As globally prevalent mental health conditions, anxiety and depression impose significant socioeconomic burdens on healthcare systems. Key risk factors for anxiety and depression were determined in the current study by examining the associations across multiple domains.
Methods
Participants were enrolled for this cross-sectional study from the Department of Gerontology at the Fifth Affiliated Hospital of Sun Yat-sen University in 2023. Anxiety and depression were evaluated based on the Hamilton Anxiety (HAM-A) and Hamilton Depression (HAMD) Rating Scales, respectively. Cognitive function, sleep quality, frailty, and nutritional status were evaluated using the Montreal Cognitive Assessment (MoCA), the Pittsburgh Sleep Quality Index (PSQI), the FRAIL scale, and the Nutrition Risk Screening (NRS)-2002, respectively. Multiple linear regression analysis was performed to determine the factors associated with anxiety and depression.
Results
A total of 1121 participants (659 women and 462 men; mean age ± SD: 64.67 ± 10.86 years) were studied. HAMD scores were significantly associated with gender (β = 0.078, P = 0.017), body mass index (β = -0.096, P = 0.003), MoCA (β = -0.141, P < 0.001), FRAIL score (β = 0.200, P < 0.001), PSQI (β = 0.494, P < 0.001), and NRS-2002 (β = 0.099, P = 0.004) after full adjustment. HAM-A scores exhibited significant and independent associations with gender (β = 0.155, P < 0.001), illness (β = -0.098, P = 0.006), MoCA (β = -0.114, P = 0.001), PSQI (β = 0.435, P < 0.001), and FRAIL score (β = 0.276, P < 0.001).
Conclusions
Male patients with superior physical and cognitive function, better sleep, favorable body mass index and nutrition status, and fewer illnesses had better mood. Targeted interventions, including cognitive preservation, sleep optimization, frailty prevention, comorbidity management, and lifestyle modification, may be correlated with reduced symptom burden. Special consideration is warranted for elderly and female populations.
anxietycognitive functiondepressionfrailtygendersleep qualityThe author(s) declared that financial support was not received for this work and/or its publication.section-at-acceptanceAging PsychiatryIntroduction
Anxiety and depression are common mental health disorders worldwide. Prior to 2020, the estimated global prevalence of anxiety and depression disorders was 3.6% and 4.4%, respectively (1). These conditions impose a substantial economic and social burden on public health systems (2). Anxiety and depression impose a substantial burden among hospitalized patients, contributing to increased healthcare utilization, longer hospital stays, and poorer recovery outcomes. However, greater than one-half of these patients do not receive appropriate mental health treatment (3). Without appropriate intervention, these disorders can progress into an illness, severely impairing mental and physical health. This finding underscores the critical importance of identifying modifiable risk factors and implementing preventive interventions for anxiety and depression in the hospital setting.
The etiology of anxiety and depression involves a complex interplay of biological, environmental, temperamental, and psychological factors (4). Well-established risk factors include female gender, physical illness, cognitive impairment, limited social contact, and a history of psychiatric disorders (5–8). Previous studies have identified strong associations between mental disorders and cognitive function, frailty, and sleep quality (9–11). The prevalence of mental disorders affects sleep and increases the risk of functional and cognitive impairment, while cognitive decline, sleep disorders, and frailty exacerbate mental disorders (12–15). Given these established associations, future inferential research should rigorously control for confounding factors. The mechanisms underlying mental disorders include hypercortisolemia, chronic inflammation, and common shared risk factors, such as vascular disease, chronic illness, unhealthy lifestyles, and malnutrition (1, 16, 17). Vascular risk factors, including body mass index (BMI), lipid and blood glucose levels, blood pressure, and cigarette smoking, induce neurodegenerative changes and mood-related symptoms (18–21). The occurrence and development of anxiety and depression are driven by interactions among multiple factors. Isolated analysis of individual determinants may therefore have limited explanatory power. However, most studies have examined risk factors from a single perspective, leaving the combined effects across multiple domains poorly understood. Consequently, adopting analytical approaches that capture this multidomain evaluation represents a critical advancement for the field.
Anxiety and depression are highly prevalent among hospitalized adults, yet they often remain underrecognized and inadequately managed in routine clinical practice. Although early prevention and multidomain interventions have shown promise in alleviating or even reversing mental health disorders, the contributing factors among general hospital inpatients have not been thoroughly investigated. Existing evidence lacks a comprehensive understanding of overall risk across multiple domains in this specific setting. This study was designed to determine the correlation between depression and anxiety with multiple risk factors, including demographic, epidemiologic, lifestyle, and clinical and biological domains in hospitalized patients. Our objective was to identify the most critical risk factors for anxiety and depression to inform the design of targeted strategies for promoting mental health.
MethodsStudy design and participants
Participants in this cross-sectional study were consecutively recruited from the Department of Gerontology at the Fifth Affiliated Hospital of Sun Yat-sen University (Zhuhai, China) from August 2023 to April 2024. Prior to recruitment, a sample size estimation was performed for the planned multiple linear regression analysis. Based on the 10–15 observations per predictor variable rule of thumb of and anticipating approximately 20 key predictors in the final model, a minimum of 200–300 participants was required. To ensure robust parameter estimation, account for missing data, and allow for model exploration, a recruitment target of 700 participants was set.
Following a standardized procedure, all eligible inpatients admitted during this period were screened against the inclusion and exclusion criteria upon admission. Those inpatients who met all criteria were invited to participate. The inclusion criteria were: (1) age≥45 years; (2) absence of any acute illnesses within the past 4 weeks; (3) overall health status deemed stable by the primary physician; and (4) sufficient audiovisual capacity to complete assessments. Exclusion criteria were: (1) inability to provide basic demographic information; (2) blindness or significant speech impairment; (3) acute infection or recent surgery; (4) severe functional dependency; (5) severe neurologic or psychiatric disorders; and (6) critical systemic co-morbidities, including vital organ failure, malignancy, or an expected survival < 1 year. The participant recruitment process is summarized in a flowchart (Figure 1). The study was approved by the Institutional Ethics Committee of the Fifth Affiliated Hospital of Sun Yat-sen University [No. ZDWY (2024); Lunzi No. (K169-1)], and all subjects provided written informed consent.
