Circadian Rhythm Disruption and In-Hospital Mortality in ICU Patients with Obesity

Circadian Rhythm Disruption and In-Hospital Mortality in ICU Patients with Obesity | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Circadian Rhythm Disruption and In-Hospital Mortality in ICU Patients with Obesity Hong Xue, Ge Song, Lawrence Cheskin, Hua Min This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7934062/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract Objectives Circadian rhythms are frequently disrupted in ICU patients, and obesity may exacerbate this misalignment, worsening outcomes. This study evaluates associations between circadian rhythm parameters of vital signs and in-hospital mortality, assesses whether obesity modifies these relationships, and determines if integrating circadian parameters with obesity status improves predictive model performance. Methods Retrospective observational cohort study using MIMIC-IV database (2008–2019). Included 14 064 adult patients with ≥ 24-hour ICU stays after exclusions for short stays (n = 33 107), incomplete/implausible vital signs (n = 26 010), DNR/DNI orders, or multiple admissions (latest kept). Circadian parameters (MESOR, amplitude, acrophase) derived for heart rate (HR), body temperature (BT), and mean arterial pressure (MAP) using cosinor analysis from first 24-hour vital signs. Associations assessed using logistic regression, overall and BMI-stratified (normal < 25, overweight 25–29.9, class I 30–34.9, class II 35–39.9, class III ≥ 40 kg/m 2 ). Machine learning models (logistic regression, random forest, SVM, XGBoost) trained on 70/30 split and 10-fold cross-validation. Results Of 14 064 patients, 1 938 (13.8%) died in-hospital. HR MESOR (OR 1.027, 95% CI 1.024–1.030, p < 0.001), BT amplitude (OR 1.092, 95% CI 1.059–1.127, p < 0.001), MAP amplitude (OR 1.001, 95% CI 1.001–1.002, p < 0.001), class III obesity (OR 1.626, 95% CI 1.404–1.883, p < 0.001) increased mortality risk; MAP MESOR protective (OR 0.998, 95% CI 0.998–0.999, p < 0.001). HR MESOR was significant across BMI groups (ORs 1.021–1.040, p < 0.01 to < 0.001); BT acrophase in class III (OR 1.219, 95% CI 1.041–1.428, p = 0.014). The proposed random forest model achieved the best predictive performance (AUC 0.764; accuracy 85.8%; recall 88.2%; precision 87.1%), outperforming the baseline models. Conclusions Circadian rhythm parameters were significantly associated with in-hospital mortality in ICU patients, with effects modified by obesity status, and their inclusion significantly improved predictive performance. Health sciences/Risk factors Health sciences/Diseases/Endocrine system and metabolic diseases/Obesity Figures Figure 1 Figure 2 Introduction Circadian rhythms are approximately 24-hour cycles that regulate most physiological processes and synchronize internal functions with environmental cues. 1 In humans, the suprachiasmatic nucleus (SCN) acts as the central pacemaker, coordinating peripheral clocks and influencing sleep–wake patterns, hormone secretion, cardiovascular activity, and thermoregulation (Fig. 1 ). 2 – 4 Under normal conditions, circadian rhythms optimize physiological functions, providing a survival advantage by aligning with predictable daily tasks. 5 However, in critically ill patients, these rhythms are frequently disrupted. 6 , 7 While the underlying illness contributes to this dysregulation, intensive care unit (ICU)-specific factors, such as continuous artificial lighting, noise, mechanical ventilation, irregular feeding schedules, reduced mobility, and frequent interventions, exacerbate the loss of endogenous rhythmicity. 5 , 8 Consequently, alterations in circadian patterns of sleep, heart rate (HR), core body temperature (BT), and blood pressure may correlate with increased disease severity and poor outcomes. 6 , 9 Obesity adds another layer of complexity. It is a recognized risk factor for adverse outcomes in critical illness and is associated with altered circadian biology. Studies showed obesity is associated with dysregulated hormonal and metabolic rhythms, affecting HR variability, blood pressure control, and thermoregulation. 