ABSTRACT
Introduction
There is a potential link between Metabolic Dysfunction‐Associated Steatotic Liver Disease (MASLD) and depression in later life. This study aims to examine whether MASLD is associated with elevated clinically relevant depressive symptoms in community‐dwelling older adults.
Methods
This study utilised longitudinal data from a prospective cohort to quantify the association between MASLD at baseline and depressive symptom trajectories. MASLD, introduced by multinational liver societies to replace Non‐Alcoholic Fatty Liver Disease (NAFLD), highlights the central role of metabolic dysfunction in the pathogenesis of steatotic liver disease. Depression trajectories were estimated across a median follow‐up of 4.6 years (IQR: 0.1–7.1 years) and were identified as four distinct patterns of symptoms: ‘non‐depressed’, ‘subthreshold depression’, ‘persistent depression’, and ‘emerging depression’. Multivariable multinomial logistic regressions were performed to examine the association between baseline MASLD presence and membership of the depression trajectories, adjusting for sociodemographic and lifestyle factors, anthropometric indices, cognitive function, polypharmacy and the number of morbidities. Additional subgroup analyses were also performed.
Results
Of 9097 individuals (mean age 75.1 ± 4.2 years; 55.0% males), 2998 (33%) had MASLD at baseline. Participants with MASLD were significantly more likely to belong to the persistent depression trajectory (RRR = 1.13; 95% CI: 1.02–1.24) and the subthreshold depression trajectory (1.43; 1.22–1.67) than those without MASLD.
Conclusion
This study shows that the presence of MASLD is significantly associated with an increased risk of a worse depression trajectory. In subgroup analysis, this relationship was particularly pronounced amongst females and among individuals with key cardiometabolic morbidities such as dyslipidaemia.
Keywords: ageing, depression, fatty liver, mental health, metabolic dysfunction‐associated steatotic liver disease, metabolic syndrome, psychiatry
Highlights
This study used longitudinal data from a prospective cohort to quantify the association between MASLD at baseline and depressive symptom trajectories were estimated across a median follow‐up of 4.6 years (IQR: 0.1–7.1 years).
MASLD is prospectively associated with the risk of depression among community‐dwelling older adults.
The association between MASLD and worse depression outcomes was particularly evident among females and those with dyslipidaemia.
1. Introduction
Metabolic Dysfunction‐Associated Steatotic Liver Disease (MASLD) was proposed by multinational liver societies in 2023 as an alternative term to Non‐alcoholic Fatty Liver Disease (NAFLD), reflecting a shift in focus towards the role of metabolic dysfunction in steatotic liver diseases [1, 2]. MASLD is defined as hepatic steatosis (≥ 5% of hepatocytes containing fat) along with at least one of five components of the metabolic syndrome [3, 4]. This reclassification is useful as MASLD emphasises the association between metabolic disorders and liver fat accumulation, highlighting the importance of addressing both aspects in diagnosis and management [5, 6, 7]. It is associated with an increased risk of a range of chronic physical conditions, including cardiovascular disease (CVD), hypertension, diabetes, frailty and physical disability [8, 9, 10, 11, 12, 13].
Late‐life depression (LLD) is a significant public health concern associated with impaired functioning, increased healthcare utilisation, multiple comorbidities and elevated mortality rates [14]. It is estimated that around 10% of community‐dwelling older Australians experience depression [15]. Although metabolic disorders are strongly linked to late‐life depression [16, 17, 18, 19, 20, 21, 22, 23], depression risk and prevalence have yet to be systematically evaluated in midlife and older adults with MASLD [24, 25, 26]. The newly proposed definition, emphasising key metabolic criteria, provides an opportunity to better characterise dysmetabolic comorbidities that may influence mental health outcomes. The two parallel constructs of neuroprogression and somatoprogression imply that common risk factors and operative biological pathways might drive risk and outcomes of neuropsychiatric and somatic disorders, respectively [27].
This study aimed to examine whether MASLD was prospectively associated with the risk of depression among community‐dwelling older adults. Examining how MASLD intersects with established risk factors, such as sex, obesity and dyslipidamia, could clarify the role of metabolic dysfunction in late‐life depression and ultimately enhance clinical diagnosis and management strategies in older adults and their theoretical foundations. We utilised longitudinal data from the ASPirin in Reducing Events in the Elderly (ASPREE) cohort, which includes a large sample of Australians aged 70 years and older. The association between MASLD and depression trajectories was investigated, adjusting for a range of sociodemographic characteristics, lifestyle factors, morbidities and medication use.
2. Materials and Methods
2.1. Study Population
This secondary analysis focuses on a subset of Australian older adults enroled in the ASPREE cohort who had a valid MASLD measure. ASPREE was a randomised controlled trial that recruited over 19,000 participants aged 70 years or older in Australia and the United States between 2010 and 2014. This cohort aimed to investigate whether daily low‐dose aspirin could prolong disability‐free survival in healthy older adults. Exclusion criteria included self‐reported or physician‐diagnosed dementia, a Modified Mini‐Mental State examination (3MS) [28] score below 78, prior or existing cardiovascular or cerebrovascular disease, a serious illness with a life expectancy under five years, or difficulty performing one or more of six basic activities of daily living. Participants with absolute indications for antiplatelet or anticoagulant therapy were also excluded. Follow‐up was conducted through annual in‐person visits, medical record reviews and telephone calls. The design and findings of the primary cohort have been extensively detailed in previous publications [29, 30, 31].
At baseline and annual visits, in‐person interviews and assessments collected self‐reported medical, and lifestyle information, including cognitive assessments, anthropometric measurements, and physical function markers. Clinically relevant depressive symptoms were measured with the self‐completed Centre for Epidemiologic Studies Depression Scale (CES‐D‐10) [32]. Baseline anthropometric measures included weight, height, body mass index (BMI) and abdominal circumference. Baseline laboratory data, including glucose, triglycerides and lipid profiles, were collected at local pathology laboratories. The ASPREE Healthy Ageing Biobank sub‐study collected serum samples, which were stored at −80°C, from Australian participants. Thawed serum samples underwent biochemical analyses to measure alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma‐glutamyl transferase (GGT), and other markers. These analyses were conducted centrally at Alfred Health Pathology using the Abbott Alinity ci analyzer and Abbott reagents (Abbott Diagnostics, Macquarie Park, Australia).
Additionally, this analysis incorporated data from the ASPREE Longitudinal Study of Older Persons (ALSOP), a complementary cohort involving approximately 90% of Australian ASPREE participants (including 12,572 participants). ALSOP broadens the scope of the original study by examining environmental and lifestyle factors, including social factors and self‐reported intensity of physical activity, that influence ageing [33]. Ethics approval was not required for this secondary analysis.
2.2. Outcome Variable
2.2.1. Clinically Relevant Depression Symptoms
Clinically relevant depression symptoms were measured using the short version of the CES‐D‐10 scale [34, 35]. The trajectories of depressive symptoms prevalence for ASPREE participants were estimated using Nonlinear Latent Class Mixed Models (LCMM) (Figure 1) [22, 36]. These trajectories were considered as the study outcome.