Flow diagram of participant screening, enrollment, and inclusion in the study. From August 2023 to April 2024, 1158 inpatients were consecutively screened for eligibility. Thirty-two patients were excluded based on predefined criteria. Of the 1126 eligible patients invited, 5 declined participation, resulting in a final analytic sample of 1121 participants.
Flowchart illustrating participant selection: one thousand one hundred fifty-eight assessed for eligibility, thirty-two excluded for various reasons, one thousand one hundred twenty-six invited, five declined to participate, resulting in one thousand one hundred twenty-one included in final analysis.Clinical data and blood tests
Participants completed a detailed questionnaire upon enrollment, providing information on basic parameters, social demographics, and medical history, including age, gender, height, weight, education level, employment status, smoking history, and disease and medication history. Height and weight were measured with participants in underwear without shoes. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m²). Blood pressure was measured after a 10-minute rest, and the average value was recorded. Venous blood was collected after a 10-hour fast for biochemical analysis, including routine hematology, serum creatinine, fasting blood glucose, total serum protein, serum albumin, and lipid profiles.
Depression and anxiety
Depression was assessed using the observer-rated Hamilton Depression Rating Scale (HAMD), which was designed to measure depressive symptom severity (22). Most items use a 5-point scale (0–4), while some use a 3-point scale (0–2). Depression severity scores were: 7–17, suggest possible depression; 17–20 indicate mild depression; and > 20 signify moderate-to-severe depression. Higher scores indicate greater severity. Anxiety was assessed using the Hamilton Anxiety Rating Scale [HAM-A] (23). This 14-item scale uses a 5-point scale (0–4 per item; total range 0-54). Anxiety severity scores were: > 7, indicate anxiety symptoms; 7–13, indicate mild anxiety; 14–21, indicate moderate anxiety; and > 21, indicate severe anxiety.
Assessment of Cognitive Function
Cognitive function was evaluated using the Montreal Cognitive Assessment [MoCA] (24, 25). The Montreal Cognitive Assessment (MoCA) evaluates multiple cognitive domains, including memory, language, attention, and executive function. The maximum score is 30 points, with a score above 26 considered normal. An additional point is added for individuals with fewer than 12 years of education.
Sleep quality
Sleep quality was evaluated using the Pittsburgh Sleep Quality Index [PSQI] (26). This reliable and valid instrument measures the following seven dimensions of sleep: subjective quality; duration; latency; disturbances; efficiency; daytime dysfunction; and medication use. Each dimension is scored 0–3, yielding a total score from 0–21. A score of 7 or higher indicates a sleep disorder with higher scores denoting poorer sleep quality.
FRAIL Scale
Frailty was assessed using the FRAIL scale (27, 28). This instrument consists of the following five self-reported components: fatigue (frequent tiredness in the past 4 weeks); resistance (difficulty climbing 10 stairs unaided); walking ability (difficulty walking 500–600 meters unaided); co-morbidity (five or more diagnosed conditions, e.g., hypertension, diabetes, and stroke); and weight loss (≥ 5% body weight loss in the past year). Scores range from 0–5 with 0 indicating robust health, 1–2 indicating pre-frailty, and ≥ 3 indicating frailty.
Nutrition and activities of daily living
The evaluation of nutrition risk was performed with the Nutrition Risk Screening (NRS)-2002 (29). This instrument evaluates the following five parameters: (1) unexplained weight loss in the past 3 months; (2) BMI; (3) appetite; (4) disease severity; and (5) age over 70 years. A total score < 3 indicates no nutritional risk, while a score ≥ 3 signifies high nutritional risk, indicating the need for nutritional support. Activities of daily living were evaluated using the Barthel Index (29, 30). Scores range from 0–100 with lower scores indicating greater impairment in daily activities.
All assessment scales, including the HAMD, MoCA, and other instruments, were administered in person by trained research assistants. Prior to the study commencing, all raters completed a standardized training session on the proper administration and scoring of each tool to ensure assessment consistency and reliability.
Statistical analysis
Normality was assessed using the Kolmogorov-Smirnov test. Normally distributed data are presented as mean ± standard deviation (SD) and compared using one-way analysis of variance (ANOVA). Non-normally distributed data are presented as median (interquartile range) and compared using the Kruskal-Wallis H-test. Categorical data are presented as counts and compared using the χ2 test or Fisher probabilities. Correlations between variables were analyzed using Spearman’s correlation coefficient. Multiple linear regression analysis was employed to adjust for potential confounding variables. Statistical significance was defined as a P-value < 0.05. All analyses were performed using SPSS 26.0.
ResultsParticipant characteristics
Between August 2023 and April 2024, a total of 1158 participants were screened. After applying the inclusion and exclusion criteria, a final sample of 1121 participants was enrolled and included in the analysis, which exceeded the initial target sample size. The cohort had a mean age of 64.67 years (SD = 10.86) and was comprised of 659 (58.8%) women and 462 (41.2%) men. Most participants (57.28%) had a secondary education or below, while 22.39% completed high school and 20.33% held a university degree or above. Nearly one-half (47.3%) were unemployed or homemakers, followed by retirees (38.2%), laborers (8.5%), and technical workers or cadres (6.0%).
Basic data and clinical variables of participants based on HAMD and HAM-A groups
The basic and clinical characteristics according to the HAMD scores (<17, 17–20, and ≥ 20) are shown in Table 1. All variables, except systolic blood pressure (SBP), diastolic blood pressure (DBP), polypharmacy, fasting blood glucose (FBG), total cholesterol (TC), low-density lipoprotein (LDL), serum creatinine (SCr), total protein (TP) and C-reactive protein (CRP) were significantly different among the three HAMD groups (all P < 0.05). The following parameters increased in all participants with progression from low-to-high HAMD scores (all P < 0.05): age, serum cystatin C (CYSC), high-density lipoprotein (HDL), Fazekas score, FRAIL scale score; PSQI, and NRS-2002; while the BMI, education level, hemoglobin (HGB), uric acid (UA), estimated glomerular filtration rate (eGFR), triglycerides (TG), albumin (ALB), MoCA, and activities of daily living (ADL) decreased.
