Psychocardiological impact of depression on medication adherence, ventricular function, and readmission in heart failure: A retrospective cohort study

Abstract

BACKGROUND

Depression is a highly prevalent and clinically significant psychiatric comorbidity in patients with heart failure (HF), exerting multidimensional effects that extend beyond emotional symptoms to influence physiological outcomes and disease progression. Emerging evidence in psychocardiology highlights the bidirectional interplay between mood disorders and cardiovascular dysfunction through neuroendocrine, autonomic, and behavioral pathways. This study aims to explore the real-world effect of depression severity - measured by the Patient Health Questionnaire-9 (PHQ-9) - on left ventricular systolic function and one-year cardiovascular readmission in patients with HF, providing insights into its prognostic relevance within a psychocardiological framework.

AIM

To investigate the impact of depression severity on medication adherence, ventricular function, and readmission in patients with HF.

METHODS

A total of 160 patients hospitalized for HF between January 2020 and December 2023 were included in this real-world retrospective cohort study. Depression severity was assessed by using the PHQ-9, with scores ≥ 10 indicating moderate-to-severe depression. Cardiac function was evaluated through transthoracic echocardiography to determine left ventricular ejection fraction (LVEF). Medication adherence was assessed at three and six months postdischarge by employing the four-item Morisky Medication Adherence Scale (MMAS-4) and categorized as high (score = 0), moderate (1-2), or low (3-4). Data on antidepressant or anxiolytic prescriptions and psychological interventions during hospitalization were collected. Patients were followed up for one year to capture cardiovascular-related readmissions. Kaplan-Meier analysis was employed to estimate event-free survival, and Cox regression identified independent predictors of readmission.

RESULTS

Patients with moderate-to-severe depression (PHQ-9 ≥ 10) presented with significantly lower LVEF at baseline, higher N-terminal pro-B-type natriuretic peptide levels, and more severe HF symptoms than other patients. They also demonstrated poorer medication adherence postdischarge, with a higher proportion classified as low adherence on the MMAS-4 scale, and were less likely to receive β-blockers or angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker at discharge than other patients. At three and six months postdischarge, PHQ-9 and MMAS-4 scores were inversely correlated with LVEF, suggesting a behavioral pathway linking depression to impaired cardiac recovery. During the one-year follow-up period, 30.0% of patients experienced cardiovascular-related readmissions, predominantly due to worsening HF (54.1%). In multivariable Cox regression analysis, high PHQ-9 scores, reduced LVEF, old age, elevated N-terminal pro-B-type natriuretic peptide levels, New York Heart Association class IV, and absence of β-blocker therapy were independently associated with readmission risk.

CONCLUSION

In patients with HF, depression severity independently predicts impaired ventricular function, low medication adherence, and increased one-year cardiovascular readmission. These findings highlight the psychocardiological relevance of depression screening and behavioral intervention in optimizing adherence and clinical outcomes in routine HF care.

Key Words: Heart failure; Depression; Cardiac dysfunction; Cardiovascular readmission; Readmission; Psychocardiology

Core Tip: This real-world retrospective cohort study investigated the psychocardiological impact of depression on heart failure outcomes. Depression severity, measured by the Patient Health Questionnaire-9, was associated with poorer medication adherence, reduced left ventricular ejection fraction, and higher 1-year readmission risk. These findings underscore the need to incorporate psychosocial screening and interventions into routine heart failure management.

INTRODUCTION

Heart failure (HF) remains a leading cause of morbidity and mortality worldwide, affecting approximately 64 million people and accounting for substantial healthcare costs due to frequent hospitalizations and poor long-term survival[1]. Despite advances in pharmacologic and device-based therapies, 30%-50% of HF patients require rehospitalization within one year of discharge, highlighting the need for improved risk stratification and management strategies[2,3].

Depression is increasingly recognized as a significant yet underdiagnosed comorbidity in HF, with a prevalence ranging from 20% to 40%, exceeding that of the general population[4]. Evidence suggests that depression is associated with increased hospitalizations, worse medication adherence, reduced quality of life, and higher mortality in HF patients[5]. The pathophysiological mechanisms underlying this relationship remain incompletely understood but are thought to involve autonomic dysfunction, systemic inflammation, endothelial dysfunction, and dysregulation of the hypothalamic-pituitary-adrenal axis, all of which contribute to worsening cardiac function and adverse cardiovascular outcomes[5-7].

Although previous studies have established an association between depression and adverse HF outcomes, several critical knowledge gaps remain[8]. The direct impact of depression severity on cardiac function, particularly left ventricular ejection fraction (LVEF), has not been systematically quantified, limiting our understanding of its pathophysiologic significance[9]. While depression has been linked to increased hospitalization risk, its specific role in predicting cardiovascular event-related readmission remains poorly characterized, making it difficult to determine whether it independently contributes to recurrent cardiovascular events[10]. Furthermore, despite mounting evidence supporting the adverse impact of depression in HF, current HF guidelines do not integrate depression screening or management into routine care, underscoring the need for studies that establish its prognostic value and inform potential therapeutic interventions[11-14].

