Secondary Logo

Journal Logo

Original Articles

Association between gastrointestinal symptoms and depression among older adults in Taiwan: A cross-sectional study

Huang, Mao-Hsuana,b; Wang, Yen-Pob–e; Wu, Po-Shanc,f; Chan, Yee-Lam E.g; Cheng, Chih-Mingb,c,h; Yang, Cheng-Hungb,h; Tsai, Shih-Jenb,h; Lu, Ching-Liangb–e; Tsai, Chia-Fenb,c,h,*

Author Information
Journal of the Chinese Medical Association: March 2021 - Volume 84 - Issue 3 - p 331-335
doi: 10.1097/JCMA.0000000000000460



Huang, Mao-Hsuan; Wang, Yen-Po; Wu, Po-Shan; Chan, Yee-Lam E; Cheng, Chih-Ming; Yang, Cheng-Hung; Tsai, Shih-Jen et al. Association between gastrointestinal symptoms and depression among older adults in Taiwan: A cross-sectional study . J Chin Med Assoc. 2021;84:3,331-335. doi: 10.1097/JCMA.0000000000000460.

In the article “Association between gastrointestinal symptoms and depression among older adults in Taiwan: A cross-sectional study” (2021; 84:3). The first author’s affiliation was incorrect as “Department of Psychiatry, Suao and Yuanshan Branches of Taipei Veterans General Hospital, Hualien, Taiwan, ROC”. The correct affiliation is “Department of Psychiatry, Taipei Veterans General Hospital, Yuanshan and Suao Branch, Yilan, Taiwan, ROC”. This has been updated in the online version of the article.

Journal of the Chinese Medical Association. 84(5):555, May 2021.


Late-life depression (LLD) is a major psychiatric disorder in older adults, with prevalence rates ranging from 7% to 49%.1 Older adults with depression may have more somatic and gastrointestinal (GI) symptoms than either older adults without depression2 or younger adults with major depressive disorder (MDD).3 Several studies have identified some generalized physical complaints as predictors of psychiatric comorbidity.4 However, whether GI symptoms are predictive of elevated rates of depression among older adults remains unclear. Furthermore, few studies on older adults with depression have investigated GI symptoms using GI-specific symptom questionnaire scales, such as the Gastrointestinal Symptom Rating Scale (GSRS). In other fields of research, the bidirectional route of communication between the gut and central nervous system, referred to as the gut-brain axis, has been of interest in recent years.5 Psychological factors, such as depression, anxiety, and the occurrence of traumatic life events, have been reported to be closely related to the causal mechanisms underlying functional GI disorders.6 Dysbiosis and inflammation of the gut have also been identified as causes of several mental illnesses, including anxiety and depression.7 Studies investigating the relationship between depression and GI symptoms have not controlled for crucial variables, such as diet, that may influence both depression and functional GI disorders.8 Furthermore, relatively little research on this topic has focused on older adults.

Thus, this study investigated the correlation between specific GI symptoms and depressive symptoms in older adults after diet is controlled for. Specifically, we examined the influence of a Mediterranean diet pattern (MDP) on the risk of depression in older adults. We hypothesized that older adults with certain GI symptoms have a higher risk of having more depressive symptoms.


2.1. Participants

Our study was conducted at the psychiatric outpatient department of Taipei Veterans General Hospital. We recruited patients between the ages of 60 and 89 years who had a Diagnostic and Statistical Manual of Mental Disorders, 5th Edition diagnosis of MDD made by a board-certificated psychiatrist9 into the LLD group. Participants in the control group were recruited using posters placed in the community surrounding the hospital. The exclusion criteria were having (1) a diagnosis of a major neurocognitive disorder; (2) any one major psychiatric comorbidity (such as schizophrenia or bipolar disorder); (3) any one physical comorbidity (such as organic GI diseases—including liver cirrhosis, fatty liver disease, peptic ulcer, inflammatory bowel disease, or any malignancy—or a history of receiving chemotherapy or radiation for any cancer); (4) any known active bacterial, fungal, or viral infection; (5) received new prebiotic or probiotics within 90 days prior to enrollment; and (6) a record of GI tract surgery, appendectomy, or cholecystectomy surgery in the preceding 1 year. The participants in the control group underwent the Mini-International Neuropsychiatric Interview with a psychiatrist to exclude patients with any psychiatric illness. We conducted a detailed review of the participants’ medical history and performed anthropometric measurements to exclude those with physical comorbidities. The study was conducted in accordance with the Declaration of Helsinki and was approved by the local ethics review committee. All participants provided written informed consent prior to participating in the study.