Overall variables of the participants based on HAMD groups.
Variable
HAMD
<17(n=848)
17-20(n=165)
≥20 (n=58)
P value
Age (years)
63.21 ± 10.33
69.40 ± 10.95**
69.40 ± 10.45**
<0.001
Gender
<0.001
Male (%)
371 (43.8)
59 (35.8)
10 (17.2)
BMI (kg/m²)
24.37 ± 3.46
23.44 ± 3.78**
21.89 ± 3.00**##
<0.001
SBP (mmHg)
126.72 ± 13.11
128.7 ± 13.29
128.17 ± 14.82
0.164
DBP (mmHg)
78.19 ± 9.31
77.26 ± 9.15
75.59 ± 8.21
0.071
Occupational status
<0.001
Unemployed/housewife/house husband
359 (43.2)
99 (61.1)
38 (66.7)
Retirement pension
80 (9.6)
8 (4.9)
1 (1.8)
Laborer
334 (40.2)
52 (32.1)
16 (28.1)
Technical worker and cadre
58 (7.0)
3 (1.8)
2 (3.6)
Education level (years)
10.01 ± 4.19
8.82 ± 4.45**
7.38 ± 5.05**#
<0.001
Cigarette smoking
0.011
Never smoker
721 (85.7)
141 (86.5)
57 (98.3)
Former smoker
30 (3.6)
11 (6.7)
0
Current smoker
90 (10.7)
11 (6.7)
1 (1.7)
Illness (n, %)
0.020
None
187 (22.2)
21 (12.9)
8 (13.8)
One or two diseases
291 (34.6)
53 (32.5)
21 (36.2)
Three or more diseases
363 (43.2)
89 (54.6)
29 (50.0)
Polypharmacy
0.086
Yes
140 (16.8)
39 (23.9)
12 (20.7)
HGB (g/L)
132.35 ± 15.90
127.8 ± 15.62**
125.00 ± 15.90**
<0.001
FBG (mmol/L)
5.58 ± 1.60
5.68 ± 1.74
5.54 ± 1.28
0.763
SCr (mmol/L)
77.81 ± 22.98
78.90 ± 26.49
79.17 ± 29.09
0.812
eGFR(ml/min.1.73m²)
82.31 ± 16.85
76.74 ± 19.04**
74.86 ± 20.84**
<0.001
UA (μmol/L)
356.17 ± 99.94
325.9 ± 96.97**
300.27 ± 66.66**
<0.001
CYSC (mg/L)
0.90 ± 0.20
0.95 ± 0.21**
0.96 ± 0.27*
0.005
TG (mmol/L)
1.59 ± 1.34
1.39 ± 0.78
1.18 ± 0.45*
0.014
TC (mmol/L)
4.82 ± 1.15
4.83 ± 1.31
4.78 ± 1.08
0.957
HDL (mmol/L)
1.28 ± 0.36
1.35 ± 0.42*
1.41 ± 0.43*
0.007
LDL (mmol/L)
2.87 ± 0.98
2.85 ± 1.14
2.82 ± 1.01
0.934
TP (g/L)
67.49 ± 5.08
67.14 ± 5.29
67.23 ± 5.60
0.699
ALB (g/L)
40.64 ± 3.01
39.61 ± 3.76**
39.73 ± 3.42*
<0.001
CRP (mg/L)
1.04 (0.07-2.50)
1.05 (0.18-2.37)
0.55 (0.07-0.95)
0.271
Fazekas scores
1.14 ± 0.65
1.27 ± 0.71
1.47 ± 0.84*
0.041
Frailty scores
0.96 ± 1.02
1.84 ± 1.18**
2.45 ± 1.18**##
<0.001
MoCA
20.68 ± 5.28
16.93 ± 5.99**
15.24 ± 6.29**#
<0.001
PSQI
8.43 ± 3.86
12.98 ± 3.32**
14.97 ± 3.50**##
<0.001
ADL
94.89 ± 9.62
87.04 ± 17.23**
82.50 ± 17.65**#
<0.001
NRS
0.00 (0.00-1.00)
1.00 (0.00-2.00)**
1.00 (0.00-2.00)**
<0.001
Normally distributed data are expressed as the mean ± SD and the differences in data were examined using one-way analysis of variance (ANOVA). Data that were not normally distributed are shown as a median and interquartile range. Comparisons of these variables among groups were performed using the Kruskal–Wallis H-test. Least significant difference or Dunnett’s T3 analyses were used for comparisons between pairs of groups according to the homogeneity of variance test. P-values are the outcomes of ANOVA or the Kruskal–Wallis H-test. *#Significant difference (P < 0.05). **##Significant difference (P < 0.01) when compared to the < 17 and 17–20 groups. ADL, activities of daily living; ALB, serum albumin; CRP, C-reactive protein; CYSC, cystatin C; BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FBG, fasting blood glucose; HDL, high-density lipoprotein; HGB, hemoglobin; LDL, low-density lipoprotein; NRS, nutritional risk; SBP, systolic blood pressure; SCr, serum creatinine; TC, total cholesterol; TG, triglycerides; TP, total serum protein; UA, uric acid.
The variables of the participants are shown according to HAM-A groups in Table 2. A higher HAM-A score was significantly associated with higher age, HDL, FRAIL score, PSQI and NRS-2002 but a lower DBP, education level, HGB, eGFR, UA, TG, ALB, MoCA, and ADL. There were no significant differences in BMI, SBP, FBG, SCr, CYSC, TC, LDL, TP, or Fazekas score with respect to HAM-A scores.
Overall variables of the participants based on HAM-A groups.