10 – 12 Individuals with obesity often display disrupted expression of clock genes, blunted hormonal oscillations, and reduced metabolic flexibility. 13 , 14 This bidirectional interaction suggests that ICU patients with obesity may be particularly sensitive to circadian rhythm disruption, compounding physiological instability and clinical weakening. Although emerging evidence links obesity and circadian rhythms to health outcomes 15 , 16 , little is known about their combined influence on outcomes. Only a small number of studies have examined circadian variation in vital signs as prognostic markers 17 , 18 , and none have systematically incorporated both circadian rhythm parameters and obesity into predictive models of mortality. This study addresses this gap by: (1) examining the impact of circadian rhythm parameters in vital signs (heart rate, body temperature, and mean arterial pressure) and obesity status on in-hospital mortality; (2) assessing whether obesity modifies the association between these circadian parameters and mortality; and (3) incorporating both elements into ML algorithms to enhance mortality prediction. We hypothesize that obesity amplifies the prognostic value of circadian disruptions and that their inclusion will improve model performance, with implications for personalized ICU management. Related Works Circadian rhythms regulate a wide range of physiological functions, including sleep–wake cycles, endocrine activity, cardiovascular dynamics, and body temperature. 19 , 20 In critically ill patients, these rhythms are often disrupted. Studies have shown that circadian patterns of core temperature and hormone secretion are diminished or lost in the ICU, and that the degree of disruption is linked to illness severity and outcomes. 2 , 6 It is no surprise that the ICU environment itself, which is characterized by continuous light exposure, noise, sedation, irregular feeding times, and frequent interventions, further contributes to circadian misalignment. 5 , 8 Despite these challenges, some circadian rhythmicity in vital signs can persist during ICU stays. Recent work by Davidson et al. 17 and Yang et al. 18 demonstrated that circadian rhythmicity in vital signs remains detectable during ICU stays, and these patterns may hold prognostic value. It suggests that, even during critical illness, crunching the numbers on these rhythms could sharpen risk assessments and possibly lead to better care. On the other hand, the relationship between obesity and circadian rhythms is gaining more attention lately. It has been associated with altered clock gene expression, blunted hormonal oscillations, and impaired metabolic regulation. 10 , 13 Individuals with obesity appear to be more sensitive to circadian misalignment, showing disturbances in cardiovascular control, thermoregulation, and metabolic flexibility. 11 , 14 This interaction between obesity and circadian disruption has been implicated in insulin resistance, inflammation, and cardiovascular dysregulation. 12 , 15 Despite the current evidence, research that integrates circadian parameters and obesity in predicting ICU outcomes is limited. To date, only one published study has included circadian metrics in a mortality prediction model for ICU patients. 17 Comprehensive models that incorporate multiple vital sign rhythms, alongside obesity status, are still lacking. Methods Data Data wew extracted from the Medical Information Mart for Intensive Care (MIMIC-IV), a large, publicly available database containing electronic health records of patients admitted to the intensive care units or emergency department of the Beth Israel Deaconess Medical Center between 2008 and 2019. 