| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry, Haukeland University Hospital, Bergen, Norway.
Address correspondence and reprint requests to Tone Tangen Haug, MD, PhD, Department of Psychiatry, Haukeland University Hospital, Bergen 5021, Norway. E-mail: tone.haug{at}psyk.uib.no
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
METHOD: The HUNT-II study invited all inhabitants aged 20 years and above in Nord-Trøndelag County of Norway to have their health examined and sent a questionnaire asking about physical symptoms, demographic factors, lifestyle, and somatic diseases. Anxiety and depression were recorded by the Hospital Anxiety and Depression Scale. Of those invited, 62,651 participants (71.3%) filled in the questionnaire. A total of 10,492 people were excluded due to organic diseases, and 50,377 were taken into the analyses.
RESULTS: Women reported more somatic symptoms than men (mean number of symptoms women/men: 3.8/2.9). There was a strong association between anxiety, depression, and functional somatic symptoms. The association was equally strong for anxiety and depression, and a somewhat stronger association was observed for comorbid anxiety and depression. The association of anxiety, depression, and functional somatic symptoms was equally strong in men and women (mean number of somatic symptoms men/women in anxiety: 4.5/5.9, in depression: 4.6/5.9, in comorbid anxiety and depression: 6.1/7.6, and in no anxiety or depression: 2.6/3.6) and in all age groups. The association between number of somatic symptoms and the total score on Hospital Anxiety and Depression Scale was linear.
CONCLUSION: There was a statistically significant relationship between anxiety, depression, and functional somatic symptoms, independent of age and gender.
Key Words: anxiety, • depression, • functional somatic symptoms.
Abbreviations: ME = myalgic encephalomyelitis;; FSS = functional somatic symptoms;; ECA = Epidemiological Catchment Area Study;; HADS = Hospital Anxiety and Depression Scale;; HADS-A = anxiety subscale of Hospital Anxiety and Depression Scale;; HADS-D = depression subscale of Hospital Anxiety and Depression Scale;; HADS-AD = comorbid anxiety and depression on Hospital Anxiety and Depression Scale;; HADS-T = total score on Hospital Anxiety and Depression Scale;; OR = odds ratio.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Clusters of medically unexplained symptoms have appeared under different labels like "chronic fatigue syndrome," "food intolerance," "myalgic encephalomyelitis (ME)," "fibromyalgia," "postviral syndrome," "candida syndrome," "chronic pain syndrome," "irritable bowel syndrome," and "amalgam syndrome." Even if there are few or no objective findings by clinical investigation, FSS, and especially musculoskeletal complaints, account for about 50% of long-term sickness compensation and disability in Norway (7).
Anxiety and depression are prevalent in FSS (8). Somatic symptoms like chest pain, dizziness, fatigue, abdominal discomfort, and musculoskeletal pains are common in both anxiety and depression. In primary care, patients with anxiety and depression usually present such somatic symptoms, whereas emotional symptoms are less likely to be mentioned if they are not specifically asked about by the interviewer (9,10). Physical complaints are seldom attributed to psychological causes, and the focus for clinical examination is kept on somatic conditions (11).
Several studies of the relationship among anxiety, depression, and somatic symptoms in general have been conducted in primary care. A depressive or anxiety disorder was present in about 30% of patients presenting with physical complaints (8). There are, however, few community studies on the prevalence and severity of FSS and associated psychiatric comorbidity. In the Epidemiological Catchment Area Study (ECA), the presence of physical symptoms was associated with at least a two-fold increase in anxiety or depressive disorders (3,12). Because anxiety and worry about symptoms are major reasons for consultations in primary care, and women more often seek medical consultation than men, there might be a selection bias of reported symptoms in studies from primary care and medical settings (13). Studies from the general population are, therefore, needed in order to estimate less biased prevalence of FSS.
The health study of Nord-Trøndelag County of Norway (HUNT-II) is a large community-based study, and we have earlier reported data from this survey on the association among anxiety, depression, and gastrointestinal symptoms (2,14). In this paper, we expand the analyses to all functional somatic symptoms. Our hypothesis was that the association between functional somatic symptoms and anxiety and depression is also statistically significant, indicating that there is no difference if the functional somatic syndromes are taken separately or together in a single entity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
From August 1995 to April 1997, HUNT-II sent personal invitations to take part in the study with an appointment date for physical examination and blood test as well as a questionnaire covering demographic factors, physical and mental health, and lifestyle factors like consumption of coffee, alcohol use, and smoking, to be brought back to the appointment. Among those invited aged 20 to 89 years (N = 92,100), 62,651 took part, which gave a response rate of 71.3%. Among these, 60,869 people filled in at least five of seven items on both the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS-A) and the depression subscale of the Hospital Anxiety and Depression Scale (HADS-D).