Variable
HAMA
<14(n=254)
14-21(n=662)
≥21(n=155)
P value
Age (years)
61.06 ± 9.56
64.74 ± 10.56**
69.06 ± 11.38**##
<0.001
Gender
<0.001
Male (%)
139 (54.7)
267 (40.3)
34 (21.9)
BMI ((kg/m²)
24.19 ± 3.05
24.19 ± 3.62
23.55 ± 3.92
0.119
SBP (mmHg)
127.08 ± 13.76
127.05 ± 13.11
127.42 ± 13.07
0.951
DBP (mmHg)
79.39 ± 9.45
77.77 ± 9.37*
76.08 ± 7.89**#
0.002
Occupational status
<0.001
Unemployed/housewife/house husband
74 (30.0)
331 (51.0)
91 (59.1)
Retirement pension
37 (15.0)
44 (6.8)
8 (5.2)
Laborer
106 (42.9)
246 (37.9)
50 (32.5)
Technical worker and cadre
30 (12.1)
28 (4.3)
5 (3.2)
Education level
11.19 ± 3.65
9.51 ± 4.31**
7.95 ± 4.64**##
<0.001
Cigarette smoking
0.029
Never smoker
212 (84.1)
562 (85.7)
145 (94.2)
Former smoker
9 (3.6)
30 (4.6)
2 (1.3)
Current smoker
31 (12.3)
64 (9.8)
7 (4.5)
Illness
0.012
None
70 (27.8)
124 (18.9)
22 (14.3)
One or two diseases
79 (31.3)
229 (34.9)
57 (37.0)
Three or more diseases
103 (40.9)
303 (46.2)
75 (48.7)
Polypharmacy
0.010
Yes
32 (12.9)
121 (18.6)
38 (24.8)
HGB (g/L)
135.91 ± 14.07
130.5 ± 16.33**
126.52 ± 15.76**##
<0.001
FBG (mmol/L)
5.47 ± 1.45
5.62 ± 1.65
5.66 ± 1.64
0.388
SCr (mmol/L)
79.10 ± 20.95
77.67 ± 24.47
77.99 ± 25.93
0.723
eGFR(ml/min.1.73m²)
83.80 ± 15.76
81.28 ± 17.54
75.60 ± 19.50**##
<0.001
UA (μmol/L)
364.79 ± 91.98
349.31 ± 103.12*
318.23 ± 86.11**##
<0.001
CYSC (mg/L)
0.88 ± 0.15
0.91 ± 0.22
0.93 ± 0.20
0.054
TG (mmol/L)
1.67 ± 1.58
1.53 ± 1.18
1.33 ± 0.78*
0.037
TC (mmol/L)
4.89 ± 1.04
4.81 ± 1.19
4.73 ± 1.27
0.370
HDL (mmol/L)
1.25 ± 0.35
1.29 ± 0.37
1.38 ± 0.42**#
0.004
LDL (mmol/L)
2.93 ± 0.86
2.86 ± 1.02
2.74 ± 1.15
0.189
TP (g/L)
67.43 ± 5.01
67.37 ± 5.14
67.60 ± 5.36
0.885
ALB (g/L)
40.85 ± 2.99
40.37 ± 3.22*
40.01 ± 3.30*
0.026
CRP (mg/L)
0.99 (0.07-2.13)
1.02 (0.07-2.74)
0.90 (0.25-1.92)
0.102
Fazekas scores
1.05 ± 0.63
1.21 ± 0.66
1.23 ± 0.74
0.058
Frailty scores
0.61 ± 0.79
1.18 ± 1.09**
1.94 ± 1.27**##
<0.001
MoCA
22.19 ± 4.51
19.58 ± 5.66**
16.90 ± 6.18**##
<0.001
PSQI
6.53 ± 3.55
9.64 ± 3.83**
13.65 ± 3.61**##
<0.001
ADL
97.15 ± 5.62
93.16 ± 11.90**
85.61 ± 17.17**##
<0.001
NRS
0.00 (0.00-1.00)
0.00(0.00-1.00)**
1.00 (0.00-2.00)**##
<0.001
Abbreviations as in Table 1. Normally distributed data are expressed as the mean ± SD and the differences in these data were examined using one-way analysis of variance (ANOVA). Data that were not normally distributed are shown as a median and interquartile range. Comparisons of these variables among groups were performed using the Kruskal–Wallis H-test. Least significant difference or Dunnett’s T3 analyses were used for comparisons between pairs of groups according to the homogeneity of variance test. P-values are the outcome of ANOVA or the Kruskal–Wallis H-test. *#Significant difference (P < 0.05). **##Significant difference (P < 0.01) when compared to <14 and 14–21 groups.
As shown in Figure 2, HAMD and HAM-A scores varied significantly across different patient subgroups. Subsequent post-hoc tests confirmed that all inter-group comparisons were statistically significant (all P < 0.001). This finding included the following comparisons: older and middle-aged patients (Figures 1a, b); robust, pre-frail, and frail groups (with each group differing from the others; Figures 1c, d); good and poor sleep quality groups (Figures 1e, f); and normal cognition and mild cognitive impairment groups (Figures 1g, h).
Distribution of HAMD and HAM-A scores across key subgroups. The figure illustrates the distribution patterns of depression (HAMD; left panels) and anxiety (HAM-A; right panels) scores among participants, stratified by (a, b) age, (c, d) frailty, (e, f) sleep quality (PSQI), and (g, h) cognitive function (MoCA). Each panel compares the score distributions between men and women within the specified categories.
Grouped bar chart panels compare HAMD and HAMA scores by gender for four categories: age, frailty, PSQI scores, and MOCA scores, with each panel showing stratified means and error bars by subgroup color.Correlation of HAMD and HAM-A scores with basic characteristics and clinical variables
Correlations between HAMD and HAM-A scores and basic and clinical parameters are shown in Table 3. There was a weak correlation between the HAMD scores and age, occupational status, education level, HGB, MoCA, ADL, and NRS-2002 (0.2 < r < 0.4; P < 0.01). There was a moderate correlation between the HAMD scores and FRAIL score (r = 0.401; P < 0.01). There was a strong correlation between the HAMD score and PSQI (r = 0.605; P < 0.01). The HAMA score was correlated with age, gender, education level, HGB, MoCA, ADL, NRS-2002, FRAIL score (0.2 < r < 0.4; P < 0.01), and PSQI (r = 0.553; P < 0.01).
