Emotional Eating in Hospitalized Older Adults Tied to Age, Not Weight

Emotional eating—the tendency to eat in response to negative emotions rather than hunger—has mostly been studied in younger and community-dwelling adults, leaving limited evidence for the very old. A cross-sectional study of 300 hospitalized adults aged 65 years and older, published in BMC Geriatrics, examined whether emotional eating is linked to anthropometric measurements, blood chemistry, and selected cardiometabolic risk indicators in this population. The findings point to a more nuanced picture than the popular narrative suggests: emotional eating in these older patients tracked with age, alcohol consumption, lipid-lowering medication use, and a kidney-related blood marker, but not with the conventional measures of metabolic risk that many clinicians and patients assume are connected to eating behavior.

The participants underwent a standardized battery of assessments. Researchers recorded sociodemographic characteristics, measured body mass index (BMI) and waist-to-hip ratio (WHR), and collected biochemical values including high-density lipoprotein (HDL) cholesterol, blood glucose, and creatinine, all interpreted against the hospital laboratory’s reference ranges. Emotional eating was assessed with the Emotional Eating Scale (EES). The metabolic picture of the cohort was striking: 58.3 percent of participants were overweight or obese, and low HDL cholesterol (below 45 mg/dL) and elevated blood glucose (above 109 mg/dL) were each observed in 51.7 percent of the sample. In other words, cardiometabolic risk was common in this hospitalized group—but it did not travel in lockstep with emotional eating scores.

In the multivariable linear regression analysis, which adjusts for multiple variables simultaneously to isolate independent associations, the study identified factors linked to emotional eating scores. Younger age within the older-adult range (B = −0.171, 95% CI: −0.308 to −0.034; p = 0.015), alcohol consumption (B = −4.347, 95% CI: −8.422 to −0.271; p = 0.037), and use of lipid-lowering medication (B = −4.309, 95% CI: −7.251 to −1.366; p = 0.004) were each independently associated with higher emotional eating scores in the authors’ reported model. Elevated creatinine levels, a marker related to kidney function, pointed in the opposite direction, being associated with lower emotional eating scores (B = 2.598, 95% CI: 0.295 to 4.900; p = 0.027). To test the robustness of these results, the team ran a sensitivity analysis that additionally adjusted for income status, and the main associations remained consistent.

What did not emerge is arguably the more consequential finding. No statistically significant associations were observed between emotional eating and BMI, waist circumference, waist-to-hip ratio, blood glucose, total cholesterol, or triglycerides. For a behavior often blamed—both in popular culture and in parts of the clinical literature—for driving obesity and metabolic dysfunction, the absence of a measurable link with these conventional risk indicators in hospitalized older adults is notable. The authors interpret this as evidence that emotional eating may relate to selected sociodemographic, behavioral, and clinical characteristics rather than to a broad cardiometabolic risk profile in this age group.

The study’s design imposes important limits on interpretation. As a cross-sectional analysis, it captures a single moment in time and cannot establish whether emotional eating causes any of the measured outcomes, or whether the observed associations reflect reverse pathways, confounding, or the physiological and pharmacological realities of hospitalization. The sample was also drawn from hospitalized patients, a group whose eating patterns, medication regimens, and metabolic values may differ from those of community-dwelling older adults, which limits generalizability. The findings should therefore be read as associations within one specific clinical setting rather than as statements about cause and effect or about older adults more broadly.

Even with those caveats, the results carry practical weight for anyone thinking about eating behavior in late life. If emotional eating in hospitalized older adults is not simply a proxy for obesity or dyslipidemia, then assuming a connection between body size, standard metabolic panels, and the urge to eat under distress may not hold in this population. The association with lipid-lowering medication use raises questions about whether patients on such therapies differ in underlying health status or in how they cope with chronic disease, while the link between elevated creatinine and lower emotional eating scores points toward kidney-related blood chemistry as a variable worth examining in future studies of this kind.

For now, the study adds a carefully measured data point to a field dominated by findings from younger cohorts: among hospitalized people aged 65 and older, the urge to eat under emotional distress appears to be woven into age, drinking behavior, medication use, and kidney-related blood markers—rather than into the tape measure or the lipid panel.

Subject of Research: Emotional eating and cardiometabolic risk indicators in hospitalized older adults

Article Title: Evaluation of emotional eating and selected cardiometabolic risk indicators among hospitalized older adults: a cross-sectional study

Article References: Umur, M. B., Sofuoglu, Z., & Eyuboglu, M. (2026). Evaluation of emotional eating and selected cardiometabolic risk indicators among hospitalized older adults: a cross-sectional study. BMC Geriatrics. https://doi.org/10.1186/s12877-026-08574-5

Image Credits: AI Generated

DOI: 10.1186/s12877-026-08574-5

Keywords: emotional eating, older adults, cardiometabolic risk, cross-sectional study, BMI, HDL cholesterol, creatinine, lipid-lowering medication, alcohol consumption, hospitalized patients, geriatric nutrition, BMC Geriatrics