2.2. Assessment of depressive and GI symptoms

All participants completed the 17-item Hamilton Depression Rating Scale (HAMD)10 and 15-item GSRS.11 The 17-item HAMD is a commonly used and widely validated measure of depressive symptoms, where ratings are given by clinicians. The GSRS comprises sections on the five symptom clusters of reflux syndrome, diarrhea, constipation, abdominal pain, and dyspepsia, where ratings are given by clinicians. The GSRS uses a 7-point scale with 1 representing “no bothersome symptoms” and 7 representing “very bothersome symptoms.” The GSRS has been reported to be valid and reliable when used to measure GI symptoms, such as gastroesophageal reflux disease and dyspepsia.12

2.3. Assessment of food consumption

Food consumption was assessed with the assistance of a trained dietitian using a semiquantitative food frequency questionnaire with documented validity.13 Participants were asked how often they consumed various types of food and beverages that fall under one of nine categories during the previous 1 month. We evaluate the adherence of participants to the MDP according to a 9-point dietary score, which was previously reported by Trichopoulou et al.14 The Mediterranean diet score (MedDietScore) was calculated by assigning a score of 0 or 1 to each of the nine food groups; scores of 0 and 1 indicated that the participant’s consumption patterns in the given food category did not and did accord with the MDP, respectively. Specifically, for each food category, 1 point was awarded for food intake with a (1) high ratio of monounsaturated fatty acids to saturated fatty acids; (2) high intake of legumes, at one or more servings per week; (3) high intake of whole grains, at one or more servings per day; (4) high intake of nuts, at one or more servings per week; (5) high intake of fruits, at three or more servings per day; (6) high intake of vegetables, at four or more servings per day; (7) high intake of fish, at four or more servings per week; (8) low intake of red and processed meat, at less than two and three servings per day for women and men, respectively; and (9) moderate intake of alcohol, at 0.5 to 1 drink daily for women and 1 to 2 drinks daily for men. The total score ranged from 0 (minimal adoption) to 9 (maximum adoption).

2.4. Other covariates assessments

For the baseline assessment, we gathered sociodemographic data (eg, sex, age, and education) and anthropometric data (eg, weight and height). The weight and height of the participants were measured by an assisting nurse, and body mass index (BMI) was calculated, defined as weight (in kilograms) divided by squared height (in meters).

2.5. Statistical analyses

Categorical variables were compared using chi-square tests, and continuous variables were compared using Student’s t tests. All tests were based on two-tailed alternatives. Analysis of variance with post hoc analysis was used to compare HAMD scores and GSRS scores between the four groups of patients with LLD who were taking different types of antidepressant drugs. Linear regression was used to assess the correlation between total HAMD score and GSRS score, with adjustment for age, sex, years of education, BMI, and MedDietScore. Sex-stratified subgroup analyses were also conducted to investigate the role of sex in the association between HAMD score and GSRS score among patients with LLD. The HAMD-17 has two items that are related to appetite and GI (items 11 and 12), which may overlap with the psychometric construct of the GSRS. Thus, to prevent any consequently spurious inferences as to the predictive power of GI symptoms, we summed all subscores of HAMD except item 11 and item 12 in the linear regression model. Before the analysis, we tested for multicollinearity using the variance inflation factor. We also performed logistic regression with adjustment for demographic data (age, sex, years of education, BMI, and diabetes mellitus) to calculate the odds ratios (ORs) of developing LLD. A subgroup analysis using logistic regression models for subscales of GSRS was performed to assess the risks of developing LLD. Furthermore, logistic regression analyses for subgroups stratified by depression severity were also conducted to explore the effects of disease severity on the risk of LLD. With reference to a large study that established severity cutoff scores on the HAMD,15 we classified patients into mild depression (8-16), moderate depression (17-23), and severe depression (≥24) groups according to the ranges of HAMD. For all variables, significance was defined as a two-tailed p value of <0.05. All data processing and statistical analyses were performed using Statistical Package for Social Science software version 17 and Statistical Analysis Software (version 9.1; SAS Institute, Cary, NC, USA).