FIGURE 1.
Trajectory of depressive symptoms: a) model‐based trajectory patterns; b) summary statistics of trajectories across years. Four distinct longitudinal trajectories were identified across a median follow‐up of 4.6 years (IQR: 0.1–7.1 years) and labelled as low (‘non‐depressed’), consistently mild (‘subthreshold depression’), consistently elevated (‘persistent depression’), and initially low but increasing (‘emerging depression’) CES‐D‐10 depressive symptoms [22].
2.3. Exposure Variables
2.3.1. Metabolic Dysfunction‐Associated Steatotic Liver Disease
The Fatty Liver Index (FLI) was calculated for participants whose plasma biochemistry analysis was collected within 90 days of their baseline visit using an algorithm incorporating BMI, waist circumference, triglycerides and gamma‐glutamyl transferase (GGT) levels [11]. The FLI ranges from 0 to 100; with a score of ≥ 60 indicating fatty liver disease (i.e., hepatic steatosis) [11]. Participants with a FLI ≥ 60 were excluded if they had excessive alcohol intake (males more than 210 g/week, females more than 140 g/week) or were prescribed steatogenic medications (e.g., methotrexate, amiodarone, glucocorticoids, and/or tamoxifen). Those with a FLI ≥ 60 were classified as having MASLD if they met at least one of the five cardiometabolic comorbidities ─ hypertension, diabetes mellitus (DM), obesity (BMI ≥ 25 kg/m2), hypertriglyceridemia (fasting triglycerides ≥ 150 mg/dL), low high‐density lipoprotein (HDL) cholesterol (or treatment for dyslipidamia) ─ as defined by the MASLD diagnostic criteria [2] (Supporting Information S1: Figure S1). Participants with a FLI of < 60 were considered the no‐MASLD comparator.
2.3.2. Covariates
We studied the peer‐reviewed literature on the relationship between MASLD and depression in both the general population and older adults to identify potential confounders. The following variables were considered: demographic variables, lifestyle factors, global cognitive function, polypharmacy (defined as the use of five or more prescribed medications [37]), morbidities and medication use, social health, and physical activity. Detailed definitions and measurement approaches for these covariates are provided in the Supplementary eMethods. While BMI and alcohol use were not treated as primary confounders, given their inclusion in the MASLD definition, additional sensitivity analyses were conducted to examine their potential influence on the observed associations.
2.4. Statistical Analysis
The baseline characteristics of the study sample were summarised according to the MASLD condition, using means and standard deviations (SD) for continuous variables and frequencies (%) for categorical variables.
To investigate the association between baseline MASLD (i.e., the exposure) and depressive symptom trajectories across the follow‐up period (i.e., the outcome), we employed multivariable multinomial logistic regressions. Results are presented as relative risk ratios (RRRs) with corresponding 95% confidence intervals (CIs). Additional models investigated the role of social isolation and social support collected from the ALSOP sub‐study (n = 7275). Further adjustments for anthropometric indices, including BMI (continuous), waist‐to‐height ratio (WtHR), and waist circumference (WC), were performed as sensitivity analyses. These confounders were included in separate models to address potential collinearity.
We conducted subgroup analyses based on prevalent medical morbidities, the ASPREE study treatment arm (aspirin vs. placebo), and the use of anti‐inflammatory, antidepressant, or antipsychotic medications to explore the association between MASLD and depression risk across these factors.
Effect modifications on the relationship between MASLD and depression trajectory membership were assessed by examining the two‐way interactions between MASLD and potential effect modifiers.
MASLD was additionally categorised into No MASLD (FLI < 30), indeterminate (30 ≤ FLI < 60), and MASLD (FLI ≥ 60) groups, and we assessed the association of the three‐category MASLD definition with depression trajectories. Missing data were generally less than 1% in the key covariates and were imputed using a single imputation when necessary.
Analyses were done using STATA software, version 17.0 (StataCorp LLC, College Station, TX, USA).
3. Results
Of the 16,703 ASPREE Australian participants, 11,914 provided blood samples for the biobank, and 9097 with a valid and calculable FLI at baseline and who met the MASLD criteria were included (see Supporting Information S1: Figure S1). Baseline characteristics of participants with and without available FLI data are presented in Supporting Information S1: Table S2 and were broadly comparable across most demographic, lifestyle, and clinical characteristics. Table 1 displays baseline participants' characteristics stratified by MASLD status. The mean (SD) age was 75.0 (4.2) years, with 55% of participants being female, and most individuals identifying as White (98.16%). Among individuals with MASLD, 66.4% were classified as having obesity, compared to only 6.5% among those without MASLD. Participants with and without MASLD reported similar current alcohol use, at 76.7% and 79.5%, respectively. Almost 44% of participants with MASLD were former smokers, compared with 38% for no‐MASLD participants. The number of medical morbidities at baseline was higher in those with MASLD (26.2%) than those without MASLD (12.9%), including hypertension (82.9% vs. 69.3%) and dyslipidamia (68.4% vs. 31.6%). Similarly, a higher proportion of MASLD patients (23.0%) were prescribed five or more medications compared to 17.0% of those without MASLD.
TABLE 1.
Baseline socio‐demographic, medical, and lifestyle characteristics of participants by presence of Metabolic Dysfunction‐Associated Steatotic Liver Disease (MASLD) (n = 9097).
