L. Raison, W. C. Reeves, and C. Heim Psychiatric Comorbidity in Persons With Chronic Fatigue Syndrome Identified From the Georgia Population Psychosom Med, June 1, 2009; 71(5): 557 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Fuller-Thomson and J. Nimigon Factors associated with depression among individuals with chronic fatigue syndrome: findings from a nationally representative survey Fam. Pract., December 1, 2008; 25(6): 414 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Nelesen, Y. Dar, K. Thomas, and J. E. Dimsdale The Relationship Between Fatigue and Cardiac Functioning Arch Intern Med, May 12, 2008; 168(9): 943 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. ter Wolbeek, L. J. P. van Doornen, A. Kavelaars, and C. J. Heijnen Predictors of Persistent and New-onset Fatigue in Adolescent Girls Pediatrics, March 1, 2008; 121(3): e449 - e457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Yamazaki, S. Fukuhara, Y. Suzukamo, S. Morita, T. Okamura, T. Tanaka, and H. Ueshima Lifestyle and work predictors of fatigue in Japanese manufacturing workers Occup. Med., June 1, 2007; 57(4): 262 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Garver, A. Lazarus, K. Rajagopalan, L. Lamerato, L. M. Katz, L. S. Stern, M. Dolgitser, and J. J. Doyle Racial Differences in Medication Switching and Concomitant Prescriptions in the Treatment of Bipolar Disorder Psychiatr Serv, May 1, 2006; 57(5): 666 - 672. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