We enrolled 69 patients with LLD and 37 controls. The patients with LLD had a lower average educational level compared with the controls and reported higher levels of depressive symptoms and GI symptoms (Table 1). Specific GI symptoms that were more prevalent in patients with LLD were reflux syndrome, abdominal pain, and dyspepsia (all p < 0.01) (Table 1). MedDietScore did not significantly differ between patients with LLD and the controls. Among the 69 patients with LLD, 14 (20.3%) did not take antidepressants and 55 (79.7%) took antidepressants. Among these 55 patients taking antidepressants, 18 (26.1%) took selective serotonin reuptake inhibitors, 14 (20.3%) took serotonin-norepinephrine reuptake inhibitors, and 23 (33.3%) took agomelatine. To investigate the influence of the various medications on GI symptoms, analysis of variance tests were performed, and no significant difference in any of the symptom clusters of GSRS was noted among patients taking medications in the different groups (Table 2). Linear regression results indicated a modest correlation between GSRS total score and HAMD total score (β: 0.47, p < 0.01) (Table 3). The subscales were also included in the analyses, and we found that reflux syndrome (β: 1.47, 95% CI: 0.46-2.48), abdominal pain (β: 1.98, 95% CI: 0.41-3.55), and dyspepsia (β: 1.02, 95% CI: 0.36-1.68) were significantly correlated with HAMD total score without somatic subscores (Table 3). Sex-stratified subgroup analyses further indicated that the correlation between HAMD score and GSRS score was only present among female patients with LLD (Table 4).

Table 1 - Clinical and demographic characteristics of the patients with late-life depression and the controls
Mean (SD) or n (%) LLD (n = 69) Control (n = 37) p
Age 72.4 (8.3) 67.5 (5.5) <0.01
Sex (female), n (%) 53 (76.8) 22 (59.5) 0.05
Edu (y) 9.0 (5.1) 12.3 (4.5) <0.01
BMI (kg/m2) 23.8 (3.0) 24.0 (3.4) 0.80
DM (yes), n (%) 14 (20.3) 5 (13.5) 0.28
MedDietScore 1.8 (1.0) 1.8 (1.3) 0.97
HAMD total score 14.4 (6.4) 1.8 (1.8) <0.01
GSRS total score 6.3 (4.1) 3.5 (3.3) <0.01
 GSRS: Reflux syndrome 1.2 (1.4) 0.5 (0.9) <0.01
 GSRS: Diarrhea syndrome 1.1 (1.1) 1.0 (1.1) 0.61
 GSRS: Constipation syndrome 1.4 (1.5) 0.8 (1.0) 0.02
 GSRS: Abdominal pain 0.7 (1.0) 0.3 (0.5) <0.01
 GSRS: Dyspepsia 2.1 (2.0) 1.0 (1.4) <0.01
BMI = body mass index; DM = diabetes mellitus; GSRS = Gastrointestinal Symptom Rating Scale; HAMD = Hamilton Depression Rating Scale; LLD = late-life depression; MedDietScore = Mediterranean diet score.

Table 2 - Comparison of gastrointestinal symptoms among patients with late-life depression taking different types of antidepressant drugs (n = 69)
GSRS score SSRI (n = 18) SNRI (n = 14) Agomelatine (n = 23) Nonmedication (n = 14) p
GSRS total score 6.3 (3.4) 6.1 (4.3) 6.1 (4.5) 6.9 (4.5) 0.96
 GSRS: Reflux syndrome 1.2 (1.0) 0.9 (1.2) 1.5 (1.8) 1.3 (1.2) 0.65
 GSRS: Diarrhea syndrome 1.2 (1.1) 0.9 (1.0) 1.0 (1.1) 1.4 (1.1) 0.66
 GSRS: Constipation syndrome 1.2 (1.2) 2.0 (1.6) 1.5 (1.9) 0.9 (0.9) 0.27
 GSRS: Abdominal pain 0.8 (0.9) 0.4 (0.6) 0.8 (1.3) 0.8 (0.7) 0.66
 GSRS: Dyspepsia 2.1 (2.1) 1.9 (1.9) 2.0 (1.7) 2.7 (2.4) 0.68
GSRS = Gastrointestinal Symptom Rating Scale; SNRI = serotonin-norepinephrine reuptake inhibitor; SSRI = selective serotonin reuptake inhibitor.