| MASLD f | Total | |||
|---|---|---|---|---|
| No (6099) | Yes (2998) | |||
| Sex | Male | 2545 (41.7%) | 1548 (51.6%) | 4093 (45.0%) |
| Female | 3554 (58.3%) | 1450 (48.4%) | 4804 (55.0%) | |
| Age, mean (SD) | 75.25 ± 4.38 | 74.66 ± 3.91 | 75.05 ± 4.24 | |
| Race/Ethnicity | White/Caucasian | 5997 (98.3%) | 2964 (99.0%) | 8964 (98.5%) |
| Else | 99 (1.6%) | 31 (1.0%) | 130 (1.4%) | |
| Body mass index (kg/m2) | Underweight/Normal | 2455 (40.3%) | 32 (1.1%) | 2478 (27.3%) |
| Overweight | 3249 (53.3%) | 975 (32.5%) | 4224 (46.4%) | |
| Obese | 395 (6.5%) | 1991 (66.4%) | 2386 (26.2%) | |
| Waist circumference (WC) (cm), mean (SD) | 90.40 ± 9.29 | 107.59 ± 9.18 | 96.7 ± 12.29 | |
| Waist‐height ratio (WHtR), mean (SD) | 0.55 ± 0.05 | 0.65 ± 0.05 | 0.58 ± 0.71 | |
| Education level | ≤ 12 yrs | 3537 (58.0%) | 1912 (63.8%) | 5449 (59.9%) |
| > 12 yrs | 2562 (42.0%) | 1086 (36.2%) | 3648 (40.1%) | |
| Smoking status | Current | 208 (3.4%) | 70 (2.3%) | 278 (3.1%) |
| Former | 2993 (37.6%) | 1311 (43.7%) | 4304 (39.6%) | |
| Never | 3598 (59.0%) | 1617 (53.9%) | 5215 (57.3%) | |
| alcohol status a | Current | 4848 (79.5%) | 2300 (76.7%) | 7148 (78.6%) |
| Former | 269 (4.4%) | 168 (5.6%) | 437 (4.8%) | |
| Never | 982 (16.1%) | 530 (17.7%) | 1512 (16.6%) | |
| self‐reported physical activity b | Rare | 1329 (27.0%) | 976 (42.1%) | 2305 (31.8) |
| More than light | 2630 (53.4%) | 1090 (47.1%) | 3720 (51.4%) | |
| More than moderate | 964 (19.6%) | 250 (10.8%) | 1214 (16.8%) | |
| Social isolation | No | 4672 (94.1%) | 2189 (93.4%) | 6861 (93.9%) |
| Yes | 292 (5.9%) | 154 (6.6%) | 446 (6.1%) | |
| Social support | No | 2158 (43.7%) | 1004 (43.0%) | 3162 (43.5%) |
| Yes | 2782 (56.3%) | 1331 (57.0%) | 4113 (56.5%) | |
| Living situation | At home alone/In a residential home | 1950 (32.0%) | 896 (29.9%) | 2846 (31.3%) |
| Living at home with family | 4149 (68.0%) | 2102 (70.1%) | 6251 (68.7%) | |
| Systolic blood pressure (SBP) (mmHg), mean (SD) | 139.00 ± 16.41 | 140.40 ± 15.65 | 139.46 ± 16.17 | |
| Diastolic blood pressure (DBP) (mmHg), mean (SD) | 76.55 ± 9.80 | 78.32 ± 9.92 | 77.14 ± 9.87 | |
| Hypertension g | No | 1870 (30.7%) | 513 (17.1%) | 2383 (26.2%) |
| Yes | 4229 (69.3%) | 2485 (82.9%) | 6714 (73.8%) | |
| Diabetes mellitus g | No | 5763 (94.5%) | 2495 (83.2%) | 8258 (90.8%) |
| Yes | 336 (5.5%) | 503 (16.8%) | 839 (9.2%) | |
| Metabolic syndrome g | No | 5031 (83.1%) | 921 (31.0%) | 5952 (66.0%) |
| Yes | 1020 (16.9%) | 2052 (69.0%) | 3072 (34.0%) | |
| 3MS cognitive score [mean (SD)] | 93.81 (4.35) | 93.32 (4.47) | 93.65 (4.40) | |
| Pulmonary disease g | No | 5226 (86.3%) | 2526 (84.3%) | 7788 (85.6%) |
| Yes | 837 (13.7%) | 472 (15.7%) | 1309 (14.4%) | |
| Chronic kidney disease g | No | 5601 (91.8%) | 2581 (86.1%) | 8182 (89.9%) |
| Yes | 498 (8.2%) | 417 (13.9%) | 915 (10.1%) | |
| Gout g | No | 5862 (96.1%) | 2697 (90.0%) | 8559 (94.1%) |
| Yes | 237 (3.9%) | 301 (10.0%) | 538 (5.9%) | |
| Dyslipidamia g | No | 2167 (35.5%) | 948 (31.6%) | 3115 (34.2%) |
| Yes | 3932 (64.5%) | 2049 (68.4%) | 5981 (65.8%) | |
| GORD c | No | 4403 (72.2%) | 1894 (63.2%) | 6297 (69.2%) |
| Yes | 1696 (27.8%) | 1104 (36.8%) | 2800 (30.8%) | |
| number of morbidities d | 0 | 336 (5.5%) | 58 (1.9%) | 394 (4.3%) |
| 1 | 1401 (23.0%) | 406 (13.5%) | 1807 (19.9%) | |
| 2 | 2170 (35.6%) | 863 (28.8%) | 3033 (33.3%) | |
| 3 | 1408 (23.1%) | 886 (29.6%) | 2294 (25.2%) | |
| 4 or more | 784 (12.9%) | 785 (26.2%) | 1569 (17.2%) | |
| polypharmacy e | No | 5066 (83.1%) | 2308 (77.0%) | 7374 (81.1%) |
| Yes | 1033 (16.9%) | 690 (23.0%) | 1723 (18.9%) | |
| Anti‐inflammatory use g | No | 5091 (83.5%) | 2400 (80.1%) | 7491 (82.3%) |
| Yes | 1008 (16.5%) | 598 (19.9%) | 1606 (17.7%) | |
| Antipsychotics use g | No | 5976 (98.0%) | 2930 (97.7%) | 8906 (97.9%) |
| Yes | 123 (2.0%) | 68 (2.3%) | 191 (2.1%) | |
| Antidepressant use g | No | 5491 (90.0%) | 2616 (87.3%) | 8107 (89.1%) |
| Yes | 608 (10.0%) | 382 (12.7%) | 990 (10.9%) | |
| Aspirin use | Yes | 3070 (50.3%) | 1474 (49.2%) | 4544 (50.0%) |
| No | 3029 (49.7%) | 1524 (50.8%) | 4553 (50.0%) | |
Alcohol status (based on the Australian guidelines to reduce health risks from drinking alcohol): Never: Participants who never consumed alcohol, Former alcohol use: Participants who formerly consumed alcohol but abstained during the study period, Current use: Participants who consumed alcohol at the time of the questionnaire.
Self‐reported physical activity intensity in a typical week; low (never), more than light, and more than moderate (moderate to vigorous).
GORD is defined as using any proton pump inhibitor or H2‐receptor antagonist.
Morbidity is defined as the number of chronic health conditions (hypertension, diabetes mellitus, dyslipidamia, chronic kidney disease, gastroesophageal reflux disease, pulmonary disease, osteoarthritis, and gout).
Polypharmacy is defined as the use of five or more prescribed medications.
Metabolic dysfunction‐associated steatotic liver disease (MASLD) has been defined similarly to NAFLD (≥ 5% of hepatocytes containing fat in the absence of secondary causes of steatosis) but with the addition of requiring the individual with hepatic steatosis to also meet at least one cardiometabolic criterion.
Definitions of all variables, including full names for abbreviations, are detailed in the Supplementary Materials (Supporting Information S1: Table S1).
Trajectories of depressive symptoms for our study sample are illustrated in Figure 1, showing patterns of consistently low, consistently elevated, consistently high, and initially low but increasing symptoms over time. The numbers in each trajectory were follows: ‘non‐depressed’ (N = 4188, 46%; mean [SD] CES‐D at baseline: 1.25 [1.51]); ‘subthreshold depression’ (N = 3472, 38.2%; 4.35 [2.59]); ‘persistent depression’ (N = 829, 9.1%; 8.41 [4.38]); and ‘emerging depression’ (N = 608, 6.7%; 1.07 [1.26]). Box plots in Figure 1 illustrate the trajectories across the follow‐ups.
Illustrated in Table 2, the distribution of MASLD patients varied across the depression trajectory classes. MASLD was slightly more prevalent in the subthreshold and persistent depression trajectories (39.1% and 10.9%, respectively). The smallest proportion of individuals with MASLD fell within the emerging depression trajectory, where the distribution was nearly equal between those with and without MASLD. Furthermore, a lower proportion of individuals with MASLD was observed in the non‐depressed trajectory compared to those without MASLD (43.4% vs. 47.4%).