21 The database includes detailed information on demographics, vital signs, laboratory values, medications, procedures, diagnoses, and outcomes. This study linked patient information, admission, ICD diagnoses, ICU stays, lab, and chart events. Study Sample The study sample consists of adult patients (18 years and older) with an ICU stay of at least 24 hours. For patients with multiple admissions, only the latest ICU admission was retained to avoid correlated observations. We required complete vital sign data for at least three daytime (06:00–19:59) and three nighttime (20:00–05:59) measurements of heart rate, body temperature, and mean arterial pressure (MAP) during the first 24 hours of ICU admission. Patients with do-not-resuscitate (DNR), do-not-intubate (DNI), or “comfort measures only” codes or similar orders were excluded to focus on those receiving active treatment interventions. Patients with measurements of vital signs outside of the broad physiological bounds (50 mmHg < MAP < 183 mmHg, 30 bpm < HR < 240 bpm, 34°C < T < 40°C) were also excluded. Variables The primary outcome was in-hospital mortality, coded as a binary variable: died in the hospital (1) or survived (0). The primary independent variables were circadian rhythm parameters of three vital signs: body temperature (BT), heart rate (HR), and mean arterial pressure (MAP). Circadian rhythm parameters were estimated using cosinor analysis, which characterizes rhythmicity by three components: MESOR (midline estimating statistic of rhythm, representing the rhythm-adjusted mean), amplitude (half the difference between peak and trough, indicating variation around the MESOR), and acrophase (timing of the rhythm peak within the 24-hour cycle). Obesity status was defined by BMI-based categories: normal weight (< 25), overweight (25–29.9), class I obesity (30–34.9), class II obesity (35–39.9), and class III obesity (≥ 40). Covariates included age, sex, race/ethnicity, and the Charlson Comorbidity Index (CCI), selected for their established influence on ICU outcomes and physiological regulation. Analysis We first applied cosinor analysis to each patient’s 24-hour vital sign records to derive MESOR, amplitude, and acrophase for HR, BT, and MAP. Cosinor analysis is a statistical technique commonly used in chronobiology to analyze rhythmic (cyclical) data. This method is particularly useful for assessing patterns in biological processes that demonstrate circadian rhythms, such as hormone levels, body temperature, and heart rate, which typically follow a 24-hour cycle. Cosinor analysis is used to fit a cosine wave to the time series, with the model specified as $$\:Y=M+A*\text{c}\text{o}\text{s}\left[2\pi\:\right(T-\varphi\:)/24]$$ where Y is the vital sign value (heart rate, body temperature, or MAP), M represents the MESOR (midline estimating statistic of rhythm), A the amplitude, T the time of day in hours, and ϕ the acrophase of the circadian rhythm (Fig. 2 ). According to the cosinor model, the three parameters MESOR, amplitude, and acrophase characterize the circadian rhythm of heart rate (HR), body temperature (BT), and mean arterial pressure (MAP). Logistic regression models were then used to examine associations between circadian parameters, obesity categories, and in-hospital mortality, adjusting for covariates. Stratified analyses were conducted to explore whether associations differed across obesity subgroups. For prediction, data were randomly split into a training set (70%) and a testing set (30%). Two sets of models were built. Baseline models include obesity status and covariates only. Proposed models include circadian parameters in addition to baseline variables. We evaluated four algorithms: logistic regression, random forest, support vector machine, and extreme gradient boosting. To address class imbalance in the training data, we applied the Synthetic Minority Oversampling Technique. Model performance was assessed in the testing set using the area under the receiver operating characteristic curve (AUC), accuracy, precision, and recall. Ten-fold stratified cross-validation was used to reduce overfitting and ensure model robustness. Results From 73 181 ICU patients in MIMIC-IV, 40 074 adults had at least one ICU admission lasting 24 hours. After applying exclusion criteria for incomplete or implausible vital sign records, 14,064 patients remained in the analytic cohort (Fig. 3). Of these, 1 938 (13.8%) died before hospital discharge. Baseline characteristics are summarized in Table 1 . Among the 12 126 patients who survived, the average age was 63.05 years (SD 17.86), and about 45.5% were women. For the 1 938 who