The Tohoku Journal of Experimental Medicine
Online ISSN : 1349-3329
Print ISSN : 0040-8727
ISSN-L : 0040-8727
Regular Contributions
A Diagnosis of Depression Should Be Considered in Patients with Multiple Physical Symptoms in Primary Care Clinics
Kyoko YamamotoSeiji ShiotaShigeki OhnoAkiko KurodaAoi YoshiiwaKoh AbeKazunari MurakamiToshio Fujioka
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML

2013 Volume 229 Issue 4 Pages 279-285

Details
Abstract

Although the vast majority of depressed patients visit primary health care clinics, they often remain undiagnosed and untreated. Therefore, early detection in primary care settings is important. There is a high correlation between number of physical symptoms and the presence of depression, yet little has been reported regarding this relationship in Japanese primary care clinics. We examined number of physical symptoms and depression in a department of general medicine of a Japanese hospital. We included patients with unexplained symptoms after multiple tests to rule out organic diseases. Twenty-one common symptoms were assessed using a symptom checklist. Depression was diagnosed using the Patient Health Questionnaire-9, a self-administered questionnaire designed to diagnose depression. Among 386 patients, 105 (27.2%) (average age: 49.7 ± 20.9 years, 28 men and 77 women) met the criteria for depression. Among the 21 symptoms, 14 were significantly more frequent in patients with depression than in those without depression. When patients had neither general fatigue, nor sleep disturbance nor appetite loss, none met the criteria for depression. Number of symptoms was significantly higher in patients with compared with those without depression. The prevalence of depression increased with number of symptoms: 2% (2/100) for 0 or 1 symptom, 42.4% (42/99) for four to five symptoms and 68.7% (22/32) for more than nine symptoms. Japanese primary care physicians can often rule out depression when patients have neither general fatigue, nor sleep disturbance nor appetite loss. A diagnosis of depression should be considered in patients who report multiple physical symptoms.

Introduction

Although depression and anxiety are common in primary care settings (Brown and Schulberg 1998; Valenstein et al. 2001), these disorders remain undiagnosed and untreated at least half the time (Schulberg and Burns 1988; Zich et al. 1990; Ohtsuki et al. 2010). In all parts of the world, the vast majority of depressed patients are reported to have visited a primary health care clinic (Patel 2001). Therefore, early detection in primary care is important for decreasing the severity of patient depression (Greden 2003).

Evidence indicates that there is a high correlation between physical symptoms, pain and depression (Greden 2003). Thirty percent of patients with depression experience physical symptoms for more than 5 years before receiving the proper diagnosis (Lesse 1983). These individuals often present to their primary care physician with atypical symptoms either because they are too ashamed to discuss psychological problems or because subjective somatic symptoms are the main reason for their consultation (Wittkampf et al. 2007). Indeed, physical rather than emotional symptoms are the predominant complaint in patients with psychiatric disorders who seek care in primary care settings (Mathew et al. 1981; Katon et al. 1982; Bridges and Goldberg 1985; Katon et al. 1991; Simon and VonKorff 1991). Various symptoms such as fatigue, headache, back pain and constipation have been associated with depression (Bair et al. 2003; Sugahara et al. 2004; Nakao and Yano 2006). Moreover, there is a high correlation between number of symptoms and the presence of depression (Kroenke et al. 1994).

In Japan, Sugahara et al. (2004) found that several symptoms, including general fatigue, appetite loss, headache and diarrhea, were associated with depression. However, that study was conducted in a psychiatric clinic, not a primary care setting. Other studies conducted during annual health examinations have revealed that physical symptoms can be useful for the screening of depression (Isshiki et al. 2004; Nakao and Yano 2006). However, little has been reported on the association between depression and specific symptoms in individuals visiting primary care settings in Japan. In addition, it remains unclear whether the number of symptoms is correlated with the presence of depression in these individuals. In this study, we aimed to examine how the type and number of symptoms are associated with depression in a Japanese primary care setting.

Materials and Methods

Subjects and assessment

We enrolled patients who visited the outpatient clinic in the Department of General Medicine at Oita University Hospital during January 2008 to December 2010. The hospital is located approximately 20 km from an urban area. The department is in a medical division that usually provides treatment for patients who visit the hospital without a referral form. The chief complaints of patients visiting this department are abdominal pain, back pain and fever. Approximately 245,000 patients visit this hospital per year, of whom 10,000 visit this department. We included patients with unexplained symptoms after multiple tests, including blood tests, to rule out organic diseases. Most of them agreed to enter this study. The protocol was approved by the ethics committee of Oita University.