To address these gaps, this study investigates the association between depression severity. We used the Patient Health Questionnaire-9 (PHQ-9) to assess depression severity. Cardiac dysfunction [LVEF, New York Heart Association (NYHA) classification], and one-year cardiovascular event-related readmission in HF patients[15]. By integrating standardized depression assessment with objective cardiac function measures and long-term clinical outcomes, this study aims to provide a more comprehensive understanding of the prognostic significance of depression in HF, which may help refine risk stratification and guide future HF management strategies.

MATERIALS AND METHODS

Study design

This was a retrospective cohort study conducted at a tertiary academic hospital specializing in cardiovascular care. The primary objective was to examine the association between depression severity and cardiac dysfunction in patients hospitalized for HF and to evaluate the prognostic significance of depressive symptoms on one-year cardiovascular event-related readmission. The study adhered to the principles outlined in the Declaration of Helsinki and followed the STrengthening the Reporting of OBservational studies in Epidemiology guidelines for reporting.

The study protocol was reviewed and approved by the Institutional Review Board of The First People’s Hospital of Yunnan Province (Approval No. 2025-YNFPH-006), which granted a waiver of informed consent due to the use of anonymized retrospective data and minimal risk to participants. Data were collected from routinely recorded electronic medical records, ensuring real-world relevance and clinical applicability.

Study population and data source

We retrospectively reviewed electronic medical records of patients hospitalized for HF between January 2020 and December 2023. Patients were included if they met European Society of Cardiology/American Heart Association HF diagnostic criteria, confirmed by clinical evaluation and echocardiographic evidence, and had completed a PHQ-9 depression assessment at admission. Only patients with at least one documented LVEF measurement during hospitalization and a minimum of one-year follow-up data for cardiovascular outcomes were included.

Exclusion criteria included end-stage renal disease, active malignancy, severe hepatic dysfunction, recent major cardiac surgery (e.g., coronary artery bypass grafting or valve replacement within the past three months), pre-existing psychiatric disorders other than depression, long-term psychiatric medication use, incomplete medical records, or loss to follow-up.

Data collection and variables

Demographic, clinical, and medication data were extracted from electronic medical records. Baseline characteristics included age, sex, body mass index, NYHA classification, prior cardiovascular history (coronary artery bypass grafting, myocardial infarction), and HF etiology (ischemic vs non-ischemic). Depression severity was assessed using PHQ-9, with no/mild depression defined as PHQ-9 ≤ 9 and moderate/severe depression as PHQ-9 ≥ 10. Cardiac function was evaluated using LVEF, with LVEF ≤ 20% considered severe systolic dysfunction. Discharge medications included β-blockers, angiotensin-converting enzyme (ACE) inhibitors, diuretics, nitrates, antidepressants, and anxiolytics.

The primary outcome was cardiovascular event-related readmission within one year, including worsening HF, acute coronary syndrome (myocardial infarction or unstable angina), arrhythmia-related hospitalization, sudden cardiac death, and other cardiovascular complications. The secondary outcomes included event-free survival (EFS), time to first cardiovascular event, and independent predictors of readmission.

Assessment of medication adherence

Medication adherence was assessed using the 4-item Morisky Medication Adherence Scale (MMAS-4) at both 3 months and 6 months post-discharge. The scale consists of four yes/no questions addressing forgetfulness, discontinuation, and regimen complexity. Each “yes” response scores 1 point, yielding a total score range of 0 to 4, with higher scores indicating poorer adherence. Patients were classified into high (score = 0), moderate (score = 1-2), or low adherence (score = 3-4) groups. MMAS-4 was analyzed both as a categorical variable and as a continuous score in correlation and regression analyses.

Follow-up and outcome assessment

All patients were followed prospectively for 12 months from the date of hospital discharge. Follow-up was conducted through a review of electronic medical records and scheduled outpatient visits at 3 months, 6 months, and 12 months. Additional information was obtained via structured telephone interviews when necessary to ensure completeness of outcome data. The primary outcome was cardiovascular event-related readmission, defined as hospitalization due to worsening HF, acute coronary syndrome, arrhythmia, sudden cardiac death, or other major cardiovascular complications. Each readmission event was independently adjudicated and classified by two board-certified cardiologists blinded to patients’ depression status, with discrepancies resolved by consensus.

Secondary outcomes included time to first cardiovascular event and event-free survival. Patients who died during follow-up without prior readmission were considered to have experienced the primary endpoint. Those lost to follow-up or with incomplete outcome data were excluded from the final analysis.