TABLE 2.
Distribution of depressive symptom trajectories by MASLD status.
| MASLD a | Total | |||
|---|---|---|---|---|
| No (6099) | Yes (2998) | |||
| Trajectories of depressive symptoms b | Subthreshold depression | 2299 (37.7%) | 1173 (39.1%) | 3472 (38.2%) |
| Persistent depression | 502 (8.2%) | 327 (10.9%) | 829 (9.1%) | |
| Emerging depression | 410 (6.7%) | 198 (6.6%) | 608 (6.7%) | |
| Non‐depressed | 2888 (47.4%) | 1300 (43.4%) | 4188 (46.0%) | |
Metabolic dysfunction‐associated steatotic liver disease (MASLD) has been defined similarly to NAFLD (≥ 5% of hepatocytes containing fat in the absence of secondary causes of steatosis) but with the addition of requiring the individual with hepatic steatosis to also meet at least one cardiometabolic criterion.
Depression trajectories are identified using a latent class mixed‐model approach. The four distinct patterns of depressive symptoms over time were labelled as: consistently low, elevated, emerging, and persistently high.
Table 3 displays model‐adjusted relative risk ratio estimates to quantify the association between MASLD presence and membership likelihood (i.e., relative risk ratio) of subthreshold depression, persistent depression, and emerging depression trajectories compared with a non‐depressed trajectory (i.e., the reference group). In Model 1, after adjusting for baseline age, sex, education, living status, cognition score, smoking, and alcohol use, MASLD was associated with a 43% (RRR = 1.43; 95% CI: 1.22, 1.67) and a 13% (1.13; CI: 1.02, 1.24) increase in the likelihood of being in the persistent and subthreshold depression trajectory, respectively, compared with non‐depressed trajectory. However, there was no association between MASLD and the likelihood of emerging trajectory membership (1.09; CI: 0.91, 1.31). Relative risk ratio estimates for the confounders that were included in Model 1 are displayed in Supporting Information S1: Table S3.
TABLE 3.
Model‐adjusted relative risk ratios and 95% CI for investigating the association between MASLD and depression trajectories.
| Depression trajectories |
Number, Baseline mean (SD) |
RRR (95% CI) | |||
|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 4 | ||
| Non‐depressed (ref) |
4188 1.25 (1.51) |
— | — | — | — |
|
Subthreshold depression versus Non‐depressed |
3472 4.35 (2.59) |
1.13 (1.02, 1.24) | 1.14 (1.03, 1.25) | 1.10 (0.99, 1.21) | 1.06 (0.96, 1.17) |
|
Persistent depression versus. Non‐depressed |
829 8.41 (4.38) |
1.43 (1.22, 1.67) | 1.42 (1.22, 1.66) | 1.34 (1.14, 1.57) | 1.28 (1.09, 1.51) |
|
Emerging depression versus. Non‐depressed |
608 1.07 (1.26) |
1.09 (0.91, 1.31) | 1.09 (0.91, 1.31) | 1.06 (0.88, 1.27) | 1.04 (0.86, 1.26) |
| Chi‐square (df) p‐value | — |
21.59 (3) < 0.0001 |
21.55 (3) < 0.0001 |
13.83 (3) 0.0031 |
9.44 (3) 0.0239 |
Note: Model 1 adjusted for baseline age, sex, education level, living situation, cognition score, smoking status. Model 2 adjusted for baseline age, sex, education level, living situation, cognition score, smoking status, alcohol use. Model three in addition to model 1 covariates, the model adjusted for polypharmacy. Model four in addition to model 1 covariates, the model adjusted for number of morbidities (hypertension, diabetes mellitus, dyslipidamia, chronic kidney disease, gastroesophageal reflux disease, pulmonary disease, and gout).
Abbreviations: df: degree of freedom. RRR (95% CI): relative risk ratio with its 95% Confidence Interval.
Further adjustments for alcohol use, polypharmacy, or the number of medical morbidities indicated that the relative risk ratios remained almost unchanged in models 2 to 4 but showed a small reduction and a non‐significant association for the subthreshold trajectory in models 3 and 4.
We explored subgroup analyses based on sex, dyslipidamia status, treatment arm (aspirin vs. placebo), and self‐reported use of prescribed medications, including anti‐inflammatories, and antidepressant or antipsychotic medications. A similar pattern to Model 1 was observed for females, the MASLD condition was associated with a 69% increase in the likelihood of belonging to the persistent depression trajectory (1.69; CI: 1.37, 2.08). In contrast, the RRRs for males were nonsignificant, with a trend towards a higher likelihood of belonging to the persistent depression trajectory (1.19; CI: 0.94, 1.51). In participants with dyslipidamia, individuals with MASLD had a higher likelihood of belonging to both the persistent depression trajectory (1.48, 95% CI: 1.22–1.79) and the subthreshold depression trajectory (1.16, 95% CI: 1.03–1.31), although no significant association was found for the emerging trajectory. Among participants who used anti‐inflammatory or antipsychotic/antidepressant medications, those with MASLD showed a similar trend to Model 1, but without statistical significance. Among participants taking aspirin, MASLD was associated with a significant 31% increase in the likelihood of belonging to the persistent depression trajectory (1.31; CI: 1.05, 1.64), with no significant association observed for other trajectories (Table 4).
TABLE 4.
Association between MASLD and depression trajectories: Sub‐group analysis based on sex, morbidities, and medication use.
| RRR (95% CI) | Chi‐square(df) p‐value | ||||||
|---|---|---|---|---|---|---|---|
| Depression trajectories | |||||||
| Non‐depressed |
Subthreshold depression versus non‐depressed |
Persistent depression versus non‐depressed | Emerging depression versus non‐depressed | ||||
| Sex | N [mean (SD)] CES‐D10 | Male (4,093) |
2264 1.45 (1.64) |
1924 4.68 (2.68) |
473 8.90 (4.44) |
343 1.25 (1.38) |
|
| Female (5,004) |
1924 1.01 (1.31) |
1548 3.95 (2.42) |
356 7.77 (4.24) |
265 0.85 (1.05) |
|||
| RRR | Male | Ref | 1.07 (0.93, 1.23) | 1.19 (0.94, 1.51) | 1.07 (0.82, 1.40) |
2.62 (3) 0.4548 |
|
| Female | Ref | 1.20 (1.05, 1.37) | 1.69 (1.37, 2.08) | 1.12 (0.86, 1.44) |
25.09 (3) < 0.0001 |
||
| Dyslipidamia (N = 5982) | N [mean (SD)] CES‐D10 |
2746 1.28 (1.53) |
2274 4.42 (2.64) |
559 8.46 (4.51) |
403 1.13 (1.31) |
||
| RRR | Ref | 1.16 (1.03, 1.31) | 1.48 (1.22, 1.79) | 1.07 (0.85, 1.34) |
18.29 (3) < 0.0001 |
||
| Anti‐psychotic RRR antidepressant use (n = 1130) | N [mean (SD)] CES‐D10 |
334 1.60 (1.92) |
472 4.95 (2.84) |
238 9.61 (4.76) |
86 1.12 (1.34) |
||
| RRR | Ref | 1.30 (0.96, 1.75) | 1.33 (0.94, 1.89) | 0.91 (0.53, 1.52) |
4.99 (3) 0.1725 |
||
| Anti‐inflammatory use (N = 1604) | N [mean (SD)] CES‐D10 |
656 1.31 (1.56) |
635 4.50 (2.70) |
197 8.76 (4.67) |
118 1.27 (1.31) |
||
| RRR | Ref | 1.26 (1.00, 1.59) | 1.22 (0.87, 1.71) | 1.26 (0.83, 1.91) |
4.54 (3) (0.2085) |
||
| Aspirin use (4544) | N [mean (SD)] CES‐D10 |
2084 1.21 (1.50) |
1727 4.34 (2.59) |
418 8.55 (4.51) |
315 1.10 (1.24) |
||
| RRR | Ref | 1.03 (0.90, 1.19) | 1.31 (1.05, 1.64) | 0.98 (0.76, 1.28) |
6.03 (3) 0.1102 |
||
Note: RRR (95% CI): relative risk ratio with its 95% Confidence Interval. The model was adjusted for baseline age, sex, education level, living status, cognition score, and smoking use.