died, the average age was higher at 70.40 (SD 14.59), with 47.8% women. Patients who died also had a higher Charlson Comorbidity Index score (4.28 on average versus 2.56 for survivors). The distribution of obesity categories was similar across groups, although a higher proportion of those who died had class III obesity (3.5% vs. 4.7%). Table 1 Characteristics summary N, (%) Characteristics Survived (N = 12 126) Died In Hospital (N = 1 938) Sex Female 5 522 (45.5%) 926 (47.8%) Male 6 604 (54.5%) 1 012 (52.2%) Age groups 18 ≤ Age < 29 682 (5.6%) 27 (1.4%) 30 ≤ Age < 39 686 (5.7%) 33 (1.7%) 40 ≤ Age < 49 1 275 (10.5%) 115 (5.9%) 50 ≤ Age < 59 2 146 (17.7%) 253 (13.1%) 60 ≤ Age < 69 2 527 (20.8%) 415 (21.4%) 70 ≤ Age < 79 2 251 (18.6%) 472 (24.4%) Age ≥ 80 2 559 (21.1%) 623 (32.1) Race Non-Hispanic White 8 479 (69.9%) 1 315 (67.9%) Non-Hispanic Black 1 107 (9.1%) 179 (9.2%) Hispanic 469 (3.9%) 53 (2.7%) Asian 322 (2.7%) 78 (4.0%) Other Race 1 749 (14.4%) 313 (16.2%) CCI, Mean (SD) 2.56 (2.37) 4.28 (2,65) Obesity class Normal weight 4 171 (34.4%) 690 (35.6%) Overweight 4 317 (35.6%) 733 (37.8%) Obesity I 2 280 (18.8%) 339 (17.5%) Obesity II 788 (6.5%) 109 (5.6%) Obesity III 570 (4.7%) 68 (3.5%) CCI, Charlson Comorbidity Index; SD, standard deviation. Multivariable logistic regression identified several predictors of mortality (Table 2 ). Having class III obesity was strongly associated with in-hospital death, with an odds ratio (OR) of 1.626 (95%CI 1.404–1.883, p < 0.001). Circadian parameters of vital signs also carried prognostic value. A higher MESOR of heart rate was associated with greater mortality risk (OR 1.027, 95% CI 1.024–1.030, p < 0.001), and larger amplitude of body temperature predicted higher odds of death (OR 1.092, 95% CI 1.059–1.127, p < 0.001). Conversely, higher MESOR of mean arterial pressure was protective (OR 0.998, 95% CI 0.998–0.999, p < 0.001). Other circadian parameters, including acrophase of heart rate, body temperature, and MAP, were not significant in the overall cohort. Table 2 Logistic regression results Key independent variables Odds ratio SE 95% CI P-value Obesity class Normal weight Ref. Overweight 0.911 0.092 (0.747, 1.111) 0.356 Obesity I 0.835 0.106 (0.652, 1.070) 0.155 Obesity II 0.809 0.154 (0.556, 1.175) 0.265 Obesity III 1.626 0.122 (1.404, 1.883) 0.000 Heart rate Mesor 1.027 0.002 (1.024, 1.030) 0.000 Amplitude 1.004 0.003 (0.998, 1.011) 0.196 Acrophase 0.993 0.015 (0.96,4 1.022) 0.628 Body Temperature Mesor 1.017 0.015 (0.988, 1.047) 0.261 Amplitude 1.092 0.017 (1.059, 1.127) 0.000 Acrophase 1.003 0.016 (0.972, 1.034) 0.871 Mean arterial pressure Mesor 0.998 0.000 (0.998, 0.999) 0.000 Amplitude 1.001 0.000 (1.001, 1.002) 0.000 Acrophase 1.009 0.015 (0.980, 1.038) 0.547 SE, standard error; CI: confidence interval. Stratified analyses by obesity category showed heterogeneity in these associations (Table 3 ). Heart rate MESOR consistently predicted higher mortality in all groups, with odds ratios ranging from 1.028 among patients with class II obesity (95%CI 1.022–1.034, p < 0.01) to 1.040 among patients with class III obesity (95% CI 1.022–1.059, p < 0.001). Body temperature amplitude was significantly associated with mortality among patients with overweight (OR = 1.146, 95% CI 1.076–1.220, p < 0.001) and among patients with class I obesity (OR 1.108, 95% CI 1.009–1.217, p < 0.05), but not among patients with normal weight or higher obesity categories. MAP MESOR was inversely associated with mortality among patients with normal weight (OR = 0.999, 95% CI 0.997–1.001, p < 0.01), among patients with overweight (OR = 0.999, 95% CI 0.997–1.001, p < 0.1), and among patients with class I obesity (OR = 0.998, 95%CI 0.996-1.000, p < 0.05). Among patients with class III obesity, both heart rate MESOR (OR = 1.040, 95% CI 1.022–1.059, p < 0.001) and body temperature acrophase (OR = 1.219, 95% CI 0.979–1.518, p < 0.05) were significantly associated with mortality, suggesting that both the magnitude and timing of circadian variation become clinically relevant among patients with severe obesity. Table 3 Stratified analysis by obesity status OR (95% CI) Obesity status Variables Normal weight Overweight Obesity I Obesity II Obesity III Heart rate MESOR 1.028 (1.022, 1.034)**** 1.026 (1.020, 1.032)**** 1.026 (1.018, 1.034)**** 1.021 (1.007, 1.035)*** 1.040 (1.022, 1.059)**** Amplitude 1.008 (0.996, 1.020) 0.999 (0.989, 1.009) 0.998 (0.982, 1.014) 1.027 (0.997, 1.058). 