Statistical considerations

A power analysis was conducted to ensure an adequate sample size for detecting significant differences in cardiovascular readmission rates between depression severity groups. The final cohort included 160 patients, providing sufficient power for statistical analysis. Missing data were assessed, and appropriate statistical methods, including multiple imputation, were applied where necessary.

Despite the inherent limitations of a retrospective cohort study, including selection bias, missing data, and unmeasured confounders, the use of strict inclusion criteria, standardized data collection, and robust statistical methodologies enhance the validity and reliability of the findings.

Statistical analysis

All statistical analyses were performed using R software (version 4.2.1, R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were expressed as mean ± SD or median with interquartile range (IQR), and categorical variables as frequencies and percentages. Between-group comparisons were conducted using the Student’s t-test or Mann-Whitney U test for continuous variables, and the χ² test or Fisher’s exact test for categorical variables, as appropriate.

Kaplan-Meier survival analysis was employed to estimate EFS, with differences assessed using the log-rank test. Univariable Cox proportional hazards models were used to identify potential predictors of one-year cardiovascular event-related readmission. Variables with a P value < 0.10 in univariable analysis were entered into the multivariable Cox regression model using a forward stepwise approach. Hazard ratios (HRs) with 95% confidence intervals (CIs) were reported.

The proportional hazards assumption was verified using Schoenfeld residuals. Missing data were handled using multiple imputations by chained equations under the assumption of missing at random, generating five imputed datasets with pooled estimates. Statistical significance was defined as a two-sided P-value < 0.05.

RESULTS

Depressive symptom severity is associated with worse cardiac profile and reduced neurocardiological protection

A total of 160 patients with HF were included, with a mean age of 75 ± 9 years. Patients with moderate or severe depression were significantly older than those with no or mild depression (78 ± 9 years vs 73 ± 5 years, P = 0.045), and less likely to be male (23% vs 40%, P = 0.021). Heart rate was higher in the depressed group (86 ± 15 bpm vs 80 ± 13 bpm, P = 0.041), whereas body mass index and blood pressure values were comparable between groups. Patients with moderate/severe depression showed more advanced HF features, including a higher prevalence of NYHA class IV (12% vs 8%, P = 0.039), lower mean LVEF (26 ± 6% vs 34 ± 7%, P = 0.031), and a greater proportion with LVEF ≤ 20% (P = 0.042). Ischemic etiology was significantly less common in the depressed group (15% vs 45%, P = 0.018). Comorbid conditions such as coronary artery disease (14% vs 38%, P = 0.026), prior myocardial infarction (11% vs 27%, P = 0.048), hypertension (17% vs 44%, P = 0.035), and hyperlipidemia (10% vs 26%, P = 0.043) were less prevalent in the depressed cohort. No differences were observed in diabetes prevalence. Patients with moderate/severe depression were more likely to have lower educational attainment (17% vs 28%, P = 0.038) and to be divorced or separated (12% vs 10%, P = 0.042). While N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentrations were significantly higher in the depressed group (median 3900 pg/mL vs 2700 pg/mL, P = 0.038), there were no group differences in C-reactive protein, hemoglobin, serum albumin, creatinine, or hospital length of stay. The use of psychotherapy was more frequent in the moderate/severe depression group (22% vs 4%, P = 0.011). At discharge, patients with moderate/severe depression were significantly less likely to receive β-blockers (22% vs 43%, P = 0.027) or ACE inhibitors (22% vs 48%, P = 0.023), but more likely to be prescribed antidepressants (12% vs 8%, P = 0.036) and anxiolytics (13% vs 5%, P = 0.046; Table 1).

Table 1 Baseline characteristics of heart failure patients with depression, mean ± SD, % (n/total).