Correlation of HAMD and HAM-A scores with clinical variables, Rp (Spearman’s correlation coefficient).
Variable
HAMD
HAMA
rp
p
rp
p
Age (years)
0.286**
<0.001
0.237**
<0.001
Gender
0.143**
<0.001
0.218**
<0.001
BMI ((kg/m²)
-0.168**
<0.001
-0.054
0.080
SBP (mmHg)
0.011
0.712
-0.007
0.827
DBP (mmHg)
-0.129**
<0.001
-0.110**
<0.001
Occupational status
-0.251**
<0.001
-0.171**
<0.001
Education level
-0.229**
<0.001
-0.216**
<0.001
Cigarette smoking
-0.067*
0.029
-0.076*
0.013
Illness
0.131**
<0.001
0.110**
<0.001
Polypharmacy
0.096**
0.002
0.103**
0.001
HGB (g/L)
-0.223**
<0.001
-0.204**
<0.001
FBG (mmol/L)
0.034
0.269
0.050
0.103
SCr (mmol/L)
-0.023
0.462
-0.081**
0.009
eGFR (ml/min. 1.73m²)
-0.158**
<0.001
-0.117**
<0.001
UA (μmol/L)
-0.199**
<0.001
-0.160**
<0.001
CYSC (mg/L)
0.128**
<0.001
0.052
0.114
TG (mmol/L)
-0.104**
0.001
-0.056
0.068
TC (mmol/L)
-0.056
0.068
-0.055
0.072
HDL (mmol/L)
0.076*
0.014
0.091**
0.003
LDL (mmol/L)
-0.062*
0.045
-0.069*
0.025
TP (g/L)
0.015
0.620
0.021
0.504
ALB (g/L)
-0.149**
<0.001
-0.098**
0.001
CRP (mg/L)
0.013
0.806
0.084
0.100
Fazekas scores
0.138**
0.002
0.087*
0.048
Frailty scores
0.401**
<0.001
0.381**
<0.001
MoCA
-0.366**
<0.001
-0.279**
<0.001
PSQI
0.605**
<0.001
0.553**
<0.001
ADL
-0.310**
<0.001
-0.282**
<0.001
NRS
0.311**
<0.001
0.214**
<0.001
Abbreviations as in Table 1. P-values are from analysis of variance.
*Significant difference (P < 0.05).
**Significant difference (P < 0.01).
Relationship between the HAMD and HAM-A scores and clinical variables using multiple linear regression analysis
Multiple linear regression (stepwise entry) identified factors independently associated with HAMD and HAM-A scores (Table 4). The model included parameters correlated in pairwise analysis and traditional risk factors. After controlling for confounding factors, the HAMD score had a significant and independent correlation with gender (β = 0.078, P = 0.017), BMI (β = -0.096, P = 0.003), MoCA (β = -0.141, P < 0.001), PSQI (β = 0.494, P < 0.001), NRS-2002 (β = 0.099, P = 0.004), and FRAIL score (β = 0.200, P < 0.001). The HAMA score was significantly correlated with gender (β = 0.155, P < 0.001), illness (β = -0.098, P = 0.006), MoCA (β = -0.114, P = 0.001), PSQI (β = 0.435, P < 0.001), and FRAIL score (β = 0.276, P < 0.001).
Relationship between HAMD and HAM-A scores and clinical variables using a stepwise multiple regression model.
Variable
HAMD
HAM-A
Model1
Model2
Model1
Model2
Beta
p
Beta
p
Beta
p
Beta
p
Gender
0.080**
0.005
0.078*
0.017
0.141**
<0.001
0.155**
<0.001
BMI ((kg/m²)
-0.112**
<0.001
-0.096**
0.003
0.002
0.938
0.013
0.709
Illness
-0.017
0.571
-0.021
0.536
-0.102**
0.002
-0.098**
0.006
MoCA
-0.161**
<0.001
-0.141**
<0.001
-0.129**
<0.001
-0.114**
0.001
Frailty scores
0.231**
<0.001
0.200**
<0.001
0.285**
<0.001
0.276**
<0.001
PSQI
0.482**
<0.001
0.494**
<0.001
0.428**
<0.001
0.435**
<0.001
NRS
0.063*
0.046
0.099**
0.004
-0.023
0.500
-0.027
0.461
Beta, standardized coefficients. Model 1, unadjusted model; Model 2, fully adjusted for age, gender, BMI, SBP, DBP, HGB, FBG, TG, TC, LDL, SCr, eGFR, UC, CYSC, TP, ALB, working status, education level, illnesses, ADL, frailty scores, NRS, PSQI, and MoCA. Standardized coefficients and P values were the outcome of regression analyses.
*Significant difference (P < 0.05).
**Significant difference (P < 0.01).
Discussion
This study aimed to address a significant clinical gap by systematically investigating modifiable risk factors for anxiety and depression among general hospital patients. It contributes comprehensive evidence on sociodemographic, lifestyle, clinical, and biological determinants. This evidence base can inform the development of targeted interventions and support future strategies aimed at improving mental health outcomes in this patient population.
Our results confirm previous evidence linking cognitive impairment to anxiety and depression (12). After controlling for confounders, the MoCA score was significantly and independently correlated with the HAMD and HAM-A scores (all P < 0.01). Therefore, the results confirmed that cognitive impairment is an independent risk factor for mental disorders. Cognitive and mental disorders have been shown to have a bidirectional causal relationship (12). Mental disorders may be precursors to cognitive impairment and early non-cognitive manifestations of dementia (31, 32). Neurobiological research demonstrated that psychological disorders and dementia often have the same neuropathologic changes, particularly in the limbic system. Therefore, neurodegenerative changes within this system may result in emotion-related symptoms (33). Mechanisms potentially linking cognitive impairment and mental disorders include hypercortisolemia (34), chronic inflammation (35), and common shared risk factors, such as vascular disease, chronic illnesses, sleep disturbances, and unhealthy lifestyles (35). Our analysis, incorporating key confounders like vascular risk factors, co-morbidities, and sleep quality, strengthens previous findings by demonstrating the independent association of cognitive function with depression and anxiety.