Physical symptoms were examined using a checklist that accounts for over 90% of symptoms reported in the outpatient setting, according to previous reports (Kroenke et al. 1988; Schappert 1992; Kroenke et al. 1993). Twenty-one common symptoms were selected for analysis: fever, headache, vomiting, cough, sputum production, nasal discharge, throat pain, chest pain, shortness of breath, palpitations, abdominal pain, diarrhea, heartburn, back pain, joint pain, dysesthesia, dizziness/vertigo, weight loss, appetite loss, general fatigue and sleep disturbance. Presence of these symptoms was determined by the following question: “During the past two weeks, have you often been bothered by 1) abdominal pain? 2) back pain? ...” For each symptom, patients simply checked “yes” or “no.”

We used the Patient Health Questionnaire-9 (PHQ-9), a self-administered questionnaire developed as a diagnostic tool for depression (Spitzer et al. 1999; Gilbody et al. 2007; Wittkampf et al. 2007). It includes the nine criteria for major depression in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; DSM-IV (American Psychiatric Association 1994). The PHQ-9 has been translated into Japanese, and studies of the validity of the Japanese version have been published (Muramatsu et al. 2007). We used a categorical algorithm of the PHQ-9. The two core symptoms were “depressive mood” and “loss of interest or pleasure.” When at least five out of the nine criteria for major depressive episode were reported, the patient was diagnosed as having depression.

Statistical analysis

The univariate association between each group was quantified using the unpaired t-test, Mann-Whitney U-test, Fisher’s exact test and chi-square test. A multivariate logistic regression model was used to calculate the odds ratio (OR) of depression to adjust by age, gender and specific symptoms. All determinants with P-values < 0.10 were entered together in the full logistic regression model, and the model was reduced by excluding variables with P-values > 0.10. The OR and 95% confidence interval (CI) were used to estimate risk. A P-value less than 0.05 was accepted as statistically significant. The SPSS statistical software package version 19.0 (SPSS, Inc., Chicago, IL) was used for all statistical analyses.

Results

A total of 386 outpatients (average age: 51.3 ± 19.4 years, 114 men and 272 women) were enrolled in this study. Among 386 of them, 105 (27.2%) (average age: 49.7 ± 20.9 years, 28 men and 77 women) met the criteria for depression. There was no difference in average age or gender between depressed and non-depressed patients (Table 1). In those with depression, general fatigue (96.2%) was the most frequently reported symptom followed by sleep disturbance (81.9%), appetite loss (73.3%) and weight loss (40.0%). The prevalence of these symptoms was significantly higher in patients with depression than in those without depression (P < 0.001). In addition, reports of headache, chest pain, shortness of breath, palpitations, heartburn and dizziness/vertigo were significantly more frequent in patients with depression (all P < 0.001). Vomiting, back pain, dysesthesia and fever were also significantly associated with depression (all P < 0.01). Importantly, among 108 patients without general fatigue, sleep disturbance and appetite loss, none met the criteria for depression.

Multivariate analysis after adjustment by age and gender showed that general fatigue, appetite loss, sleep disturbance and headache were independently associated with depression. Among them, general fatigue showed the highest OR (OR = 13.90, 95% CI = 4.72-40.90), followed by appetite loss (OR = 4.68, 95% CI = 2.54-8.62), sleep disturbance (OR = 3.12, 95% CI = 1.61-6.02) and headache (OR = 1.99, 95% CI = 1.01-3.92).

Next, we examined symptoms related to depression according to gender (Tables 2 and 3). Among 114 male patients, 28 (24.5%) were diagnosed with depression. There was no difference in average age between males with and without depression. General fatigue was most frequently found in those with depression (92.9%) and was significantly associated with depression (P < 0.001) (Table 2). Appetite loss, weight loss, sleep disturbance, headache and dysesthesia were also significantly associated with depression (all P < 0.01). Of the female patients, 77 (28.3%) were diagnosed with depression. There was no significant difference in age between females with and without depression. Consistent with the case in men, general fatigue, appetite loss, weight loss and sleep disturbance were significantly associated with depression (all P < 0.001) (Table 3). In addition, females with depression more often reported headache, nausea, nasal discharge, shortness of breath, palpitations, heartburn, back pain and dizziness/vertigo than females without depression.