Variables	Total (n = 160)	No/mild depression (n = 100)	Moderate/severe depression (n = 60)	Test statistic	P value
Age, years	75 ± 9	73 ± 5	78 ± 9	t = 2.04	0.045
Male	63.1 (101/160)	40.0 (40/100)	23.3 (14/60)	χ2 = 5.3	0.021
BMI, kg/m2	25 ± 3	25 ± 3	24 ± 3	t = 0.99	0.322
Heart rate, bpm	82 ± 14	80 ± 13	86 ± 15	t = 2.06	0.041
Systolic bp, mmHg	118 ± 16	120 ± 15	115 ± 17	t = 1.84	0.067
Diastolic bp, mmHg	71 ± 10	72 ± 9	69 ± 11	t = 1.56	0.119
NYHA class II	23.8 (38/160)	31.0 (31/100)	11.7 (7/60)	χ2 = 4.6	0.032
NYHA class III	19.4 (31/160)	22.0 (22/100)	15.0 (9/60)	χ2 = 1.5	0.217
NYHA class IV	9.4 (15/160)	8.0 (8/100)	11.7 (7/60)	χ2 = 4.3	0.039
Coronary artery disease	51.9 (83/160)	38.0 (38/100)	13.3 (8/60)	χ2 = 5.0	0.026
Prior myocardial infarction	38.1 (61/160)	27.0 (27/100)	11.7 (7/60)	χ2 = 3.9	0.048
Diabetes mellitus	40.0 (64/160)	25.0 (25/100)	15.0 (9/60)	χ2 = 3.2	0.072
Hypertension	61.3 (98/160)	44.0 (44/100)	16.7 (10/60)	χ2 = 4.4	0.035
Hyperlipidemia	36.3 (58/160)	26.0 (26/100)	10.0 (6/60)	χ2 = 4.1	0.043
Low education level	45.0 (72/160)	28.0 (28/100)	16.7 (10/60)	χ2 = 3.8	0.038
Divorced/separated	21.9 (35/160)	10.0 (10/100)	11.7 (7/60)	χ2 = 4.2	0.042
Etiology of CHF: Ischemic	60.0 (96/160)	45.0 (45/100)	15.0 (9/60)	χ2 = 5.7	0.018

BMI: Body mass index; NYHA: New York Heart Association; CHF: Congestive heart failure.

Medication use and adherence by depression severity

Patients with moderate to severe depression exhibited distinct pharmacological and behavioral treatment patterns compared to those with no or mild depressive symptoms. Prescription rates of β-blockers (53.3% vs 68.0%, χ² = 3.5, P = 0.061), ACE inhibitors/angiotensin II receptor blocker (56.7% vs 70.0%, χ² = 3.7, P = 0.054), and diuretics (61.7% vs 75.0%, χ² = 3.0, P = 0.084) were numerically lower among patients with higher depressive burden, although differences approached but did not reach conventional statistical significance. Conversely, mental health interventions were more commonly utilized in the depressed subgroup, with higher proportions receiving antidepressants (16.7% vs 9.0%, χ² = 2.6, P = 0.107), anxiolytics (13.3% vs 7.0%, χ² = 2.9, P = 0.091), and psychotherapy (18.3% vs 6.0%, χ² = 4.1, P = 0.043), the latter reaching statistical significance.

Adherence to HF medications, as assessed by the MMAS-4 scale, differed meaningfully across groups. Patients with moderate to severe depression were significantly more likely to report low adherence (35.0% vs 22.0%, χ² = 3.9, P = 0.048) and less likely to report high adherence (25.0% vs 40.0%, χ² = 4.6, P = 0.031). Moderate adherence rates were similar between groups (40.0% vs 38.0%, χ² = 0.1, P = 0.794; Table 2).

Table 2 Medication use and adherence stratified by depression severity, % (n/total).

Variables	No/mild depression (n = 100)	Moderate/severe depression (n = 60)	Test statistic	P value
β-blocker use	68.0 (68/100)	53.3 (32/60)	χ2 = 3.5	0.061
ACE inhibitor/ARB use	70.0 (70/100)	56.7 (34/60)	χ2 = 3.7	0.054
Diuretic use	75.0 (75/100)	61.7 (37/60)	χ2 = 3.0	0.084
Antidepressant use	9.0 (9/100)	16.7 (10/60)	χ2 = 2.6	0.107
Anxiolytic use	7.0 (7/100)	13.3 (8/60)	χ2 = 2.9	0.091
Psychotherapy	6.0 (6/100)	18.3 (11/60)	χ2 = 4.1	0.043
High adherence (MMAS-4 = 0)	40.0 (40/100)	25.0 (15/60)	χ2 = 4.6	0.031
Moderate adherence (MMAS-4 = 1-2)	38.0 (38/100)	40.0 (24/60)	χ2 = 0.1	0.794
Low adherence (MMAS-4 = 3-4)	22.0 (22/100)	35.0 (21/60)	χ2 = 3.9	0.048

ACE: Angiotensin-converting enzyme; ARB: Angiotensin II receptor blocker; MMAS-4: Morisky Medication Adherence Scale.

Association of depression and medication adherence with cardiac function at follow-up

At 3 months and 6 months post-discharge, both depressive symptom severity and medication adherence were significantly associated with left ventricular systolic function. Baseline PHQ-9 scores were modestly but consistently inversely correlated with LVEF at 3 months (r = -0.26, P = 0.008) and 6 months (r = -0.24, P = 0.009), indicating that patients with higher depressive burden exhibited delayed or impaired cardiac recovery. Similarly, higher MMAS-4 scores, reflecting poorer adherence, were independently linked to lower LVEF at both timepoints (r = -0.23 at 3 months, P = 0.012; r = -0.22 at 6 months, P = 0.014).