Somatic Symptoms and Disorders
Somatic symptoms recorded by the questionnaire were chest pain, headache, pain in the neck, shoulders, hands, breast-abdomen, upper-back, low back, hips, knees, nausea, heartburn, constipation, diarrhea, palpitation, breathlessness, impaired hearing, blurred vision, trouble walking, reduced motor activity, headache, and irregular menstrual bleeding during the last year. The questions asked were: "Have you been bothered by pain or discomfort during the last year from the back, neck stomach etc.?" and for somatic disorders: "Has your doctor told you that you have the following disorders: angina, arthrosis, hypertension, etc?" There were no physical examination or lab test, so the somatic disorders were identified by self-report alone. Patients who reported somatic disorders like myocardial infarction, stroke, hypertension, other cardiac diseases, diabetes, thyroid disease, rheumatoid arthritis, and cancer were excluded from the actual study (N = 10,492). A total of 50,377 people with no organic disease, who reported at least one physical symptom during the last year, were included into the study.
Two subgroups of patients were defined by the authors: one with 12 or more somatic symptoms labeled "somatization syndrome" and one with 5 or more somatic symptoms according to the concept of "abridged somatization" for application in primary care (15). These two subgroups were analyzed separately as to their associations with anxiety and/or depression.
Measurement of Anxiety and Depression
Anxiety and depression were measured by the HADS that consists of 14 items, 7 for the HADS-A and 7 for the HADS-D, which can be added up to a total score (HADS-T). Each item is rated from 0 (not present) to 3 (maximum), and the sum score range on the HADS-A and HADS-D, therefore, is from 0 to 21. The HADS is formulated in a language that is readily understandable (16). To avoid false-positive cases with screening in somatic contexts, somatic symptoms common in anxiety disorder and depression have been excluded from the scale. The HADS is well accepted in both psychiatric and nonpsychiatric settings and is completed in 2 to 6 minutes (17). The HADS-D focuses mainly on the reduced pleasure response aspect (anhedonia) of depression and the HADS-A on generalized anxiety and panic. The psychometric properties of HADS are excellent according to a recent review (18–21).
A valid rating of depression was defined as at least five completed items on the HADS-D and HADS-A. Those who filled in five or six items were also included in the study, and their score was based on the sum of completed items multiplied by 7/5 or 7/6, respectively.
Based on accepted cutoff levels on HADS (17), three groups of psychopathology were defined in this study: case level anxiety (HADS-A 
8, HADS-D < 8, group A), case level depression (HADS-D 8, HADS-A < 8, group D), and comorbid anxiety and depression (HADS-A 8, HADS-D 8, group AD).
Statistics
The data were analyzed by SPSS version 11.0. All analyses were performed with weighting to adjust for difference in response rate according to age and gender, and also age and gender differences between the population of Nord-Trøndelag County and the population of Norway, as in the National Comorbidity Study (NCS) (22).
Gender differences in population characteristics were tested by chi-square tests and independent samples t tests (Table 1), and chi-square tests were also used for tests of gender differences in prevalence of somatic symptoms (Table 2). Mean numbers of somatic symptoms and the prevalence of "somatization syndrome (12 or more somatic symptoms)" and "abridged somatization (5 or more somatic symptoms)" in groups A, D, and AD were presented with 95% confidence intervals (CI) in Table 3. Prevalence odds ratios for having five or more somatic symptoms as a function of HADS groups were calculated using binary logistic regression analysis, and results were presented both with and without adjustment for age, gender, education, smoking, work status, and a questionnaire for alcoholism evaluation (CAGE). Hypotheses on interaction by age and gender in the association between the HADS groups and somatic symptoms were tested using logistic regression analyses (Figures 1 and 2
).
|
|
|
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Table 2 lists the somatic symptoms examined. The prevalence of all somatic symptoms except breathlessness and functional impairment were significantly higher in women than in men (p < .05). The most prevalent symptom in females was headache (40.3%). Musculoskeletal complaints were also prevalent: 28.9% of the women complained of pain in the shoulders and 26.9% of pain in the neck. About one fourth of the women complained of constipation and heartburn (26.1% and 25.5%, respectively). In men, the most prevalent somatic symptom was heartburn (30.9%), followed by headache (24.1%) and low back pain (20.0%).
The association between somatic symptoms, case-level HADS anxiety, and/or case-level HADS depression is shown in Table 3. There are four major findings: 1) there was generally a strong association between case-level anxiety (group A) and/or case-level depression (group D) and somatic symptoms in both men and women; 2) group A and group D were both associated with somatic symptoms and the associations were equally strong; 3) the association between group AD and somatic symptoms was stronger than for group A and group D, respectively (mean number of somatic symptoms: in group A men/women, 4.5/5.9; group D, 4.6/5.9; group AD, 6.1/7.6; in neither A nor D, 2.6/3.6). This finding was consistent for mean number of somatic symptoms, as well as for "abridged somatization" and "somatization syndrome"; 4) even though the prevalence of somatic symptoms was higher in women than in men, the association between somatic symptoms, case-level HADS anxiety, and/or depression was present in both genders. This is illustrated in Figure 1, where the mean level of the somatic symptoms increased linearly with the HADS-T score. The plots in men and women are parallel, illustrating no gender interaction.