Significantly higher PSQI scores (poorer sleep) were observed in groups with higher HAMD/HAM-A scores, which are consistent with prior research. The PSQI score remained significantly correlated with the HAMD and HAM-A scores after full adjustment. The bidirectional link between affective disorders and sleep disturbances is well-established. Nearly all patients with mental disorders have sleep disturbances. Patients with affective disorders accompanied by sleep abnormalities typically show more severe symptoms. Mechanisms linking poor sleep and mental disorders include autonomic nervous system dysregulation and abnormalities in brain regions among individuals with sleep disturbances involved in regulating emotions (1). Circadian rhythm disruption and chronic insomnia can negatively impact mood and increase the risk of anxiety and depression (10). Our study confirms that poor sleep quality is independently associated with affective disorders. Furthermore, the study results suggested that cognitive function and sleep influence emotions through both shared and distinct pathways. Thus, valuing cognitive function and improving sleep may be associated with a lower risk of affective disorders.
A strong relationship was previously confirmed between affective disorders and frailty, which agrees with the findings in the current study. Frailty scores differed significantly between HAMD/HAM-A groups (P < 0.01). The patients with affective disorders were more likely to have frailty, even after accounting for risk factors, such as age, gender, BMI, cigarette smoking, cognition, ADL, nutrition, and illnesses. This strong correlation was sustained after controlling for the main potential confounding factors. The mechanism underlying frailty and emotion disorders include accelerated biological aging, chronic inflammation, and shared risk factors, such as chronic illnesses, cognitive deficits, and malnutrition (14). Frail individuals are more likely to increase emotional disturbance due to functional impairment and social withdrawal (11). Given the above findings, the current study provides theoretical support for the view that effective management of frailty may be correlated with a slower progression of emotional health disturbances.
A higher BMI was associated with a lower likelihood of depression in the current study. The findings were in contrast to previous studies involving European populations in which obesity and overweight are correlated with a higher probability of depression (18, 36). The correlation between adiposity and depression in Europeans was explained by social factors, worsening physical health, and the potential appearance later in life (18). Our results align with evidence of a negative association between BMI and depression in East Asian populations (21), possibly reflecting cultural perspectives where higher body weight is sometimes perceived positively (“happy mind and fat body”). Differences in biological factors and cultural perspectives on obesity may explain the different effects observed herein. Nutritional risk (NRS-2002) was significantly associated with depression, which is consistent with the crucial role of nutrition in brain function and metabolic pathways affecting mental health (37). Nutrients, like vitamin B12, folate, iron, and essential fatty acids, are vital for neurotransmission, enzymatic activity, and cellular processes (38). The significant association between co-morbidities and anxiety aligns with prior research (39). Anxiety can heighten awareness of physical symptoms, reduce treatment adherence, and increase complications in chronic illness. Conversely, the symptom burden and functional limitations caused by illness can worsen anxiety.
Females, especially elderly females, exhibited a higher risk of affective disorders than males, which is consistent with numerous studies (40–42). Proposed explanations for the increased risk of affective disorders in females include gender differences in response to inflammation and stress, sexual differentiation of brain structures and the neuroimmune system, hormonal differences, social and cultural norms, different gender roles, and disadvantages and empowerment throughout life. However, the precise mechanisms underlying the gender differences in affective disorder risk require further elucidation. A systematic understanding of the impact of gender on affective disorders could benefit prevention and treatment strategies, warranting larger confirmatory studies.
This study explored modifiable risk factors for anxiety and depression in inpatients by integrating sociodemographic, lifestyle, clinical, and biological data. The resulting insights provide practical tools for bedside risk stratification and identify potential intervention targets, thereby aiding in clinical management. Nevertheless, several limitations need to be emphasized. First, due to the cross-sectional design of the study, the direction of the causal relationship could not be established from the present data set. The use of longitudinal or experimental designs in future research is critical for inferring causality. Second, regarding generalizability, all participants were recruited from one hospital. Although the sample reflects a typical inpatient profile, the findings may not be fully applicable to outpatients, community populations, or healthcare settings with markedly different demographics or resources. Future multi-center studies across diverse regions in China are warranted to enhance external validity. Finally, owing to limitations in measurement methods and variable selection, this study did not analyze some potentially influential factors, such as specific in-hospital treatments (e.g., surgical procedures). The impact on psychological outcomes could not be assessed and should be incorporated into future analyses.
Conclusion
This study analyzed the correlation between depression and anxiety with multiple risk factors, including sociodemographic, lifestyle, and clinical and biological domains in hospitalized patients. The analysis revealed specific correlation patterns. For example, a higher PSQI score (β = 0.435, P < 0.001) and FRAIL score (β = 0.276, P < 0.001) were positively correlated with greater anxiety severity. Other significant factors independently associated with increased symptom burden included lower BMI, multimorbidity, poorer cognitive function, and nutritional risk (all P < 0.05). These quantified associations underscore the necessity of early screening and preventive strategies focused on modifiable factors. Clinical support should be prioritized for higher-risk subgroups, such as women and older adults, who present with compounded vulnerabilities, including affective disorders, fatigue, and sleep disturbances. Interventions aimed at improving sleep, mitigating frailty and illness progression, preserving cognitive function, and promoting healthy lifestyles hold promise for an association with a reduced burden of mental disorders. Further research is needed to confirm and expand upon these findings.