The associations between MASLD and depressive symptom trajectories remained robust after further adjustments for social isolation, social support, BMI (continuous), WtHR, and WC (Supporting Information S1: Table S4).
Potential effect modifiers in the association between MASLD and depressive symptom trajectories were evaluated through two‐way interaction terms between MASLD and the respective modifiers. No significant interactions were identified for demographic variables, anthropometric indices, cognitive scores, morbidities, or physical activity. However, social support demonstrated a significant interaction effect with MASLD (Supporting Information S1: Table S5).
When MASLD was grouped based on three strata: no‐MASLD (FLI < 30), indeterminate (30 ≤ FLI < 60), and MASLD (FLI ≥ 60), most associations were no longer statistically significant. However, a significant association remained for individuals in the persistent depression trajectory, consistent with the findings using the binary MASLD classification (Supporting Information S1: Table S6).
4. Discussion
This large‐scale prospective study of community‐dwelling older adults provides several key insights into the relationships between the presence of MASLD and subsequent depressive symptom trajectories over time. First, individuals with MASLD were more likely to experience adverse depression trajectories, with a higher likelihood of persistent and subthreshold depression compared to those without MASLD. Second, the distribution of MASLD varied across depression trajectories, with a higher prevalence in the subthreshold and persistent depression groups. Third, subgroup analyses revealed important sex differences, with females showing a stronger and statistically significant association between MASLD and persistent depression. In comparison, the association in males was weaker and not significant. Additionally, among individuals with coexisting dyslipidamia and MASLD, the likelihood of persistent depression was elevated by approximately 50%. Fourth, further adjustments for social isolation, social support and anthropometric indices did not alter the primary associations, reinforcing the robustness of our findings. However, social support was identified as a potential protective modifier, reducing the risk of emerging depression in MASLD participants. These findings suggest that MASLD may contribute to the persistence and chronicity of depressive symptoms in older adults, particularly in those with additional underlying comorbidities.
The association between MASLD and persistent depression trajectories observed in this study aligns with existing evidence highlighting the interplay between metabolic dysfunction and long‐term depressive symptoms [38]. Prior research has demonstrated that individuals with MASLD show a significantly higher risk of depression, with studies reporting a nearly 1.8‐fold increased rate of severe depression compared to the general population (24). The persistence of depressive symptoms in MASLD populations may be attributed to systemic inflammation, insulin resistance, dysregulated mitochondrial function and neuroendocrine dysregulation, which are commonly observed in both conditions. Chronic low‐grade inflammation, a hallmark of MASLD, has been linked to alterations in serotonergic and dopaminergic pathways, which may contribute to sustained depressive symptoms rather than transient mood disturbances [25]. Furthermore, dysregulation of the gut‐liver‐brain axis has been proposed as a mechanistic link between MASLD and depression, with evidence suggesting that changes in gut microbiota composition and increased intestinal permeability may exacerbate neuroinflammation and contribute to persistent mood disorders [39]. These findings highlight the need for targeted mental health screening and early intervention strategies for individuals with MASLD, particularly those at risk for persistent depression, to mitigate the adverse psychological consequences associated with this condition.
An interesting finding is the absence of a significant association between MASLD and emerging depression – that those who, at baseline, do not have depressive symptoms are not more likely to subsequently develop them, despite the previously mentioned neurobiological links between MASLD and depression. Emerging depression may be a distinct phenomenon more closely linked to psychosocial vulnerabilities and baseline mental health issues [40], rather than chronic conditions such as MASLD. In our study, the lack of this link may therefore be attributed to the general unawareness of MASLD among patients, as it is often largely asymptomatic and not perceived as an immediate health concern. Unlike conditions with more apparent clinical manifestations, MASLD typically progresses gradually and silently, potentially preventing it from being recognized as a source of overt psychological distress in its early stages. Prior research has highlighted that patients rarely consider MASLD a serious condition, which may explain why it does not trigger the kind of emotional response that may lead to emerging depressive symptoms [41]. Given this, it may take some time for the progression or persistence of MASLD to impact psychological well‐being, such that it may potentially contribute to chronicity rather than new‐onset depressive symptoms. This underscores the need for improved patient education and awareness initiatives to ensure early recognition of MASLD's long‐term health implications, including its potential association with mental health outcomes.
Importantly, we also observed sex differences in the association between MASLD and depressive symptom trajectories, with older females showing more than 50% higher likelihood of belonging to the persistent depression trajectory compared to older males. This disparity may be attributed to a combination of biological and psychosocial mechanisms [1]: Hormones, particularly oestrogen, play a crucial role in mood regulation, as oestrogen exerts neuroprotective effects at least in part by modulating serotonergic and dopaminergic activity. The decline in oestrogen levels during menopause has been associated with an increased risk of depressive symptoms, potentially exacerbating the impact of MASLD on both persistent and emerging depression trajectories [2]; Sex‐based differences in inflammatory responses contribute to differential depression risks in MASLD, as females may show heightened systemic inflammation compared to males. Chronic low‐grade inflammation, a hallmark of MASLD, interacts with neuroimmune and mitochondrial pathways to increase vulnerability to depressive disorders. Markers of inflammation increase with ageing, and this pathway may be more relevant in elderly cohorts [26]. This aligns with evidence suggesting that females with MASLD face a higher risk of severe depression due to the convergence of inflammatory and metabolic dysfunction [3]; Alterations in metabolic function and gut‐liver‐brain axis dysregulation may further predispose females with MASLD to depression, as sex‐specific differences in gut microbiota composition and metabolic endotoxemia have been linked to depressive symptomatology [4]; Psychosocial factors also play a crucial role in shaping depression risk among females with MASLD. Sex role expectations may impose particular stressors on females, who may experience stigma related to health and body image, particularly about obesity, which is frequently associated with MASLD and depression [42]. These findings highlight the need for targeted mental health screenings and sex‐specific intervention strategies for individuals with MASLD, particularly females, to address both biological and psychosocial risk factors contributing to depression.