Figure 2 shows the odds ratio (OR) for more than four somatic symptoms in relation to case-level HADS anxiety and/or depression with crude effects and effects adjusted for demographic and lifestyle factors. In case-level HADS anxiety, the OR for more than four somatic symptoms was 3.0, in case-level HADS depression 2.7, and in comorbid anxiety and depression 5.1. The association between "abridged somatization" and anxiety disorder and/or depression could only to a minor degree be explained by age, gender, education, smoking, work, or harmful alcohol consumption.
The percentage of the population with case-level HADS anxiety and/or depression and the proportion of population with five or more functional somatic symptoms in relation to age are shown in Figure 3. The proportion of population with case-level HADS anxiety and/or depression and the proportion of population with five or more somatic symptoms increase with age. However, the slope of the curve for somatic symptoms is steeper than the curve for case-level HADS anxiety and/or depression, which indicates that the increase in number of somatic symptoms with age is higher than the increase in anxiety disorder and/or depression. The percentage of the population with case-level HADS anxiety and/or depression in individuals with and without five or more somatic symptoms is shown in Figure 4. In people with fewer than five somatic symptoms, the prevalence of case-level HADS anxiety and/or depression is about 15%, and this proportion increases with age. In people with 5 or more somatic symptoms, the prevalence of case-level HADS anxiety and/or depression is about 35% in all age groups. As these plots are not parallel, there is a significant interaction by age in the association between somatic symptoms and anxiety and depression (p < .01).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Somatic symptoms like fatigue, chest pain, abdominal discomfort, and dizziness are included in the diagnostic criteria of both anxiety disorders and depressions, and this may explain the high prevalence of somatic symptoms in patients with these conditions. Moreover, patients with anxiety or depression might have lower threshold for experiencing somatic symptoms in general, leading to a higher degree of symptom reporting in these conditions (24). Anxiety and depression may also be secondary phenomena to somatic symptoms. People with somatic symptoms in general will often worry about their complaints, their cause and nature, and experience symptoms of anxiety. Likewise, being bothered by somatic symptoms over time can lead to feeling of hopelessness and helplessness and consequently to depressive symptoms (25). Patients with severe major depression have more somatic symptoms than patients with mild depressive episodes (p < .05), indicating that the level of depression is closely linked to the reporting of somatic symptoms (26).
Females reported significantly more somatic symptoms than men did. This is in accordance with earlier studies, both from general practice (27) and from the community (3), where nearly all somatic symptoms in general were more common in women than in men. In our study, only heartburn, movement disability, angina, and impaired hearing were more common in men than in women. Several theories have been proposed to try to explain the increased somatic symptom report in women (27). One difference may be based on the higher prevalence of anxiety disorder and depression in women, which is strongly related to symptom reporting (6,28). Other factors may be a history of sexual or physical abuse (29–31), gender differences in social roles and responsibilities, cultural factors permitting greater expressiveness in women, lower threshold for seeking health care, and amplification of somatic symptoms (27,32). In our study, even though women reported more somatic symptoms than men did, the association between somatic symptoms and anxiety and depression was equally strong in men and women, indicating that difference in prevalence of these conditions could not explain the difference in somatic symptoms reported. Additionally, differences in men and women in their thresholds for seeking health care could not explain the gender differences in symptom reporting in our community sample.
Our findings are in accordance with findings in two earlier population-based surveys (3,33), where most somatic symptoms in general were more prevalent in women. Thus, the results from community samples are consistent and suggest that increased somatic symptoms reporting in women is not merely an artifact of higher health care utilization.
The prevalence of somatic symptoms linearly increased with age. The prevalence of anxiety disorder and depression in the population had also a tendency to increase with age, but to a lesser degree than the prevalence of somatic symptoms (Figure 4). However, the association among anxiety, depression, and somatic symptoms was nearly identical for all age groups. Earlier studies of the relationship among somatic symptoms, anxiety disorder, depression, and age have shown conflicting results. One study (34) reported that somatic complaints in the course of major depression tended to decrease with age, whereas another study (32) found that somatization in the course of depression was particularly important in older compared to younger patients. Our findings of an increasing prevalence of somatic symptoms with age, but no increase in the relationship between psychopathology and somatic symptoms with age, may indicate that older people have more physical complaints in general without medical explanation.