Table 3 - Linear regression model for the association between the depressive symptoms and gastrointestinal symptoms in patients with late-life depression (n = 69)a
GSRS score HAMD total scores without somatic subscores
β (95% CI) p
GSRS total score 0.47 (0.14-0.81) <0.01
 GSRS: Reflux syndrome 1.47 (0.46-2.48) <0.01
 GSRS: Diarrhea syndrome -0.34 (-1.64 to 0.98) 0.61
 GSRS: Constipation syndrome -0.23 (-1.17 to 0.71) 0.62
 GSRS: Abdominal pain 1.98 (0.41-3.55) <0.01
 GSRS: Dyspepsia 1.02 (0.36-1.68) <0.01
aAdjusted for demographic data and Mediterranean diet score.
GSRS = Gastrointestinal Symptom Rating Scale; HAMD = Hamilton Depression Rating Scale.

Table 4 - Linear regression model for the association between the depressive symptoms and gastrointestinal symptoms in male patients (n = 16) and female patients (n = 53)a
GSRS score HAMD total scores without somatic subscores
Male Female
β (95% CI) p β (95% CI) p
GSRS total score -0.12 (-1.46 to 1.22) 0.84 0.57 (0.25-0.9) <0.01
 GSRS: Reflux syndrome -0.9 (-5.45 to 3.65) 0.66 1.74 (0.77-2.72) <0.01
 GSRS: Diarrhea syndrome -1.46 (-5.38 to 2.46) 0.42 -0.11 (-1.47 to 1.26) 0.88
 GSRS: Constipation syndrome -0.99 (-4.83 to 2.86) 0.57 -0.18 (-1.11 to 0.74) 0.69
 GSRS: Abdominal pain 0.19 (-12.57 to 12.96) 0.97 2.13 (0.73-3.53) <0.01
 GSRS: Dyspepsia 0.39 (-2.55 to 3.34) 0.77 1.06 (0.44-1.68) <0.01
aAdjusted for demographic data and Mediterranean diet score.
GSRS = Gastrointestinal Symptom Rating Scale; HAMD = Hamilton Depression Rating Scale.

Logistic regression analyses with demographic variables and MedDietScore controlled for revealed that total GSRS score was an independent determinant of LLD (OR: 1.20, 95% CI: 1.04-1.38) (Table 5). Subgroup analyses for subscales of GSRS indicated that older adults with more reflux (OR: 1.97, 95% CI: 1.14-3.41), abdominal pain (OR: 2.41, 95% CI: 1.05-5.52), or dyspepsia (OR: 1.57, 95% CI: 1.13-2.18) had a greater risk of LLD (Table 5). Among the patients with LLD, 31, 21, and 5 were defined as having mild, moderate, and severe depression, respectively. Logistic regression analyses for subgroups stratified by depression severity indicated that GSRS score was the only variable that potentially contributed to a higher depression severity (OR: 1.25, 95% CI: 1.04-1.52) (Table 6). This correlation was only noted when the mild and moderate depression groups were compared.

Table 5 - Logistic regression analyses for adjusted odds ratio of late-life depression (n = 106)
Variables Adjusted odds ratio (95% CI) p
Age 1.11 (1.02-1.19) 0.01
Sex 2.57 (0.83-7.93) 0.10
Edu 0.93 (0.83-1.03) 0.17
BMI 0.94 (0.80-1.10) 0.43
DM 0.86 (0.19-3.88) 0.86
MedDietScore 0.69 (0.57-1.45) 0.69
GSRS total score 1.20 (1.04-1.38) 0.01
 GSRS: Reflux syndrome 1.97 (1.14-3.41) 0.02
 GSRS: Diarrhea syndrome 0.99 (0.65-1.49) 0.94
 GSRS: Constipation syndrome 1.21 (0.84-1.74) 0.32
 GSRS: Abdominal pain 2.41 (1.05-5.52) 0.04
 GSRS: Dyspepsia 1.57 (1.13-2.18) < 0.01
BMI = body mass index; DM = diabetes mellitus; GSRS = Gastrointestinal Symptom Rating Scale; MedDietScore = Mediterranean diet score.