0.999 (0.962, 1.037) Acrophase 1.009 (0.961, 1.060) 0.964 (0.922, 1.008) 1.044 (0.969, 1.125) 0.963 (0.851, 1.090) 0.911 (0.783, 1.059) Body temperature MESOR 0.969 (0.926, 1.014) 1.016 (0.966, 1.069) 1.054 (0.973, 1.142) 1.008 (0.857, 1.186) 0.992 (0.830, 1.186) Amplitude 1.126 (1.066, 1.190)**** 1.146 (1.076, 1.220)**** 1.108 (1.009, 1.217)** 0.926 (0.789, 1.087) 1.073 (0.860, 1.339) Acrophase 1.006 (0.956, 1.059) 0.972 (0.926, 1.021) 1.000 (0.930, 1.075) 1.054 (0.919, 1.209) 1.219 (0.979, 1.518)** Mean arterial pressure MESOR 0.999 (0.997, 1.001)*** 0.999 (0.997, 1.001)* 0.998 (0.996, 1.000)* 0.997 (0.991, 1.003) 0.999 (0.993, 1.005) Amplitude 1.001 (1.001, 1.001)** 1.002 (1.000, 1.004)**** 1.001 (0.999, 1.003) 1.000 (0.994, 1.006) 1.002 (0.996, 1.008) Acrophase 0.999 (0.953, 1.047) 1.023 (0.974, 1.074) 0.969 (0.907, 1.036) 1.026 (0.905, 1.163) 1.099 (0.919, 1.314) ****, p-value < 0.001; ***, p-value < 0.01; **, p-value < 0.05; *, p-value < 0.1. CI, confidence interval. The training data includes 70% (9 844) of patients, and 30% (4 220) of patients are in testing data. Prediction model performance results are presented in Table 4 . Incorporating circadian parameters consistently improved predictive accuracy compared with baseline models. In the proposed models, Random Forest achieved the best performance, with an area under the receiver operating characteristic curve (AUC) of 0.764, an accuracy of 85.8%, a recall of 88.2%, and a precision of 87.1%. Logistic regression also improved substantially with the addition of circadian variables (AUC 0.754 compared with 0.701 in the baseline model). XGBoost achieved an AUC of 0.753 with strong recall and precision, while support vector machine models showed modest but consistent improvements over the baseline. Together, these results demonstrate that circadian rhythm parameters enhance the prediction of ICU mortality, particularly when considered alongside obesity status. Table 4 Prediction model performances Model AUC Accuracy Recall Precision Baseline model Logistic regression 0.701 0.682 0.680 0.817 Random forest 0.711 0.793 0.801 0.778 Extreme gradient boosting 0.707 0.753 0.775 0.790 Support vector machine 0.702 0.710 0.731 0.801 Proposed model Logistic regression 0.754 0.684 0.681 0.835 Random forest 0.764 0.858 0.882 0.871 Extreme gradient boosting 0.753 0.765 0.807 0.810 Support vector machine 0.727 0.721 0.746 0.815 AUC: area under the ROC curve. Discussion In this large cohort of ICU patients, we found that circadian rhythm parameters of vital signs, together with obesity status, were independently associated with in-hospital mortality. Heart rate MESOR emerged as a consistent predictor across all BMI categories, while body temperature amplitude and mean arterial pressure (MAP) MESOR were significant in selected subgroups. Notably, among patients with class III obesity, both heart rate MESOR and body temperature acrophase were strong predictors, underscoring that the timing as well as the magnitude of circadian variation may play a clinically meaningful role among patients with severe obesity. Importantly, incorporating circadian parameters into machine learning models improved predictive performance compared with conventional models based only on demographics, comorbidities, and obesity status. Our findings extend previous work on circadian disruption in the ICU. Prior studies have demonstrated that circadian rhythmicity persists in vital signs during critical illness, and that attenuated or phase-shifted rhythms are associated with adverse outcomes. 2 , 17 , 18 However, most prior analyses were limited to small cohorts or single parameters such as body temperature. By applying cosinor analysis to multiple vital signs in a large ICU population, our study provides robust evidence that circadian metrics carry prognostic value beyond conventional clinical variables. The modifying role of obesity is also noteworthy. Experimental and epidemiologic studies have established that obesity alters circadian biology through changes in clock gene expression, hormonal oscillations, and metabolic regulation. 