To explore potential confounding, a sensitivity analysis restricted to patients with moderate-to-high adherence (MMAS-4 < 3) was performed. Within this subgroup, depression severity remained inversely correlated with LVEF at 6 months (r = -0.22, P = 0.034), suggesting a possible behavioral-independent effect of depression on ventricular function (Supplementary Table 1). Among patients with PHQ-9 ≥ 10, those prescribed antidepressants at discharge (n = 24) showed modestly improved adherence at 3 months (mean MMAS-4 score 1.9 vs 2.5; mean difference of -0.6, P = 0.041) compared with untreated counterparts. However, no significant difference in LVEF was observed at 6 months between treated and untreated groups (29.3% vs 28.7%, P = 0.372), implying that while antidepressant use may enhance short-term adherence, its direct impact on cardiac function recovery is limited (Supplementary Table 2).

Depression linked to early cardiovascular readmissions

During the 12-month follow-up period, 48 out of 160 patients (30.0%) experienced at least one cardiovascular event-related readmission (Table 3). The median time to the first cardiovascular event was 124 days (IQR: 93.5-150.5 days), with more than half of the events occurring within the first four months after discharge (Figure 1), highlighting an early vulnerable period for cardiovascular deterioration.

Figure 1 Association between depression severity, medication adherence, and cardiac function at 3-month and 6-month follow-up. A: Scatter plot showing the correlation between Patient Health Questionnaire-9 score and left ventricular ejection fraction (LVEF) at 3 months post-discharge. Higher depression severity was modestly associated with lower LVEF (Pearson r = -0.26, P = 0.008); B: Scatter plot showing the correlation between Morisky Medication Adherence Scale score (medication adherence) and LVEF at 3 months. Poorer adherence was associated with lower LVEF (Pearson r = -0.23, P = 0.012); C: Correlation between Patient Health Questionnaire-9 score and LVEF at 6 months. The inverse association persisted over time (Pearson r = -0.24, P = 0.009); D: Correlation between Morisky Medication Adherence Scale score and LVEF at 6 months. Self-reported nonadherence remained linked to reduced cardiac function (Pearson r = -0.22, P = 0.014). PHQ-9: Patient Health Questionnaire-9; LVEF: Left ventricular ejection fraction; MMA-S: Morisky Medication Adherence Scale.

Table 3 Clinical indicators and treatment of heart failure patients with depression, mean ± SD/% (n/total).

Variables	Total (n = 160)	No/mild depression (n = 100)	Moderate/severe depression (n = 60)	Test statistic	P value
LVEF	30 ± 8	34 ± 7	26 ± 6	t = 2.2	0.031
LVEF ≤ 20%	18.1 (29/160)	10.0 (10/100)	31.7 (19/60)	χ2 = 4.1	0.042
PHQ-9 score	7 ± 3	4 ± 2	12 ± 4	t = 6.2	< 0.001
NT-proBNP, pg/mL, median (interquartile range)	3000 (1800-4700)	2700 (1600-4100)	3900 (2100-5400)	Z = 2.1	0.038
CRP, mg/L	8.5 ± 3.2	7.9 ± 3.1	9.6 ± 3.3	t = 1.7	0.084
Hemoglobing, dL	11.2 ± 1.5	11.4 ± 1.3	10.7 ± 1.6	t = 1.8	0.076
Serum albumin, g/L	38 ± 4	39 ± 3	36 ± 4	t = 1.9	0.058
Serum creatinine, μmol/L	109 ± 36	102 ± 34	121 ± 38	t = 1.9	0.062
Length of hospital stay, days, median (interquartile range)	9 (6-13)	8 (5-11)	11 (7-15)	Z = 1.8	0.071
Use of psychotherapy	10.0 (16/160)	4.0 (4/100)	20.0 (12/60)	χ2 = 6.4	0.011
Worsening CHF	45.0 (72/160)	28.0 (28/100)	73.3 (44/60)	χ2 = 4.8	0.029
Myocardial infarction	20.0 (32/160)	14.0 (14/100)	30.0 (18/60)	χ2 = 2.9	0.086
Unstable angina	15.0 (24/160)	9.0 (9/100)	25.0 (15/60)	χ2 = 1.1	0.294
Arrhythmia	11.9 (19/160)	7.0 (7/100)	20.0 (12/60)	χ2 = 0.8	0.371
Other admission	8.1 (13/160)	5.0 (5/100)	13.3 (8/60)	χ2 = 0.5	0.478
β-blockers at discharge	65.0 (104/160)	43.0 (43/100)	21.7 (13/60)	χ2 = 5.0	0.027
Calcium blockers	30.0 (48/160)	20.0 (20/100)	10.0 (6/60)	χ2 = 3.4	0.068
ACE inhibitors	70.0 (112/160)	48.0 (48/100)	21.7 (13/60)	χ2 = 5.2	0.023
Diuretics	75.0 (120/160)	50.0 (50/100)	25.0 (15/60)	χ2 = 5.6	0.019
Nitrates	50.0 (80/160)	35.0 (35/100)	15.0 (9/60)	χ2 = 4.2	0.041
Antidepressants	20.0 (32/160)	8.0 (8/100)	11.7 (7/60)	χ2 = 4.4	0.036
Anxiolytics	18.1 (29/160)	5.0 (5/100)	13.3 (8/60)	χ2 = 3.9	0.046

LVEF: Left ventricular ejection fraction; PHQ-9: Patient Health Questionnaire-9; CRP: C-reactive protein; CHF: Congestive heart failure; ACE: Angiotensin-converting enzyme.