Data availability statement
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving humans were approved by the Institutional Ethics Committee of the Fifth Affiliated Hospital of Sun Yat-sen University [No. ZDWY (2024); Lunzi No. (K169-1)]. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
We thank the hospital staff for their efforts in recruiting patients. We also appreciate the patient support provided by the hospital staff. We thank International Science Editing for editing this manuscript.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
ReferencesCorreiaATLLipinskaGRauchHGLForshawPERodenLCRaeDE.
Associations between sleep-related heart rate variability and both sleep and symptoms of depression and anxiety: A systematic review. Sleep Med. (2023) 101:106–17. doi: 10.1016/j.sleep.2022.10.018, PMID: 36370515PatelVChisholmDParikhRCharlsonFJDegenhardtLDuaT.
Addressing the burden of mental, neurological, and substance use disorders: key messages from Disease Control Priorities, 3rd edition. Lancet (London England). (2016) 387:1672–85. doi: 10.1016/S0140-6736(15)00390-6, PMID: 26454360TillerJWG.
Depression and anxiety. Med J Aust. (2013) 199:S28–31. doi: 10.5694/mja12.10628, PMID: 25370281VittenglJR.
Who pays the price for high neuroticism? Moderators of longitudinal risks for depression and anxiety. psychol Med. (2017) 47:1794–805. doi: 10.1017/S0033291717000253, PMID: 28193303ChauRKissaneDWDavisonTE.
Risk factors for depression in long-term care: A prospective observational cohort study. Clin Gerontologist. (2021) 44:112–25. doi: 10.1080/07317115.2019.1635548, PMID: 31264523LubeckaBLubeckiMKasperczykJJośko-OchojskaJPudloR.
Risk modifying factors of anxiety and depressive disorders, using the example of a population study in the Żywiec district. Int J Environ Res Public Health. (2021) 18:10248. doi: 10.3390/ijerph181910248, PMID: 34639549StruijsSYde JongPJJeronimusBFvan der DoesWRieseHSpinhovenP.
Psychological risk factors and the course of depression and anxiety disorders: A review of 15 years NESDA research. J Affect Disord. (2021) 295:1347–59. doi: 10.1016/j.jad.2021.08.086, PMID: 34706448YanRStrandlundKCiHHuangYZhangYZhangY.
Analysis of factors influencing anxiety and depression among hospitalized patients with chronic wounds. Adv Skin Wound Care. (2021) 34:638–44. doi: 10.1097/01.ASW.0000797948.13759.ba, PMID: 34807895Barrera-CaballeroSRomero-MorenoRDel Sequeros Pedroso-ChaparroMOlmosRVara-GarcíaCGallego-AlbertoL.
Stress, cognitive fusion and comorbid depressive and anxiety symptomatology in dementia caregivers. Psychol Aging. (2021) 36:667–76. doi: 10.1037/pag0000624, PMID: 34351188DingRDingPTianLKuangXHuangBLinC.
Associations between sleep duration, depression status, and cognitive function among Chinese elderly: A community-based study. J Affect Disord. (2024) 366:273–82. doi: 10.1016/j.jad.2024.08.200, PMID: 39218319Ní MhaoláinAMFanCWRomero-OrtunoRCoganLCunninghamCKennyR-A.
Frailty, depression, and anxiety in later life. Int Psychogeriatrics. (2012) 24:1265–74. doi: 10.1017/S1041610211002110, PMID: 22333477BennettSThomasAJ.
Depression and dementia: cause, consequence or coincidence? Maturitas. (2014) 79:184–90. doi: 10.1016/j.maturitas.2014.05.009, PMID: 24931304BorgesCEllisJGRuivo MarquesD.
The role of sleep effort as a mediator between anxiety and depression. psychol Rep. (2024) 127:2287–306. doi: 10.1177/00332941221149181, PMID: 36595381ChenJ-HLeiHWanY-FZhuX-CZengL-YTangH-X.
Frailty and psychiatric disorders: A bidirectional Mendelian randomization study. J Affect Disord. (2024) 356:346–55. doi: 10.1016/j.jad.2024.04.024, PMID: 38626809WangNZhangQLiPGuoLWuXTuQ.
Association between cognitive function and emotion, sleep, frailty, and nutrition in hospitalized patients. Brain Behav. (2024) 14:e70170. doi: 10.1002/brb3.70170, PMID: 39607290GonçalvesR. S. D. S. A.MacielÁ.CCRollandYVellasBde Souto BarretoP.
Frailty biomarkers under the perspective of geroscience: A narrative review. Ageing Res Rev. (2022) 81:101737. doi: 10.1016/j.arr.2022.101737, PMID: 36162706ShahnawazZReppermundSBrodatyHCrawfordJDDraperBTrollorJN.
Prevalence and characteristics of depression in mild cognitive impairment: the Sydney Memory and Ageing Study. Acta Psychiatrica Scandinavica. (2013) 127:394–402. doi: 10.1111/acps.12008, PMID: 22943523KarageorgiouVCasanovaFO’LoughlinJGreenHMcKinleyTJBowdenJ.
Body mass index and inflammation in depression and treatment-resistant depression: a Mendelian randomisation study. BMC Med. (2023) 21:355. doi: 10.1186/s12916-023-03001-7, PMID: 37710313KotsevaKWoodDDe BacquerDEUROASPIRE investigators.
Determinants of participation and risk factor control according to attendance in cardiac rehabilitation programmes in coronary patients in Europe: EUROASPIRE IV survey. Eur J Prev Cardiol. (2018) 25:1242–51. doi: 10.1177/2047487318781359, PMID: 29873511MucciNGiorgiGDe Pasquale CerattiSFiz-PérezJMucciFArcangeliG.
Anxiety, stress-related factors, and blood pressure in young adults. Front Psychol. (2016) 7:1682. doi: 10.3389/fpsyg.2016.01682, PMID: 27840615O’LoughlinJCasanovaFFairhurst-HunterZHughesABowdenJWatkinsER.
Mendelian randomisation study of body composition and depression in people of East Asian ancestry highlights potential setting-specific causality. BMC Med. (2023) 21:37. doi: 10.1186/s12916-023-02735-8, PMID: 36726144RohanKJRoughJNEvansMHoS-YMeyerhoffJRobertsLM.