Table 6 - Adjusted odds ratios for the risk of moderate depression (n = 21) versus mild depression (n = 31)
Variables Moderate vs mild depression
Adjusted OR (95% CI) p
Age 1.07 (0.96-1.19) 0.21
Sex 0.98 (0.14-6.99) 0.98
Edu 0.97 (0.86-1.11) 0.68
BMI 1.13 (0.85-1.50) 0.40
DM 1.50 (0.22-10.36) 0.68
MedDietScore 0.83 (0.43-1.60) 0.58
GSRS total score 1.25 (1.04-1.52) 0.02
BMI = body mass index; GSRS = Gastrointestinal Symptom Rating Scale; MedDietScore = Mediterranean diet score; OR = odds ratio.


The key finding of our study was that patients with LLD had higher levels of GI symptoms, including reflux syndrome, abdominal pain, and dyspepsia, than did the controls. Certain GI symptoms, such as reflux syndrome, abdominal pain, and dyspepsia, were significantly correlated with depressive symptoms among patients with LLD. Older adults with more GI symptoms, especially reflux syndrome, abdominal pain, and dyspepsia, may have a higher risk of LLD.

Our observations that patients with LLD had more reflux symptoms accord with those in the literature. In a cross-sectional study using Taiwan’s National Health Insurance Research Database, patients over 65 years old with MDD were noted to have a higher prevalence of gastroesophageal reflux disease relative to the general population.16 This finding may be attributable to the lower threshold for bodily sensation due to psychiatric factors and to the altered perception of esophageal stimuli in patients with depression.17 An association between depression and dyspepsia was reported in the adult population,18 and our findings suggest a higher level of dyspepsia in patients with LLD than in people without depression. Dyspeptic symptoms may be attributable to visceral hypersensitivity, which was reported to be associated with emotion dysregulation.19 A previous study linked altered alpha-adrenoceptor function and depression to a G-protein beta-3 subunit gene polymorphism.20 The genetic architecture of GI syndromes and the effects of interactions between gene polymorphisms and the environment warrant further investigation to determine the roles of interoceptive awareness and central nervous system dysregulation in GI syndromes. In addition, a study reported increased activation of brain regions such as the anterior cingulate cortex, thalamus, and prefrontal cortex in response to visceral stimuli in adult patients with irritable bowel syndrome, a GI disorder characterized by abdominal pain in the absence of an organic disorder.21 The aforementioned brain regions that process GI sensory information largely overlap with regions involved in emotional regulation, which may form a structural foundation for the coexistence of abdominal pain and depression.22,23

The correlation between depression and GI symptoms cannot be explained using a single model because various interactions are involved.24 The brain-gut axis may involve bidirectional communication in the development of GI symptoms. A high depression level in patients with functional GI disorder was reported in a previous study, and depression was considered to be an important predictor of functional dyspepsia and irritable bowel syndrome.25 Our findings indicate that functional GI symptoms may be independent predictors of depression in older adults. We further stratified GI symptoms by GSRS subscale measure and found that reflux syndrome, abdominal pain, and dyspepsia are possible determinants of depression in old age. Our subgroup analyses further indicated that the association between depressive symptoms and GI symptoms was present only among female patients. However, this result may be biased because of the small number of male participants with LLD. Moreover, subgroup analysis stratified by depression severity indicated that a higher total GSRS score contributes to greater depression severity only in the comparison of groups with mild and moderate depression. No differences were observed between other groups, possibly because of our small sample of patients with severe depression.

Diet has been generally accepted for its role in the pathophysiology of GI disorders.26 Adherence to an MDP, characterized by abundant intake of plant-based foods and olive oil and a moderate intake of fish, has been linked to a lower likelihood of depressive symptoms in older adults.27 The protective role of the MDP against depression encompasses positive synergic actions, anti-inflammatory functions, and protection from oxidative stress.28,29 In Taiwan, the excessive intake of meat by men and women aged 19-64 years results in an excessive intake of protein, cholesterol, and saturated fat. Older adults (>65 years) were noted to have a low intake of dietary fiber.30 Difficulty chewing and swallowing may influence the amount of fiber consumed among older adults.30 This may explain why we found no significant difference in MedDietScore between the patients with LLD and the controls. Moreover, the GSRS total score and subscores remained related to depression even after MedDietScore was controlled for, suggesting that diet may not fully explain the presence of depressive symptoms in older patients. Despite the increasing emphasis on the brain-gut axis, the role of diet in this pathway remains uninvestigated.