The distribution and timing of cardiovascular event types are summarized in Figure 2 and Table 4. Worsening heart failure was the most frequent cause of readmission, observed in 26 patients (54.1%), followed by acute coronary syndrome - including myocardial infarction and unstable angina - in 9 cases (18.8%) and arrhythmia-related hospitalization in 7 patients (14.6%). Additionally, 4 patients (8.3%) were readmitted due to sudden cardiac death or other major adverse cardiovascular events, while 2 patients (4.2%) were readmitted for other cardiovascular conditions such as hypertensive crisis and pulmonary embolism. No significant differences in event-type distribution were observed between male and female patients or across depression-severity strata.

Figure 2 Distribution of cardiovascular event readmissions in 1-year follow-up. This pie chart illustrates the proportion of different cardiovascular-related causes of hospital readmission (n = 48 events) among heart failure patients during the 12-month follow-up. The most frequent cause was worsening heart failure (54.1%), followed by acute coronary syndrome (myocardial infarction + unstable angina; 18.8%), arrhythmia-related hospitalization (14.6%), sudden cardiac death or major adverse cardiovascular events (8.3%), and other cardiovascular causes (4.2%). MACE: Major adverse cardiovascular events; HF: Heart failure; MI: Myocardial infarction; UA: Unstable angina.

Table 4 Cardiovascular event readmissions in 1-year follow-up with median time to event, median (interquartile range).

Cardiovascular event type	Number of patients (n = 48)	Proportion of readmitted patients (%)	Median time to event (days)
Worsening HF	26	54.2	90 (60-150)
Acute coronary syndrome (MI + UA)	9	18.8	120 (90-180)
Arrhythmia-related hospitalization	7	14.6	150 (120-210)
Sudden cardiac death/mace	4	8.3	210 (180-300)
Other cardiovascular causes	2	4.2	180 (150-270)

HF: Heart failure; MI: Myocardial infarction; UA: Unstable angina.

Event-free survival and cardiovascular readmission patterns

Kaplan-Meier survival analysis demonstrated a median EFS of 124 days (IQR: 93.5-150.5 days), with 50% of cardiovascular events occurring within the first four months post-discharge. By the end of one year, the cumulative event-free survival rate was 70.0%, indicating that most patients did not require cardiovascular readmission during follow-up (Figure 3A).

Figure 3 Cardiovascular event-free survival and predictors of readmission. A: Kaplan-Meier survival curve for cardiovascular event-free survival during 1-year follow-up. The Kaplan-Meier plot shows the cumulative probability of remaining free from cardiovascular-related readmission or adverse events over a 365-day follow-up period. The steepest decline in survival probability occurred within the first 200 days, indicating a high-risk period for cardiovascular deterioration after hospital discharge; B: Multivariable Cox regression analysis identifying predictors of cardiovascular-related readmission. This forest plot presents hazard ratios and 95% confidence intervals for independent predictors of cardiovascular readmission at 12 months post-discharge. Significant predictors include higher Patient Health Questionnaire-9 score, reduced left ventricular ejection fraction, absence of β-blocker therapy, older age, elevated N-terminal pro-B-type natriuretic peptide levels, and advanced New York Heart Association class IV status. Psychotherapy and worsening congestive heart failure at admission showed trends toward risk modification but did not reach statistical significance. LVEF: Left ventricular ejection fraction; PHQ-9: Patient Health Questionnaire-9; NT-proBNP: N-terminal pro-B-type natriuretic peptide; NYHA: New York Heart Association; CHF: Congestive heart failure.