Overall, the number of symptoms was not different between males and females (3.7 ± 2.9 vs. 3.8 ± 3.1, respectively, P = 0.97). In patients without depression, number of symptoms was not related to gender (3.0 ± 2.7 in males vs. 2.8 ± 2.5 in females, P = 0.61). Furthermore, the number of symptoms in patients with depression did not differ between males and females (5.8 ± 2.3 vs. 6.4 ± 3.0, respectively, P = 0.56). Number of symptoms was significantly higher in patients with depression than in those without depression (6.2 ± 2.8 vs. 2.9 ± 2.5, respectively, P < 0.001). This gap was statistically significant in the case of female patients (6.4 ± 3.0 vs. 2.8 ± 2.5, respectively, P < 0.001). Even in males, the number of symptoms was significantly higher in patients with depression than in those without depression (5.8 ± 2.3 vs. 3.0 ± 2.7, respectively, P < 0.001).

Finally, we evaluated the relationship between the number of symptoms and the prevalence of depression (Fig. 1). In the case of 0 or 1 symptom, depression was found in only 2 out of 100 patients (2%). The prevalence of depression increased with number of symptoms. The rate was 42.4% (42/99) when patients had four to five symptoms, and 49.1% (28/57) for six to eight symptoms. Surprisingly, in patients with more than nine symptoms, depression was found in 68.7% (22/32).

Table 1. Comparison of Patients With and Without Depression.
Depression Non-depression P value
n 105 281
Age (years old) 49.7 ± 20.9 51.9 ± 18.8 0.38
Male 28 26.7% 86 30.6% 0.45
Fever 20 19.0% 30 10.7% 0.02
Headache 39 37.1% 45 16.0% < 0.0001
Vomiting 17 16.2% 22 7.8% 0.01
Cough 8 7.6% 23 8.2% 0.85
Sputum 14 13.3% 29 10.3% 0.40
Nasal discharge 17 16.2% 17 6.0% 0.002
Pain of throat 13 12.4% 19 6.8% 0.07
Chest pain 19 18.1% 36 12.8% 0.18
Shortness of breath 24 22.9% 23 8.2% < 0.0001
Palpitation 25 23.8% 28 10.0% < 0.0001
Abdominal pain 16 15.2% 30 10.7% 0.21
Diarrhea 14 13.3% 23 8.2% 0.12
Heart burn 22 21.0% 15 5.3% < 0.0001
Back pain 26 24.8% 41 14.6% 0.01
Joint pain 24 22.9% 45 16.0% 0.11
Dysesthesia 28 26.7% 44 15.7% 0.01
Dizziness/vertigo 28 26.7% 33 11.7% < 0.0001
Sleep disturbance 86 81.9% 105 37.4% < 0.0001
General fatigue 101 96.2% 119 42.3% < 0.0001
Appetite loss 77 73.3% 64 22.8% < 0.0001
Weight loss 42 40.0% 38 13.5% < 0.0001
Table 2. Comparison of Male Patients With and Without Depression.
Depression Non-depression P value
n 28 86
Age (years old) 47.5 ± 18.6 51.0 ± 17.5 0.30
Fever 4 14.3% 5 5.8% 0.14
Headache 11 39.3% 14 16.3% 0.01
Vomiting 4 14.3% 9 10.5% 0.40
Cough 3 10.7% 9 10.5% 0.60
Sputum 7 25.0% 11 12.8% 0.11
Nasal discharge 5 17.9% 6 7.0% 0.09
Pain of throat 3 10.7% 7 8.1% 0.46
Chest pain 3 10.7% 13 15.1% 0.40
Shortness of breath 6 21.4% 9 10.5% 0.12
Palpitation 6 21.4% 10 11.6% 0.16
Abdominal pain 2 7.1% 8 9.3% 0.53
Diarrhea 5 17.9% 8 9.3% 0.18
Heart burn 5 17.9% 6 7.0% 0.09
Back pain 4 14.3% 13 15.1% 0.59
Joint pain 5 17.9% 14 16.3% 0.52
Dysesthesia 10 35.7% 12 14.0% 0.01
Dizziness/vertigo 8 28.6% 13 15.1% 0.11
Sleep disturbance 20 71.4% 34 39.5% 0.003
General fatigue 26 92.9% 32 37.2% < 0.0001
Appetite loss 16 57.1% 20 23.3% 0.001
Weight loss 12 42.9% 13 15.1% 0.002
Table 3. Comparison of Female Patients With and Without Depression.
Depression Non-depression P value
N 77 195
Age (years old) 50.5 ± 21.7 52.3 ± 19.4 0.58
Fever 16 20.8% 25 12.8% 0.09
Headache 28 36.4% 31 15.9% < 0.0001
Vomiting 13 16.9% 13 6.7% 0.01
Cough 5 6.5% 14 7.2% 0.84
Sputum 7 9.1% 18 9.2% 0.97
Nasal discharge 12 15.6% 11 5.6% 0.008
Pain of throat 10 13.0% 12 6.2% 0.06
Chest pain 16 20.8% 23 11.8% 0.05
Shortness of breath 18 23.4% 14 7.2% < 0.0001
Palpitation 19 24.7% 18 9.2% 0.001
Abdominal pain 14 18.2% 22 11.3% 0.13
Diarrhea 9 11.7% 15 7.7% 0.29
Heart burn 17 22.1% 9 4.6% < 0.0001
Back pain 22 28.6% 28 14.4% 0.006
Joint pain 19 24.7% 31 15.9% 0.09
Dysesthesia 18 23.4% 32 16.4% 0.18
Dizziness/vertigo 20 26.0% 20 10.3% 0.001
Sleep disturbance 66 85.7% 71 36.4% < 0.0001
General fatigue 75 97.4% 87 44.6% < 0.0001
Appetite loss 61 79.2% 44 22.6% < 0.0001
Weight loss 30 39.0% 25 12.8% < 0.0001
Fig. 1.