Because depression in older adults remains underdiagnosed and inadequately treated, patients with LLD often have a poor long-term prognosis, a more chronic disease course, and a higher relapse rate. The findings of our study can aid clinical practice in highlighting the importance of considering an older adult patient’s history of somatic symptoms, particularly GI symptoms. Clinicians should be aware that LLD may present without typical subjective mood symptoms but with a more pronounced frequency of somatic symptoms. Knowledge of a patient’s co-occurring or even preceding GI symptoms can help physicians pay more attention to older patients with depression to provide them with suitable and timely treatment. Future studies should investigate the relationship between mental illness and the GI tract in older adults.

Our study had the following limitations. First, our sample size was small, particularly for the subsamples of the control group and male patients with LLD. Future studies should recruit more participants to produce findings with higher statistical power. Second, participants with depression were taking medication. Psychotropic medications, particularly antidepressant medications, have known GI side effects. We did not account for this influence, which potentially affected our results. Third, we had an insufficient number of participants with more severe depression. Thus, we could not investigate the association between GI symptoms and depression among subgroups of patients with different symptom severities. Fourth, we did not account for lifestyle factors, such as smoking and sedentary behavior; these factors have a known influence on reflux symptoms. Finally, GI symptoms were recorded retrospectively. This constitutes a limitation because the occurrence of mood disturbances may increase recall effects because retrospective ratings of symptoms are biased by the participant’s feelings as they complete the questionnaire, leading to positive or negative recall.

Our findings provide evidence supporting the association between GI symptoms and depressive symptoms among patients with LLD. Older people with more specific GI symptoms, such as reflux, abdominal pain, and dyspepsia, may have a higher risk of LLD. Future studies that account for neurobiological and genetic factors should be conducted to further elucidate the relationship between GI symptoms and depression. Clinicians should remain vigilant to signs of mental disorders when evaluating older patients with GI symptoms.


This study was supported by a grant from the Ministry of Science and Technology, Taiwan (MOST 104-2314-B-075-040).

We thank Dr. MHC, Dr. WYP, Ms. WPS, Dr. CCM, Dr. LCL, and Dr. TCF, who designed the study and wrote the protocol and manuscripts; Drs. CYL, CCM, YCH, TSJ, LCL, and TCF, who assisted with preparing and proofreading the manuscript; and Drs. CCM and TCF, who provided advice for statistical analysis. This manuscript was edited by Wallace Academic Editing.