Depression severity emerges as an independent psychocardiological predictor of readmission risk

In univariable analysis, one-year cardiovascular readmission was significantly associated with older age (P = 0.042), lower LVEF (P = 0.006), elevated PHQ-9 scores (P < 0.001), higher NT-proBNP levels (P = 0.041), and absence of β-blocker therapy (P = 0.027). Clinical severity indicators at index admission, including NYHA class IV (P = 0.049) and acute decompensation (P = 0.041), were also more common among readmitted patients (Table 5). Multivariable Cox regression identified baseline depressive symptom severity (per-point increase in PHQ-9; HR = 1.059, 95%CI: 1.025-1.094, P < 0.001) as an independent predictor of readmission, along with lower LVEF (HR = 0.969, 95%CI: 0.947-0.991, P = 0.009), older age (HR = 1.028, 95%CI: 1.005-1.052, P = 0.016), and elevated NT-proBNP (HR = 1.018, 95%CI: 1.003-1.033, P = 0.021). NYHA class IV and index hospitalization for worsening HF were also independently associated with higher risk, whereas β-blocker therapy remained protective (HR = 0.732, P = 0.024; Figure 3B and Table 6).

Table 5 Univariable analysis of risk factors for 1-year cardiovascular event readmission, mean ± SD/%.

Variable	Event group (n = 48)	No event group (n = 112)	P value
Age, years	77 ± 8	74 ± 9	0.042
Male	50	66	0.067
Heart rate, bpm	84 ± 14	81 ± 14	0.219
NYHA class IV	17	14	0.049
Coronary artery disease	61	52	0.284
Prior MI	46	35	0.191
Hypertension	58	63	0.548
Hyperlipidemia	31	37	0.441
Low education level	42	32	0.228
Divorced/separated	19	18	0.889
Ischemic etiology	52	65	0.113
LVEF	27 ± 7	32 ± 8	0.006
LVEF ≤ 20%	27	12	0.023
PHQ-9 score	9 ± 4	6 ± 3	< 0.001
NT-proBNP, median (interquartile range), pg/mL	3950 (2300-5600)	2900 (1600-4300)	0.041
Use of psychotherapy	17	8	0.086
β-blocker use at discharge	48	72	0.027
ACE inhibitor use	60	74	0.091
Diuretic use	65	78	0.084
Nitrate use	46	52	0.486
Antidepressant use	10	12	0.723
Anxiolytic use	15	9	0.237
Worsening CHF as admission	63	41	0.041

NYHA: New York Heart Association; MI: Myocardial infarction; LVEF: Left ventricular ejection fraction; PHQ-9: Patient Health Questionnaire-9; NT-proBNP: N-terminal pro-B-type natriuretic peptide; ACE: Angiotensin-converting enzyme; CHF: Congestive heart failure.

Table 6 Multivariable Cox regression analysis of independent predictors for 1-year cardiovascular event readmission.

Variable	Hazard ratio	95% confidence interval	P value
Age	1.028	1.005-1.052	0.016
LVEF	0.969	0.947-0.991	0.009
PHQ-9 score	1.059	1.025-1.094	< 0.001
β-blocker use	0.732	0.562-0.953	0.024
NT-proBNP	1.018	1.003-1.033	0.021
Psychotherapy	1.447	0.912-2.296	0.112
NYHA class IV	1.611	1.102-2.354	0.037
Worsening CHF at admission	1.723	1.103-2.694	0.031

LVEF: Left ventricular ejection fraction; PHQ-9: Patient Health Questionnaire-9; NT-proBNP: N-terminal pro-B-type natriuretic peptide; NYHA: New York Heart Association; CHF: Congestive heart failure.

Stratified analyses by HF phenotype revealed that the prognostic value of PHQ-9 was more pronounced among patients with reduced ejection fraction (HF with reduced ejection fraction, LVEF < 50%). In this group (n = 112), depression severity remained significantly associated with readmission risk (HR = 1.068, 95%CI: 1.027 to 1.111, P = 0.001). Among those with preserved ejection fraction (HF with preserved ejection fraction, LVEF ≥ 50%, n = 48), the association was attenuated and not statistically significant (HR = 1.021, 95%CI: 0.994-1.056, P = 0.097), suggesting phenotype-specific differences in psychocardiological vulnerability (Supplementary Table 3).

Additional subgroup analysis confirmed that patients with PHQ-9 ≥ 10 were older, more frequently female, and more likely to be readmitted than non-depressed counterparts. These trends align with established epidemiology and underscore the relevance of targeted psychosocial risk stratification in HF care (Supplementary Table 4).

DISCUSSION

In this retrospective cohort study of patients with HF, we identified a remarkable and independent association between depressive symptom severity and impaired cardiac function and an increased risk of one-year cardiovascular event-related readmission. These findings not only reinforce the growing body of evidence linking depression to adverse cardiovascular outcomes but also highlight the prognostic utility of PHQ-9 as a simple, scalable tool for risk stratification in routine clinical care.

Emerging evidence supports the notion that depression in HF is not merely a prognostic marker but a potentially modifiable risk factor. Randomized controlled trials in other chronic disease populations have shown that depression treatment - whether pharmacologic, psychotherapeutic, or collaborative care - can improve both mental health and clinical outcomes[16-18]. Although data in HF populations remain limited, recent studies have demonstrated that targeted management of depressive symptoms may improve self-care, medication adherence, and even cardiac function indices. Given our finding that depression severity independently predicted readmission, incorporating evidence-based depression treatment into post-discharge HF management protocols warrants further exploration in prospective trials.