10 , 13 Our stratified results suggest that these disturbances amplify the prognostic importance of circadian rhythms in critically ill patients. For example, body temperature amplitude predicted mortality among patients with overweight and among patients with class I obesity. However, this relationship weakened among patients with higher obesity levels, where instead the timing of rhythms (acrophase) became more relevant. This pattern may reflect impaired thermoregulatory capacity in severe obesity, leading to reliance on timing shifts rather than amplitude changes as markers of physiological instability. From a clinical standpoint, these findings highlight the potential of circadian parameters as simple, low-cost indicators for risk stratification in the ICU. Vital signs are continuously collected in most modern ICUs, and extracting circadian rhythm measures from these data could provide a dynamic, noninvasive means of assessing physiologic reserve and vulnerability. The consistent predictive improvement observed across multiple machine learning models supports the feasibility of integrating circadian metrics into real-time decision support systems. Such tools could help clinicians identify high-risk patients earlier, tailor interventions more precisely, and design ICU environments that minimize circadian disruption. However, some limitations should be acknowledged. First, the data were sourced from a single academic center, which may limit generalizability to other patient populations or care practices. Second, while cosinor analysis provides a robust framework for detecting circadian rhythmicity, ICU vital signs are influenced by multiple external factors, including sedation, mechanical ventilation, and therapeutic interventions, which may confound the rhythms observed. Third, despite adjusting for major covariates, residual confounding from unmeasured variables, such as medication timing, staff shift changes, or sleep disruption, cannot be excluded. Finally, the observational design precludes causal inference. We will extend our findings further to elucidate the role of circadian rhythms in critical care. Planned subgroup analyses will stratify patients by age groups (e.g., 75 years) to explore whether the associations between circadian parameters and mortality vary across the lifespan, potentially revealing age-specific vulnerabilities in circadian disruption among ICU patients with obesity. Additionally, incorporating established severity scores, such as the Sequential Organ Failure Assessment (SOFA) or the Acute Physiology and Chronic Health Evaluation II (APACHE II), into our predictive models will enable more comprehensive risk adjustment. This should provide a clearer understanding of how circadian parameters contribute independently, beyond the patient's initial level of illness upon entering the ICU. In summary, circadian rhythm parameters of vital signs were independently associated with in-hospital mortality, and their effects varied across obesity categories. Heart rate MESOR was a consistently significant predictor, while body temperature amplitude and MAP MESOR showed differential associations by BMI. Among patients with class III obesity, both the magnitude and timing of circadian variation emerged as important risk markers. Incorporating circadian metrics into machine learning models substantially improved predictive performance compared with conventional approaches. These findings underscore the value of circadian analysis in critical care and highlight the interaction between chronobiology and obesity in shaping outcomes. Routine ICU vital sign data can provide actionable circadian parameters, offering a low-cost, scalable approach to enhance risk stratification and guide individualized care. Future multicenter studies and interventional trials are needed to validate these observations and to determine whether preserving circadian alignment can