We observed a 50% increase in the likelihood of persistent depression among individuals with coexisting MASLD and dyslipidamia. Dyslipidamia is present in approximately 40%–70% of individuals with MASLD and contributes to a proinflammatory and insulin‐resistant state [43]. These metabolic abnormalities have been implicated in the pathogenesis of depression through neuroimmune dysregulation, oxidative stress, and impaired lipid metabolism affecting central nervous system function. Elevated triglycerides, in particular, have been positively associated with depressive symptoms, further supporting the role of lipid disturbances in mood disorders [44]. Moreover, both MASLD and dyslipidamia independently increase cardiometabolic risk, and their co‐occurrence may synergistically amplify systemic inflammatory burden, predisposing individuals to persistent depressive trajectories. The added influence of behavioural and treatment‐related factors, such as physical inactivity, poor diet and the cumulative effects of polypharmacy, may further contribute to common biological pathways for medical comorbidity and sustained depressive symptoms in this population [39, 40, 45].
Several demographic and lifestyle factors modulate the relationship between MASLD and depression, particularly in older adults. Ageing is associated with physiological changes, such as immunosenescence and vascular ageing, which may amplify the inflammatory and neurodegenerative effects of MASLD, contributing to persistent depressive symptoms. Additionally, older adults may have a longer history of metabolic dysfunction (or advanced liver fibrosis) and more comorbid illnesses that compound both physical and mental health burdens [46]. Given that MASLD is associated with an increased risk of persistent physical disability and frailty, these impairments may contribute to a loss of physical functional independence, reduced likelihood of community living, and overall decline in physical health [11, 47]. Socioeconomic disparities further influence this association, as lower education and income are linked to both increased MASLD risk and reduced mental health resilience due to chronic stress and limited healthcare access [20, 48]. Lifestyle factors, particularly poor diet, physical inactivity, smoking and alcohol use, also play a critical role. Smoking accelerates insulin resistance and systemic inflammation while also being a known risk factor for depression [49, 50]. Similarly, alcohol consumption, even at moderate levels, may worsen liver outcomes in MASLD and increase the risk of advanced disease [51]. Lastly, social support emerges as a key determinant in mitigating depressive symptoms, as individuals with strong family or community networks are more likely to adhere to treatment and experience better mental health outcomes [52]. Addressing these interconnected factors through targeted interventions, such as lifestyle modifications, psychosocial support and equitable healthcare access, may contribute to improving mental health outcomes in older adults with MASLD.
5. Conclusion
This study underscores the intricate relationship between MASLD and depression, highlighting the need for a multidimensional approach to patient care. Given the overlapping metabolic, inflammatory and psychosocial factors contributing to depressive symptom trajectories in MASLD, an integrated strategy that incorporates routine mental health screening, lifestyle modifications, and targeted psychosocial support is indicated.
Author Contributions
Marzieh Nasr Azadani: writing – original draft, software, project administration, methodology, investigation, formal analysis, data curation, conceptualization. Najmeh Davoodian: writing – review and editing, visualization, validation, methodology, conceptualization. Michael Berk: writing – review and editing, visualization, project administration, methodology, conceptualization. Daniel Clayton‐Chubb: writing – review and editing, visualization, supervision, methodology, conceptualization. Malcolm Forbes: writing – review and editing, visualization, methodology, conceptualization. Shiva Ganjali: writing – review and editing, visualization, supervision, methodology, conceptualization. William Kemp: writing – review and editing, visualization, methodology, conceptualization. Robyn L. Woods: writing – review and editing, visualization, project administration, methodology, conceptualization. Mojtaba Lotfaliany: writing – review and editing, visualization, validation, supervision, software, methodology, conceptualization. John J. McNeil: writing – review and editing, visualization, project administration, methodology, conceptualization. Julie A. Pasco: writing – review and editing, visualization, project administration, methodology, conceptualization. Stuart K. Roberts: writing – review and editing, visualization, methodology, conceptualization. Mohammadreza Mohebbi: writing – review and editing, project administration, visualization, validation, supervision, software, methodology, conceptualization.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Information S1
Acknowledgements
The ASPREE principal study is funded primarily by a grant from the US National Institute on Aging (U01AG029824, U19AG062682); with additional support from Australia's NHMRC (ID334047; 1127060), the Victorian Cancer Agency and Monash University. The ASPREE Healthy Ageing Biobank, which collected the blood samples for measures of inflammatory biomarkers for this study, was funded by a Flagship Cluster Collaboration award from the CSIRO and the National Cancer Institute through a supplement (5U01AG029824‐02) to the NIH ASPREE grant. The ASPREE‐D study exploring the depression outcome was funded by a project grant from the NHMRC (GNT1081901). MB and JMCN are supported by NHMRC Leadership 3 Investigator Grants (GNT2017131 and GNT2033013, respectively). ND is supported by the GNT 2035912 ‐ SYNDICAT: Synergy for the Development of Innovative Clinical Treatments in Schizophrenia. Open access publishing facilitated by Deakin University, as part of the Wiley ‐ Deakin University agreement via the Council of Australasian University Librarians.