Prevalence of depression according to number of symptoms.

In the case of 0 or 1 symptom, depression was found in only 2 out of 100 patients (2%). In patients with 2-3 symptoms, 11.2% (11/98) had depression. The rate was 42.4% (42/99) when patients had four to five symptoms. In patients with more than nine symptoms, the depression rate was 68.7% (22/32).

Discussion

Although the vast majority of depressed patients visit primary health care clinics, they often remain undiagnosed and untreated. Therefore, early detection in primary care settings is important. 77% of Japanese individuals with depression have reported somatic symptoms as the reason for visiting a physician (Simon et al. 1999). In our study, we found several symptoms associated with the presence of depression in both genders, including in particular, general fatigue, sleep disturbance, loss of appetite and weight loss. These findings are reasonable because fatigue, sleep disturbance and loss of appetite are part of the DSM-IV criteria for major depression (American Psychiatric Association 1994). Importantly, among 108 patients without general fatigue, sleep disturbance and appetite loss, none met the criteria for depression. This suggests that when patients have none of these three symptoms, primary physicians can often rule out depression. Additionally, headache, chest pain, shortness of breath, palpitations, heartburn, dizziness/vertigo, vomiting, back pain, dysesthesia and fever were more frequent in patients with depression. Furthermore, number of symptoms was positively correlated with rate of depression. Findings such as ours can help physicians diagnose depression in primary care settings in Japan. Physicians can more easily identify patients with depression at an earlier stage with an assessment of physical symptoms and number of such symptoms.

Consistent with previous studies, headache, chest pain and back pain were more frequent in patients with depression than those without depression. Previous studies have reported that more than 50% of patients with depression experience comorbid pain, including headache, neck and back pain, chest pain and other nonspecific generalized pain (Bair et al. 2003; Katona et al. 2005). A common pathogenesis may be associated with the development of pain and depression. Serotonin and norepinephrine likely share neurochemical mechanisms that tie physical symptoms and depression (Greden 2003). Through the noradrenergic and serotonergic systems, pain and depression may share a common biological pathway. However, further studies are needed to clarify the mechanisms of pain and depression.

In addition to pain, prevalence of dysesthesia is reportedly greater in male patients with depression. Nutritional problems such as hypomagnesemia (Berkelhammer and Bear 1985) and cobalamin deficiency (Lindenbaum et al. 1988) due to an intake imbalance caused by anorexia or bulimia are possible contributing biological factors, although these conditions are more common in females (Han et al. 1997). In future studies, it is important to examine the level of these factors in male patients with depression.

