The timing, frequency, and quality of food intake significantly impact health. Night eating, defined as consuming food between 22:00 and 04:00, has been linked to various health problems, including cancer and diabetes. However, its long-term effects on mortality remain unclear. This study aimed to address this gap by investigating the association between night eating characteristics (timing, frequency, and food quality) and the risk of all-cause, cancer, and diabetes mortality in a large sample of US adults. The existing literature presents conflicting findings regarding the impact of night eating. Some studies suggest that later eating times are associated with increased waist circumference and type 2 diabetes risk, while others show that higher daily eating frequency may be associated with lower all-cause mortality in individuals with diabetes. The impact of food quality consumed during night eating is also debated. Some studies suggest that high-energy-density foods at night increase diabetes and cancer risk, while others indicate that small, low-energy meals might be harmless or even beneficial. This lack of clarity underscores the need for a comprehensive study analyzing the multifaceted nature of night eating and its relationship with mortality, considering timing, frequency, and food quality simultaneously.

Existing research on night eating and its health consequences is inconsistent. Studies have linked later meal timing to increased waist circumference and higher type 2 diabetes mellitus risk. Conversely, some studies suggest a potential protective effect of higher daily eating frequency on all-cause mortality in individuals with diabetes. The influence of food quality consumed during night eating is also inconclusive, with some studies indicating a correlation between high-energy-density nighttime meals and increased risk of type 2 diabetes mellitus or cancer, while others report that small, low-energy-density nighttime meals might even be cardiometabolically protective. This disparity in findings highlights the need for a comprehensive analysis incorporating the timing, frequency, and food quality of night eating to clarify its impact on long-term health outcomes.

This study utilized data from the US National Health and Nutrition Examination Survey (NHANES) (2002-2018). The study population comprised 41,744 participants after excluding individuals with unreliable dietary recalls, those younger than 20, extreme energy intake values, missing mortality data, incomplete dietary information, and pregnant individuals at baseline. Dietary intake was assessed using the first 24-hour dietary recall interview, with nutrient values determined using the Food and Nutrient Database for Dietary Studies (FNDDS). Night eating was defined as food consumption between 22:00 and 04:00. The timing of night eating was categorized into seven groups (no eating, 22:00–23:00, 23:00–00:00, 00:00–01:00, 01:00–02:00, 02:00–03:00, and 03:00–04:00), frequency into three groups (no eating, one time, and two or more times), and food quality was assessed using latent class analysis based on energy intake, low-energy-density food intake, and high-energy-density food intake. The outcomes were all-cause, cancer, and diabetes mortality, identified using the National Death Index and ICD-10 codes. Covariates included age, sex, race/ethnicity, education, income, smoking status, drinking status, body mass index, physical activity, sleep hours, dietary energy intake, adherence to HEI-2015, dietary supplement use, glycohemoglobin, triglycerides, fasting glucose, total cholesterol, OGTT, hypertension, hyperlipidemia, CVD, diabetes, and cancer. Weighted Cox proportional hazards regression models were used to assess the associations, adjusting for various confounders in three models of increasing complexity. Subgroup analyses were conducted to explore potential effect modification. Linear regression analysis was used to examine the correlation between night eating characteristics and biochemical variables. Sensitivity analyses were performed to assess the robustness of the findings, including the exclusion of participants with over 50% of energy intake from night eating, the addition of an eating time category (09:00–10:00), and the additional adjustment for sleep times.

During a median follow-up of 8.7 years, 6066 deaths were documented (1381 from cancer, 206 from diabetes). Compared to no night eating, later timing of night eating was significantly associated with higher risks of all-cause (P-trend < 0.05) and diabetes mortality (P-trend < 0.05), with the highest risk observed for 00:00–01:00 and 23:00–00:00, respectively. The increased risk was not observed for 22:00–23:00. Higher frequency of night eating was associated with increased all-cause (P < 0.05) and diabetes mortality (P < 0.05). Night eating with high dietary energy density significantly increased the risk of all-cause (aHR 1.21, 95% CI 1.06–1.38) and diabetes mortality (aHR 1.97, 95% CI 1.13–3.45), but not in the low-dietary-energy-density group. Correlation analysis revealed positive associations between night eating and glycohemoglobin, fasting glucose, and OGTT, and a negative correlation with triglycerides (depending on timing, frequency, and food quality). Subgroup analyses showed no significant effect modification by age, sex, BMI, smoking status, drinking status, HEI-2015 score, or sleep hours. Sensitivity analyses confirmed the robustness of the findings.

This study provides novel evidence linking night eating characteristics to increased all-cause, cancer, and diabetes mortality risks. The findings extend previous research by showing a specific timing-dependent effect, with the highest mortality risk concentrated in the period between 23:00 and 01:00. The significant association between higher frequency and high-energy-density night eating with increased mortality underscores the importance of both the frequency and quality of nighttime food consumption. The observed correlations between night eating and altered glucose and lipid metabolism may explain these increased mortality risks. These associations might be mediated by disruptions in circadian rhythms, which have been linked to increased mortality risk, diabetes onset, and tumor progression. The protective effect observed with early night eating (before 23:00) and low dietary-energy-density foods suggests strategies to mitigate these risks. The results suggest a potential benefit of dietary and lifestyle interventions aimed at improving eating patterns and promoting a healthier relationship with food, especially during nighttime hours.

This study demonstrates a significant association between night eating patterns (later timing, higher frequency, and high dietary energy density) and increased all-cause, cancer, and diabetes mortality. The findings highlight the importance of eating before 23:00 and consuming low-energy-density foods during the night to reduce these risks. Future research should investigate the underlying mechanisms, specifically the role of circadian rhythm disruption, and explore the efficacy of interventions targeting nighttime eating habits.

This study's limitations include reliance on 24-hour dietary recall at baseline, which might not fully capture long-term eating habits. Although multiple confounders were considered, unmeasured confounding might exist. The limited number of mortality events in some analyses, particularly relating diabetes mortality and timing of night eating, might have impacted the statistical power and significance of certain findings. Future research could benefit from longitudinal studies with more frequent dietary assessments and a larger sample size to further enhance the precision and accuracy of the findings.