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
References
- 1. Alboraie M., Fouad Y., and Eslam M., “A Multi‐Society Delphi Consensus Statement on New Fatty Liver Disease Nomenclature,” Hepatology 10 (2024): 1097, 10.1097/HEP.0000000000000966. [DOI] [Google Scholar]
- 2. Rinella M. E., Lazarus J. V., Ratziu V., et al., “A Multisociety Delphi Consensus Statement on New Fatty Liver Disease Nomenclature,” Hepatology 78, no. 6 (2023): 1966–1986, 10.1016/j.aohep.2023.101133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Eslam M., Sanyal A. J., George J., et al. “MAFLD: A Consensus‐Driven Proposed Nomenclature for Metabolic Associated Fatty Liver Disease.” Gastroenterology. (2020);158, no. 7:1999–2014, 10.1053/j.gastro.2019.11.312. [DOI] [PubMed] [Google Scholar]
- 4. Moon J. H., Jeong S., Jang H., Koo B. K., and Kim W., “Metabolic Dysfunction‐Associated Steatotic Liver Disease Increases the Risk of Incident Cardiovascular Disease: A Nationwide Cohort Study,” EClinicalMedicine 65 (2023): 102292, 10.1016/j.eclinm.2023.102292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hagström H., Vessby J., Ekstedt M., and Shang Y., “99% of Patients With NAFLD Meet MASLD Criteria and Natural History Is Therefore Identical,” Journal of Hepatology 80, no. 2 (2024): e76–e77, 10.1016/j.jhep.2023.08.026. [DOI] [PubMed] [Google Scholar]
- 6. Song S. J., Lai J. C.‐T., Wong G. L.‐H., Wong V. W.‐S., and Yip T. C.‐F., “Can We Use Old NAFLD Data Under the New MASLD Definition?,” Journal of Hepatology 80, no. 2 (2024): e54–e56, 10.1016/j.jhep.2023.07.021. [DOI] [PubMed] [Google Scholar]
- 7. Younossi Z. M., Paik J. M., Stepanova M., Ong J., Alqahtani S., and Henry L., “Clinical Profiles and Mortality Rates Are Similar for Metabolic Dysfunction‐Associated Steatotic Liver Disease and Non‐Alcoholic Fatty Liver Disease,” Journal of Hepatology 80, no. 5 (2024): 694–701, 10.1016/j.jhep.2024.01.014. [DOI] [PubMed] [Google Scholar]
- 8. Chan W. K., Petta S., Noureddin M., Goh G. B. B., and Wong V. W. S., “Diagnosis and Non‐Invasive Assessment of MASLD in Type 2 Diabetes and Obesity,” Alimentary Pharmacology and Therapeutics 59, no. 1 (2024): 23–40, 10.1111/apt.17866. [DOI] [PubMed] [Google Scholar]
- 9. Clayton‐Chubb D., Kemp W. W., Majeed A., et al., “Late‐Life Metabolic Dysfunction‐Associated Steatotic Liver Disease and Its Association With Physical Disability and Dementia,” Journals of Gerontology (2024): glae011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Clayton‐Chubb D., Roberts S. K., Majeed A., et al. “Associations Between MASLD, Atrial Fibrillation, Cardiovascular Events, Mortality and Aspirin Use in Older Adults,” GeroScience 47, no. 1 (2025): 1303–1318, 10.1007/s11357-024-01435-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Clayton‐Chubb D., Kemp W. W., Majeed A., et al. “Metabolic Dysfunction‐Associated Steatotic Liver Disease in Older Adults Is Associated With Frailty and Social Disadvantage,” Liver International 44, no. 1 (2024): 39–51, 10.1111/liv.15725. [DOI] [PubMed] [Google Scholar]
- 12. Lee H.‐H., Lee H. A., Kim E.‐J., et al. “Metabolic Dysfunction‐Associated Steatotic Liver Disease and Risk of Cardiovascular Disease,” Gut 73, no. 3 (2024): 533–540, 10.1136/gutjnl-2023-331003. [DOI] [PubMed] [Google Scholar]
- 13. Targher G., Byrne C. D., and Tilg H., “MASLD: A Systemic Metabolic Disorder With Cardiovascular and Malignant Complications,” Gut 73, no. 4 (2024): 691–702, 10.1136/gutjnl-2023-330595. [DOI] [PubMed] [Google Scholar]
- 14. Taylor W. D., “Depression in the Elderly,” New England Journal of Medicine 371, no. 13 (2014): 1228–1236, 10.1056/nejmcp1402180. [DOI] [PubMed] [Google Scholar]
- 15. Mohebbi M., Agustini B., Woods R. L., et al. “Prevalence of Depressive Symptoms and Its Associated Factors Among Healthy Community‐Dwelling Older Adults Living in Australia and the United States,” International Journal of Geriatric Psychiatry 34, no. 8 (2019): 1208–1216, 10.1002/gps.5119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Labenz C., Huber Y., Michel M., et al. “Nonalcoholic Fatty Liver Disease Increases the Risk of Anxiety and Depression,” Hepatology Communications 4, no. 9 (2020): 1293–1301, 10.1002/hep4.1541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lee J. W. and Park S. H., “Association Between Depression and Nonalcoholic Fatty Liver Disease: Contributions of Insulin Resistance and Inflammation,” Journal of Affective Disorders 278 (2021): 259–263, 10.1016/j.jad.2020.09.073. [DOI] [PubMed] [Google Scholar]
- 18. Ng C. H., Xiao J., Chew N. W., et al. “Depression in Non‐Alcoholic Fatty Liver Disease Is Associated With an Increased Risk of Complications and Mortality,” Frontiers of Medicine 9 (2022): 985803, 10.3389/fmed.2022.985803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Patel P. J., Hayward K. L., Rudra R., et al. “Multimorbidity and Polypharmacy in Diabetic Patients With NAFLD: Implications for Disease Severity and Management,” Medicine 96, no. 26 (2017): e6761, 10.1097/md.0000000000006761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Shea S., Lionis C., Kite C., et al. “Non‐Alcoholic Fatty Liver Disease and Coexisting Depression, Anxiety And/Or Stress in Adults: A Systematic Review and Meta‐Analysis,” Frontiers in Endocrinology 15 (2024): 1357664, 10.3389/fendo.2024.1357664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Weinstein A. A., De Avila L., Kannan S., et al. “Interrelationship Between Physical Activity and Depression in Nonalcoholic Fatty Liver Disease,” World Journal of Hepatology 14, no. 3 (2022): 612–622, 10.4254/wjh.v14.i3.612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Agustini B., Lotfaliany M., Mohebbi M., et al. “Trajectories of Depressive Symptoms in Older Adults and Associated Health Outcomes,” Nature Aging 2, no. 4 (2022): 295–302, 10.1038/s43587-022-00203-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Forbes M., Lotfaliany M., Mohebbi M., et al. “Depressive Symptoms and Cognitive Decline in Older Adults,” International Psychogeriatrics 36, no. 11 (2024): 1039–1050, 10.1017/s1041610224000541. [DOI] [PubMed] [Google Scholar]
- 24. Åström H., Shang Y., Hagström H., and Wester A., “Persons With Metabolic Dysfunction‐Associated Steatotic Liver Disease Are at Increased Risk of Severe Depression,” Liver International (2024). [DOI] [PubMed] [Google Scholar]
- 25. Botacin E. C., Duarte S. M. B., Stefano J. T., Barbosa M. E. D., Pessoa M. G., and Oliveira C. P., “Association Between Anxiety and Depression in Metabolic Dysfunction‐Associated Steatotic Liver Disease (MASLD),” Arquivos de Gastroenterologia 61 (2024): e23128, 10.1590/s0004-2803.24612023-128. [DOI] [PubMed] [Google Scholar]
- 26. Afendy A., Kallman J., Stepanova M., et al. “Predictors of Health‐Related Quality of Life in Patients With Chronic Liver Disease,” Alimentary Pharmacology and Therapeutics 30, no. 5 (2009): 469–476, 10.1111/j.1365-2036.2009.04061.x. [DOI] [PubMed] [Google Scholar]