The definition of "somatization syndrome" as an entity with 12 or more somatic symptoms and the concept of "abridged somatization" with a symptom threshold of 5 indicate that it is possible to define a threshold for number of somatic symptoms where psychological problems should be of special importance. In our study, there was a close association among case-level anxiety depression, comorbid anxiety, and depression and both "somatization syndrome" and "abridged somatization." Case-level anxiety and depression were equally strongly associated, whereas the comorbid condition was even more strongly associated. However, we found a linear association between number of somatic symptoms and the score on HADS-T. This might indicate that rather than define categorical concepts like "somatization syndrome," "abridged somatization," and "somatization disorder" as illnesses per se, one should rather focus on a dimensional concept closely associated with anxiety and depression. This is in accordance with findings by Katon et al. (24), who suggested that somatization should be viewed as a continuum on which increasing levels of somatic symptoms are indicative of increasing distress and disability associated with anxiety disorders and depressions.
It has been claimed that in true "somatization disorder," the patients have minimal or no symptoms of anxiety or depression and that these patients are "alexithymic" with low ability to identify and express emotions (35). In our study, 2% of the males and 4.6% of the females had "somatization syndrome" (reporting 12 or more somatic symptoms), with low scores on HADS-A and HADS-D. A proportion of these patients in our population may be "true somatizers" who express their feeling of distress mainly with somatic symptoms. Earlier studies have reported a prevalence of somatization disorder (Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV)) of about 1% in the general population (36). According to the DSM-IV on somatization disorders, the symptoms must be localized to several different organ systems over a period of several years (37). These are much more strictly defined criteria than the criteria for "somatization syndrome" in our study, and this may explain why we find a higher prevalence rate .
Our study has some limitations: the symptoms were reported from a survey asking about specific symptoms. This may lead to an overendorsement bias, a tendency for patients to claim the symptoms asked about in the questionnaire, leading to overreporting of symptoms. In addition, the patients were not asked to grade the severity of their symptoms, so even mild, temporary symptoms with no clinical significance may have been reported. However, the patients were asked about symptoms during the last year, and it is likely that symptoms remembered over such a long period might have been of a certain severity or cause of functional impairment. The prevalence of somatic symptoms fit well with results from earlier studies (3). Furthermore, excluding such mild symptoms from the analyses would probably not have changed our conclusion about the close association among anxiety, depression, and somatic symptoms, because patients with anxiety and depression have a tendency to report more severe somatic symptoms than patients without these conditions (8). We have no information about clinical investigation or lab tests in the people who were included into the study. It might be possible that some of these people had somatic disorders after all. From other studies, it is known that at least one third of somatic symptoms reported lack an organic explanation (6). However, even if we had excluded a higher proportion of patients due to organic diseases, the main findings of the study of the close connection among anxiety disorders, depression, and functional somatic symptoms would not have been changed. In our study, only one sixth of patients reported an organic disorder. The reason for this might be that our study was a population-based study compared to studies conducted in primary care.
As to strengths, we believe that the use of HADS as a measure of psychopathology, was particularly suitable in this study of somatization, because there are no items in HADS about somatic manifestations of anxiety and depression. Our study is one of the largest studies of the association among anxiety, depression, and somatic symptoms, covering more than 50,000 patients. The ECA Study covered about 13,000 patients (3), whereas studies from primary care have included 1000 to 1500 patients (3,15). Our study covered the whole adult population in one Norwegian region, and the data have been weighted to fit with the general population of Norway.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication September 30, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Han, C.-U. Pae, A. A. Patkar, P. S. Masand, K. W. Kim, S.-H. Joe, and I.-K. Jung Psychometric Properties of the Patient Health Questionnaire-15 (PHQ-15) for Measuring the Somatic Symptoms of Psychiatric Outpatients Psychosomatics, November 1, 2009; 50(6): 580 - 585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Mullen and D. B. Menkes Psychiatriform Disorders: Psychiatric Analogues of Somatoform Disorders International Journal of Social Psychiatry, September 1, 2008; 54(5): 395 - 401. [Abstract] [PDF]
|
|
|
|
 |
|
 T. Rutledge, S. E. Reis, M. B. Olson, J. Owens, S. F. Kelsey, C. J. Pepine, S. Mankad, W. J. Rogers, C. N. B. Merz, G. Sopko, et al. Depression Symptom Severity and Reported Treatment History in the Prediction of Cardiac Risk in Women With Suspected Myocardial Ischemia: The NHLBI-Sponsored WISE Study. Arch Gen Psychiatry, August 1, 2006; 63(8): 874 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rutledge, S. E. Reis, M. Olson, J. Owens, S. F. Kelsey, C. J. Pepine, S. Mankad, W. J. Rogers, G. Sopko, C. E. Cornell, et al. Depression is associated with cardiac symptoms, mortality risk, and hospitalization among women with suspected coronary disease: the NHLBI-sponsored WISE study. Psychosom Med, March 1, 2006; 68(2): 217 - 223. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