The total number of symptoms is a particularly powerful correlate of depression (Kroenke et al. 1994). In one study, only 2% of patients with no physical symptoms or only one such symptom were determined to have a mood disorder, whereas the percentage jumped to 60% when patients had nine or more physical symptoms (Kroenke et al. 1994). Consistent with that research, our study showed that the prevalence of depression increased with the number of symptoms. When patients had more than four symptoms, the depression rate was 42.4%. Moreover, in patients with more than nine symptoms, the rate of depression was 68.7%. These findings suggest that physicians should consider depression if patients complain of a number of symptoms, irrespective of the individual. Nakao and Yano (2003) reported that individuals’ number of symptoms was positively associated with the prevalence of depression at annual health checkups, which was consistent with our findings. For example, among 27 subjects with more than four symptoms, depression was found in nine of them. However, that study comprised a relatively small number of subjects.

Several studies have reported that women have more physical symptoms potentially related to emotional distress (e.g., headache, back pain) than men (Kirmayer et al. 1993; Ohayon and Schatzberg 2003; Breivik et al. 2006). Different biological or psychological factors such as gender roles or sex-specific coping styles between males and females might contribute to the different outcomes. In the present study, the number of physical symptoms in patients with depression was slightly higher for females than for males. This difference did not reach statistical significance, perhaps because of the small number of male patients with depression. For females, many symptoms were reported at higher levels in patients with depression than in those without depression. In addition, for females, there was a large difference in number of symptoms between those with and those without depression. This suggests that number of symptoms can be a useful marker for depression, especially in females. On the other hand, a recent study showed that the gender difference in somatic symptoms reported is very small (Shimodera et al. 2012). It might be advisable to consider depression in cases where patients have more than four symptoms, without distinction between genders.

In our study, some patients were not diagnosed with depression despite many physical symptoms. Nakao and Yano (2006) reported that somatic symptoms (low back pain, dizziness and abdominal pain) at baseline were significant risk factors for major depression in the following year. Therefore, even though some patients do not meet the criteria for depression, those with unexplained back pain, dizziness and abdominal pain might be considered at high risk for depression in the future. In addition, these symptoms can be part of a somatoform disorder. Somatization can increase use and cost of medical resources, independent of psychiatric and medical comorbidity (Barsky et al. 2005). Furthermore, somatizing patients with comorbid anxiety or depressive disorder generally use such resources more than patients with anxiety and/or depressive disorder alone (unaccompanied by somatization) (Barsky et al. 2005). Somatization is more closely associated with outpatient visits than are depression and anxiety (Kroenke et al. 2002). Thus, it is necessary to follow individuals with unexplained symptoms, even if they do not show any organic abnormality.

Our study has several limitations. First, the prevalence of depression was high compared with previous reports from Japan (Yamazaki et al. 2005; Nakao and Yano 2006; Ohtsuki et al. 2010). This might be partly attributable to different methodologies. Previous studies were mainly conducted during annual health examinations or by random sampling of primarily healthy individuals. However, we included patients with unexplained symptoms even after multiple tests, including blood tests, ruled out organic diseases. Patients at our department may have had more psychiatric and medical comorbidity than those in the general population. Thus, selection bias could have contributed to the high prevalence of depression in our study. In addition, we used the PHQ-9 to assess the presence of depression. Although this instrument comprises the nine diagnostic criteria for depression in the DSM-IV, it detects patients with a depressive episode but not depressive disorder. The DSM-IV exclusion criteria for a depressive disorder are not included in the PHQ-9. For example, if a depressive episode is part of a bipolar disorder, the patient is diagnosed as bipolar. Therefore, other mood disorders such as bipolar disorders and dysthymia could have been present in our “depression” patients. Studies using psychiatrist-diagnosed depression in addition to the PHQ-9 are necessary to confirm our findings.

In conclusions, our study showed an association between specific symptoms and depression in both genders in a primary care setting. General fatigue, sleep disturbance, loss of appetite and weight loss were particularly associated with depression in both genders. When patients do not exhibit general fatigue, sleep disturbance, or loss of appetite, primary physicians can often rule out the presence of depression. Furthermore, the number of physical symptoms was significantly associated with depression irrespective of gender. Japanese primary care physicians should be encouraged to consider a diagnosis of depression when they encounter patients who present with multiple physical symptoms that cannot be explained in terms of organic etiology. It is important that these physicians improve their ability to recognize depression in their patients.

Acknowledgements

We thank Dr. Haruka Higuma and Dr. Shizuko Okamoto for collecting data and Ms. Megumi Abe for excellent technical assistance.

We do not have any financial support for this study.

Conflict of Interest

The authors declare no competing interest.

References
 
© 2013 Tohoku University Medical Press
feedback
Top