Consistent with previous research, we observed a high prevalence of moderate to severe depressive symptoms, particularly among older and female patients - groups historically associated with greater vulnerability to psychological stress and adverse HF trajectories[13,14,19,20]. Importantly, our study adds to the existing literature by demonstrating that higher PHQ-9 scores are independently associated with lower LVEF, advanced NYHA functional class, and elevated NT-proBNP levels[21-23]. These findings suggest a pathophysiological interplay wherein depression may exacerbate neurohumoral dysregulation, sympathetic activation, and systemic inflammation - mechanisms previously implicated in HF progression[16-18].

Our results extend the observations of prior large-scale studies such as the SADHART-congestive heart failure trial and the OPTIMIZE-HF registry, which linked depression to increased mortality and hospitalizations in HF, yet did not quantify risk using standardized depression metrics[24]. By identifying that each one-point increase in PHQ-9 score confers a nearly 6% increase in readmission risk - independent of LVEF, NT-proBNP, and age - we provide clinically actionable data that could inform early post-discharge surveillance strategies[25-27]. Moreover, the clustering of readmissions within the first 4 months underscores the importance of this vulnerable period for targeted psychosocial and pharmacologic interventions[28].

We also observed that patients with moderate/severe depression were significantly less likely to receive β-blockers or ACE inhibitors at discharge - therapies that remain the cornerstone of guideline-directed HF management[29]. This therapeutic gap may reflect a complex interplay of patient non-adherence, physician inertia, or perceived contraindications due to mental health status[30]. The protective effect of β-blocker use observed in our analysis suggests that enhancing medication adherence and eliminating barriers to optimal pharmacotherapy should be prioritized in this subgroup[25,27,28,31,32].

The observed association between depression severity and markers of cardiac dysfunction, such as reduced LVEF and elevated NT-proBNP levels, may reflect underlying pathophysiological processes shared by both conditions. Depression has been linked to autonomic nervous system dysregulation, resulting in heightened sympathetic activity and reduced heart rate variability, phenomena that are both associated with poor HF prognosis. Additionally, proinflammatory cytokines, hypothalamic pituitary adrenal axis hyperactivation, and endothelial dysfunction have been implicated in depression and HF progression. These shared pathways may create a bidirectional feedback loop; wherein psychological distress exacerbates cardiac dysfunction and worsening cardiac status amplifies depressive symptoms. Understanding these interactions could inform integrated treatment strategies that address psychological and physiological domains.

In addition to pharmacologic and psychiatric therapies, evidence-based non-pharmacologic interventions - such as structured telemonitoring, behavioral activation programs, and collaborative home-based care - have demonstrated efficacy in reducing hospital readmissions among HF patients with comorbid depression. It has been PHQ-9 scores were assessed only once at hospital admission reported that telehealth monitoring, when integrated with psychosocial support, significantly decreased 30-day readmission rates in depressed HF populations[29]. These findings support the implementation of comprehensive, multidisciplinary psychocardiological care models that extend beyond hospital discharge and encompass remote behavioral surveillance and timely intervention.

Our findings have several important clinical implications. First, they advocate for the systematic integration of depression screening - specifically PHQ-9 - into HF discharge planning and outpatient follow-up. Second, they support the need for collaborative care models that combine cardiology, psychiatry, and behavioral health interventions. Third, they suggest that existing HF guidelines may benefit from the inclusion of mental health considerations, particularly because comorbid depression appears to confer risk comparable to traditional hemodynamic predictors, such as LVEF or NT-proBNP levels[32]. Fourth, medication adherence was measured using the MMAS-4, a self-reported instrument that may be subject to recall and social desirability biases. While the MMAS-4 has been validated across diverse populations, it lacks objective confirmation through pharmacy refill records, electronic pill monitoring, or biomarker tracking. Future studies should incorporate multi-modal adherence assessment to strengthen the accuracy and validity of behavioral correlates.

Nonetheless, this study has limitations. Its retrospective, single-center design may limit generalizability, and residual confounding cannot be excluded despite multivariable adjustment. Depression severity was assessed only at index hospitalization, without accounting for symptom evolution over time[31]. Additionally, the exclusion of patients with psychiatric comorbidities other than depression may underestimate the true burden of mental health in HF populations.

CONCLUSION

This retrospective cohort study demonstrates that depression in HF patients is significantly associated with poor medication adherence, reduced LVEF, and increased risk of 1-year readmission. These findings highlight the critical psychocardiological interplay between mental and cardiac health. Early screening and targeted psychosocial interventions for depression should be incorporated into the standard management of HF to improve long-term clinical outcomes.