A protocol for the Hamilton Rating Scale for Depression: Item scoring rules, Rater training, and outcome accuracy with data on its application in a clinical trial. J Affect Disord. (2016) 200:111–8. doi: 10.1016/j.jad.2016.01.051, PMID: 27130960MaierWBullerRPhilippMHeuserI.
The Hamilton Anxiety Scale: reliability, validity and sensitivity to change in anxiety and depressive disorders. J Affect Disord. (1988) 14:61–8. doi: 10.1016/0165-0327(88)90072-9, PMID: 2963053LuJLiDLiFZhouAWangFZuoX.
Montreal cognitive assessment in detecting cognitive impairment in Chinese elderly individuals: a population-based study. J Geriatric Psychiatry Neurol. (2011) 24:184–90. doi: 10.1177/0891988711422528, PMID: 22228824TuQ-YJinHDingB-RYangXLeiZ-HBaiS.
Reliability, validity, and optimal cutoff score of the montreal cognitive assessment (changsha version) in ischemic cerebrovascular disease patients of hunan province, China. Dementia Geriatric Cogn Disord Extra. (2013) 3:25–36. doi: 10.1159/000346845, PMID: 23637698BuysseDJReynoldsCFMonkTHBermanSRKupferDJ.
The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. (1989) 28:193–213. doi: 10.1016/0165-1781(89)90047-4, PMID: 2748771FriedLPTangenCMWalstonJNewmanABHirschCGottdienerJ.
Frailty in older adults: evidence for a phenotype. Journals Gerontology. Ser A Biol Sci Med Sci. (2001) 56:M146–156. doi: 10.1093/gerona/56.3.m146, PMID: 11253156SusantoMHubbardREGardinerPA.
Validity and responsiveness of the FRAIL scale in middle-aged women. J Am Med Directors Assoc. (2018) 19:65–9. doi: 10.1016/j.jamda.2017.08.003, PMID: 28927946KondrupJRasmussenHHHambergOStangaZAd Hoc ESPEN Working Group.
Nutritional risk screening (NRS 2002): a new method based on an analysis of controlled clinical trials. Clin Nutr (Edinburgh Scotland). (2003) 22:321–36. doi: 10.1016/s0261-5614(02)00214-5, PMID: 12765673CollinCWadeDTDaviesSHorneV.
The Barthel ADL Index: a reliability study. Int Disability Stud. (1988) 10:61–3. doi: 10.3109/09638288809164103, PMID: 3403500BrownPJSneedJRRutherfordBRDevanandDPRooseSP.
The nuances of cognition and depression in older adults: the need for a comprehensive assessment. Int J Geriatric Psychiatry. (2014) 29:506–14. doi: 10.1002/gps.4033, PMID: 24123357ZhaoHChenJ.
The prevalence and clinical correlation factors of cognitive impairment in patients with major depressive disorder hospitalized during the acute phase. Front Psychiatry. (2024) 15:1497658. doi: 10.3389/fpsyt.2024.1497658, PMID: 39619340ReppermundSSachdevPSCrawfordJKochanNASlavinMJKangK.
The relationship of neuropsychological function to instrumental activities of daily living in mild cognitive impairment. Int J Geriatric Psychiatry. (2011) 26:843–52. doi: 10.1002/gps.2612, PMID: 20845500JormAF.
Is depression a risk factor for dementia or cognitive decline? A review. Gerontology. (2000) 46:219–27. doi: 10.1159/000022163, PMID: 10859462DlugajMGerwigMWegeNSiegristJMannKBröcker-PreussM.
Elevated levels of high-sensitivity C-reactive protein are associated with mild cognitive impairment and its subtypes: results of a population-based case-control study. J Alzheimer’s disease: JAD. (2012) 28:503–14. doi: 10.3233/JAD-2011-111352, PMID: 22008268MalmirHMirzababaeiAMoradiSRezaeiSMirzaeiKDadfarmaA.
Metabolically healthy status and BMI in relation to depression: A systematic review of observational studies. Diabetes Metab Syndrome. (2019) 13:1099–103. doi: 10.1016/j.dsx.2019.01.027, PMID: 31336451YahyaNFSTengNIMFDasSJulianaN.
Nutrition and physical activity interventions to ameliorate postpartum depression: A scoping review. Asia Pacific J Clin Nutr. (2021) 30:662–74. doi: 10.6133/apjcn.202112_30(4).0013, PMID: 34967195Al-WardatMAlwardatNLou De SantisGZomparelliSGualtieriPBigioniG.
The association between serum vitamin D and mood disorders in a cohort of lipedema patients. Hormone Mol Biol Clin Invest. (2021) 42:351–5. doi: 10.1515/hmbci-2021-0027, PMID: 34323062KatonWLinEHBKroenkeK.
The association of depression and anxiety with medical symptom burden in patients with chronic medical illness. Gen Hosp Psychiatry. (2007) 29:147–55. doi: 10.1016/j.genhosppsych.2006.11.005, PMID: 17336664BekhbatMNeighGN.
Sex differences in the neuro-immune consequences of stress: Focus on depression and anxiety. Brain Behavior Immun. (2018) 67:1–12. doi: 10.1016/j.bbi.2017.02.006, PMID: 28216088FaravelliCAlessandra ScarpatoMCastelliniGLo SauroC.
Gender differences in depression and anxiety: the role of age. Psychiatry Res. (2013) 210:1301–3. doi: 10.1016/j.psychres.2013.09.027, PMID: 24135551ThibautF.
The role of sex and gender in neuropsychiatric disorders. Dialogues Clin Neurosci. (2016) 18:351–2. doi: 10.31887/DCNS.2016.18.4/fthibaut, PMID: 28179806
Edited by: Francesco Panza, University of Bari Aldo Moro, Italy
Reviewed by: Yan Liu, Peking Union Medical College Hospital (CAMS), China
Swathi Ramesh, SRM Institute of Science and Technology, India