1. Djernes JK. Prevalence and predictors of depression in populations of elderly: a review. Acta Psychiatr Scand 2006; 113:372–87.
2. Jeong HG, Han C, Park MH, Ryu SH, Pae CU, Lee JY, et al. Influence of the number and severity of somatic symptoms on the severity of depression and suicidality in community-dwelling elders. Asia Pac Psychiatry 2014; 6:274–83.
3. Hegeman JM, Kok RM, van der Mast RC, Giltay EJ. Phenomenology of depression in older compared with younger adults: meta-analysis. Br J Psychiatry 2012; 200:275–81.
4. Jackson JL, Houston JS, Hanling SR, Terhaar KA, Yun JS. Clinical predictors of mental disorders among medical outpatients. Arch Intern Med 2001; 161:875–9.
5. Mayer EA, Tillisch K, Gupta A. Gut/brain axis and the microbiota. J Clin Invest 2015; 125:926–38.
6. Van Oudenhove L, Crowell MD, Drossman DA, Halpert AD, Keefer L, Lackner JM, et al. Biopsychosocial aspects of functional gastrointestinal disorders. Gastroenterology 2016; 150:1355–67.
7. Rogers GB, Keating DJ, Young RL, Wong ML, Licinio J, Wesselingh S. From gut dysbiosis to altered brain function and mental illness: mechanisms and pathways. Mol Psychiatry 2016; 21:738–48.
8. Mitsou EK, Kakali A, Antonopoulou S, Mountzouris KC, Yannakoulia M, Panagiotakos DB, et al. Adherence to the Mediterranean diet is associated with the gut microbiota pattern and gastrointestinal characteristics in an adult population. Br J Nutr 2017; 117:1645–55.
9. Blazer DG. Depression in late life: review and commentary. J Gerontol A Biol Sci Med Sci 2003; 58:249–65.
10. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23:56–62.
11. Svedlund J, Sjödin I, Dotevall G. GSRS–a clinical rating scale for gastrointestinal symptoms in patients with irritable bowel syndrome and peptic ulcer disease. Dig Dis Sci 1988; 33:129–34.
12. Kulich KR, Madisch A, Pacini F, Piqué JM, Regula J, Van Rensburg CJ, et al. Reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire in dyspepsia: a six-country study. Health Qual Life Outcomes 2008; 6:12.
13. Huang YC, Lee MS, Pan WH, Wahlqvist ML. Validation of a simplified food frequency questionnaire as used in the Nutrition and Health Survey in Taiwan (NAHSIT) for the elderly. Asia Pac J Clin Nutr 2011; 20:134–40.
14. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003; 348:2599–608.
15. Zimmerman M, Martinez JH, Young D, Chelminski I, Dalrymple K. Severity classification on the Hamilton Depression Rating Scale. J Affect Disord 2013; 150:384–8.
16. Chou PH, Lin CC, Lin CH, Tsai CJ, Cheng C, Chuo YP, et al. Prevalence of gastroesophageal reflux disease in major depressive disorder: a population-based study. Psychosomatics 2014; 55:155–62.
17. Avidan B, Sonnenberg A, Giblovich H, Sontag SJ. Reflux symptoms are associated with psychiatric disease. Aliment Pharmacol Ther 2001; 15:1907–12.
18. Mak AD, Wu JC, Chan Y, Chan FK, Sung JJ, Lee S. Dyspepsia is strongly associated with major depression and generalised anxiety disorder - a community study. Aliment Pharmacol Ther 2012; 36:800–10.
19. Tack J, Caenepeel P, Fischler B, Piessevaux H, Janssens J. Symptoms associated with hypersensitivity to gastric distention in functional dyspepsia. Gastroenterology 2001; 121:526–35.
20. Holtmann G, Siffert W, Haag S, Mueller N, Langkafel M, Senf W, et al. G-protein beta 3 subunit 825 CC genotype is associated with unexplained (functional) dyspepsia. Gastroenterology 2004; 126:971–9.
21. Chang L, Berman S, Mayer EA, Suyenobu B, Derbyshire S, Naliboff B, et al. Brain responses to visceral and somatic stimuli in patients with irritable bowel syndrome with and without fibromyalgia. Am J Gastroenterol 2003; 98:1354–61.
22. Van Oudenhove L, Demyttenaere K, Tack J, Aziz Q. Central nervous system involvement in functional gastrointestinal disorders. Best Pract Res Clin Gastroenterol 2004; 18:663–80.
23. Mayer EA, Tillisch K. The brain-gut axis in abdominal pain syndromes. Annu Rev Med 2011; 62:381–96.
24. Jansson C, Nordenstedt H, Wallander MA, Johansson S, Johnsen R, Hveem K, et al. Severe gastro-oesophageal reflux symptoms in relation to anxiety, depression and coping in a population-based study. Aliment Pharmacol Ther 2007; 26:683–91.
25. Koloski NA, Jones M, Kalantar J, Weltman M, Zaguirre J, Talley NJ. The brain–gut pathway in functional gastrointestinal disorders is bidirectional: a 12-year prospective population-based study. Gut 2012; 61:1284–90.
26. Pilichiewicz AN, Horowitz M, Holtmann GJ, Talley NJ, Feinle-Bisset C. Relationship between symptoms and dietary patterns in patients with functional dyspepsia. Clin Gastroenterol Hepatol 2009; 7:317–22.
27. Skarupski KA, Tangney CC, Li H, Evans DA, Morris MC. Mediterranean diet and depressive symptoms among older adults over time. J Nutr Health Aging 2013; 17:441–5.
28. Kyrozis A, Psaltopoulou T, Stathopoulos P, Trichopoulos D, Vassilopoulos D, Trichopoulou A. Dietary lipids and geriatric depression scale score among elders: the EPIC-Greece cohort. J Psychiatr Res 2009; 43:763–9.
29. Sánchez-Villegas A, Henríquez P, Bes-Rastrollo M, Doreste J. Mediterranean diet and depression. Public Health Nutr 2006; 9:1104–9.
30. Chen HL, Huang Yi. Fiber intake and food selection of the elderly in Taiwan. Nutrition 2003; 19:332–6.

Gastrointestinal symptoms; Late-life depression; Mediterranean diet

Copyright © 2020, the Chinese Medical Association.