- 27. Berk M. and Forbes M., “The Parallel Roads of Neuroprogression and Somatoprogression: Implications for Clinical Care,” American Journal of Geriatric Psychiatry 32, no. 8 (2024): 968–971, 10.1016/j.jagp.2024.02.011. [DOI] [PubMed] [Google Scholar]
- 28. Teng E. and Chui H., “The Modified mini‐mental State Examination (3MS),” Canadian Journal of Psychiatry 41, no. 2 (1987): 114–121, https://pubmed.ncbi.nlm.nih.gov/3611032/. [PubMed] [Google Scholar]
- 29. Ernst M. E., Broder J. C., Wolfe R., et al. “Health Characteristics and Aspirin Use in Participants at the Baseline of the Aspirin in Reducing Events in the Elderly – Extension (ASPREE‐XT) Observational Study,” Contemporary Clinical Trials 130 (2023): 107231, 10.1016/j.cct.2023.107231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. McNeil J. J., Woods R. L., Nelson M. R., et al. “Baseline Characteristics of Participants in the ASPREE (Aspirin in Reducing Events in the Elderly) Study,” Journals of Gerontology: Series A 72, no. 11 (2017): 1586–1593, 10.1093/gerona/glw342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Group A. I., “Study Design of Aspirin in Reducing Events in the Elderly (ASPREE): A Randomized, Controlled Trial,” Contemporary Clinical Trials 36, no. 2 (2013): 555–564, 10.1016/j.cct.2013.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Radloff L. S., “The CES‐D Scale:A Self‐Report Depression Scale for Research in the General Population,” Applied Psychological Measurement 1, no. 3 (1977): 385–401, 10.1177/014662167700100306. [DOI] [Google Scholar]
- 33. McNeil J. J., Woods R. L., Ward S. A., et al. “Cohort Profile: The ASPREE Longitudinal Study of Older Persons (ALSOP),” International Journal of Epidemiology 48, no. 4 (2019): 1048–9h, 10.1093/ije/dyy279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Andresen E. M., Malmgren J. A., Carter W. B., and Patrick D. L., “Screening for Depression in Well Older Adults: Evaluation of,” Preventive Medicine 10, no. 2 (1994): 77–84, 10.1016/s0749-3797(18)30622-6. [DOI] [PubMed] [Google Scholar]
- 35. Berk M., Woods R., Nelson M., et al. “ASPREE‐D: Aspirin for the Prevention of Depression in the Elderly,” International Psychogeriatrics 28, no. 10 (2016): 1741–1748, 10.1017/s104161021600079x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Proust‐Lima C., Philipps V., and Liquet B., “Estimation of Extended Mixed Models Using Latent Classes and Latent Processes: The R Package Lcmm,” Journal of Statistical Software 78, no. 2 (2017): 1–56, 10.18637/jss.v078.i02. [DOI] [Google Scholar]
- 37. Agustini B., Lotfaliany M., Woods R. L., et al. “Patterns of Association Between Depressive Symptoms and Chronic Medical Morbidities in Older Adults,” Journal of the American Geriatrics Society 68, no. 8 (2020): 1834–1841, 10.1111/jgs.16468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Liang W., Zhong K., Lai T., et al. “Causal Relationship Between Depression and Metabolic Dysfunction‐Associated Steatotic Liver Disease: A Bidirectional Mendelian Randomized Study,” Frontiers in Psychiatry 15 (2024): 1384003, 10.3389/fpsyt.2024.1384003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Mikkelsen A. C. D., Kjærgaard K., Schapira A. H., Mookerjee R. P., and Thomsen K. L., “The Liver–Brain Axis in Metabolic Dysfunction‐Associated Steatotic Liver Disease,” Lancet Gastroenterology and Hepatology 10, no. 3 (2025): 248–258, 10.1016/s2468-1253(24)00320-0. [DOI] [PubMed] [Google Scholar]
- 40. Markkula N., Marola N., Nieminen T., et al. “Predictors of New‐Onset Depressive Disorders–Results From the Longitudinal Finnish Health 2011 Study,” Journal of Affective Disorders 208 (2017): 255–264, 10.1016/j.jad.2016.08.051. [DOI] [PubMed] [Google Scholar]
- 41. Brodosi L., Stecchi M., Musio A., et al. “Anxiety and Depression in Metabolic‐Associated Steatotic Liver Disease: Relation With Socio‐Demographic Features and Liver Disease Severity,” Acta Diabetologica 61, no. 8 (2024): 1–11, 10.1007/s00592-024-02287-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Pasco J. A., Berk M., Penninx B., et al. “Obesity and Sarcopenic Obesity Characterized by Low‐Grade Inflammation Are Associated With Increased Risk for Major Depression in Women,” Frontiers in Nutrition 10 (2023): 1222019, 10.3389/fnut.2023.1222019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Bril F., Berg G., Barchuk M., and Nogueira J. P., “Practical Approaches to Managing Dyslipidemia in Patients With Metabolic Dysfunction‐Associated Steatotic Liver Disease,” Journal of Lipid and Atherosclerosis 14, no. 1 (2024): 5, 10.12997/jla.2025.14.1.5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Shi Y.‐Y., Zheng R., Cai J.‐J., and Qian S.‐Z., “The Association Between Triglyceride Glucose Index and Depression: Data From NHANES 2005–2018,” BMC Psychiatry 21, no. 1 (2021): 267, 10.1186/s12888-021-03275-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Younossi Z., Anstee Q. M., Marietti M., et al. “Global Burden of NAFLD and NASH: Trends, Predictions, Risk Factors and Prevention,” Nature Reviews Gastroenterology and Hepatology 15, no. 1 (2018): 11–20, 10.1038/nrgastro.2017.109. [DOI] [PubMed] [Google Scholar]
- 46. Tylutka A., Morawin B., Wawrzyniak‐Gramacka E., et al. “Immunosenescence in Aging‐Related Vascular Dysfunction,” International Journal of Molecular Sciences 23, no. 21 (2022): 13269, 10.3390/ijms232113269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Clayton‐Chubb D., Kemp W. W., Majeed A., et al,\. “Late‐Life Metabolic Dysfunction‐Associated Steatotic Liver Disease and Its Association With Physical Disability and Dementia,” Journals of Gerontology, Series A: Biological Sciences and Medical Sciences 79, no. 4 (2024): glae011, 10.1093/gerona/glae011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Åström H., Takami Lageborn C., and Hagstrom H., “Psychosocial Risks in Metabolic Dysfunction‐Associated Steatotic Liver Disease,” Expert Review of Gastroenterology and Hepatology (2025): (just‐accepted). [DOI] [PubMed] [Google Scholar]
- 49. Tareen K. and Winder G. S., “Tobacco, Cannabis, and the Liver,” Clinical Liver Disease 23, no. 1 (2024): e0210, 10.1097/cld.0000000000000210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Pasco J. A., Williams L. J., Jacka F. N., et al. “Tobacco Smoking as a Risk Factor for Major Depressive Disorder: Population‐based Study,” British Journal of Psychiatry 193, no. 4 (2008): 322–326, 10.1192/bjp.bp.107.046706. [DOI] [PubMed] [Google Scholar]
- 51. Marti‐Aguado D., Calleja J. L., Vilar‐Gomez E., et al. “Low‐To‐Moderate Alcohol Consumption Is Associated With Increased Fibrosis in Individuals With Metabolic Dysfunction‐Associated Steatotic Liver Disease,” Journal of Hepatology 81, no. 6 (2024): 930–940, 10.1016/j.jhep.2024.06.036. [DOI] [PubMed] [Google Scholar]
- 52. Teo R. H., Cheng W. H., Cheng L. J., Lau Y., and Lau S. T., “Global Prevalence of Social Isolation Among Community‐Dwelling Older Adults: A Systematic Review and Meta‐Analysis,” Archives of Gerontology and Geriatrics 107 (2023): 104904, 10.1016/j.archger.2022.104904. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information S1
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.