improve outcomes in critically ill patients. Declarations Competing interests The authors declare no competing interests. Author Contributions GS was responsible for designing the study protocol, conducting data extraction and analysis from the MIMIC-IV database, performing analyses, interpreting results, and drafting the manuscript. LJC provided expertise on obesity and circadian rhythms and reviewed clinical interpretations. HM assisted with statistical modeling, including machine learning algorithms, and revised the manuscript for intellectual content. HX oversaw the overall project, contributed to methodology development, supervised data analysis, and interpreted findings. All authors approved the final version and agree to be accountable for all aspects of the work. Acknowledgements This study utilized publicly available data from the MIMIC-IV database and received no external funding or grants. The authors are solely responsible for the content. Data Availability The raw de-identified ICU data were obtained from MIMIC-IV (version 2.2) and are freely available to qualified researchers after completing the required training and data use agreement at https://physionet.org/content/mimiciv/2.2/ . All relevant raw data supporting the findings of this study are freely accessible for replication purposes without breaching ethical standards. References Refinetti R. The circadian rhythm of body temperature. Front Biosci (Landmark Ed). 2010;15(2):564–94. Gazendam JAC, Van Dongen HPA, Grant DA, Freedman NS, Zwaveling JH, Schwab RJ. Altered circadian rhythmicity in patients in the ICU. Chest. 2013;144(2):483–9. Logan RW, Sarkar DK. Circadian nature of immune function. Molecular and Cellular Endocrinology. 2012;349(1):82–90. Morrison SF, Nakamura K. Central mechanisms for thermoregulation. Annual Review of Physiology. 2019;81(Volume 81, 2019):285–308. McKenna H, van der Horst GTJ, Reiss I, Martin D. Clinical chronobiology: A timely consideration in critical care medicine. Crit Care. 2018;22(1):124. Korompeli A, Muurlink O, Kavrochorianou N, Katsoulas T, Fildissis G, Baltopoulos G. Circadian disruption of ICU patients: A review of pathways, expression, and interventions. J Crit Care. 2017;38:269–77. Egi M, Bellomo R, Stachowski E, French CJ, Hart G, Stow P. Circadian rhythm of blood glucose values in critically ill patients. Crit Care Med. 2007;35(2):416–21. Telias I, Wilcox ME. Sleep and circadian rhythm in critical illness. Critical Care. 2019;23(1):82. Gao Y, Wang Q, Li J, et al. Impact of Mean Arterial Pressure Fluctuation on Mortality in Critically Ill Patients. Crit Care Med. 2018;46(12):e1167-e1174. Chaput JP, McHill AW, Cox RC, Broussard JL, Dutil C, da Costa BGG, et al. The role of insufficient sleep and circadian misalignment in obesity. Nature Reviews Endocrinology. 2023;19(2):82–97. Summa KC, Turek FW. Chronobiology and obesity: Interactions between circadian rhythms and energy regulation. Adv Nutr. 2014;5(3):312S-9S. Fatima N, Rana S. Metabolic implications of circadian disruption. Pflugers Arch. 2020;472(5):513–26. Garaulet M, Ordovás JM, Madrid JA. The chronobiology, etiology and pathophysiology of obesity. Int J Obes (Lond). 2010;34(12):1667–83. Halder SK, Melkani GC. The Interplay of Genetic Predisposition, Circadian Misalignment, and Metabolic Regulation in Obesity. Curr Obes Rep. 2025;14(1):21. Bishehsari F, Voigt RM, Keshavarzian A. Circadian rhythms and the gut microbiota: From the metabolic syndrome to cancer. Nat Rev Endocrinol. 2020;16(12):731–9. Boer-Martins L, Figueiredo VN, Demacq C, Martins LC, Consolin-Colombo F, Figueiredo MJ, et al. Relationship of autonomic imbalance and circadian disruption with obesity and type 2 diabetes in resistant hypertensive patients. Cardiovasc Diabetol. 2011;10:24. Davidson S, Villarroel M, Harford M, Finnegan E, Jorge J, Young D, et al. Day-to-day progression of vital-sign circadian rhythms in the intensive care unit. Crit Care. 2021;25(1):156. Yang Z, Xie X, Zhang X, Li L, Bai R, Long H, et al. Circadian rhythms of vital signs are associated with in-hospital mortality in critically ill patients: A retrospective observational study. Chronobiol Int. 2023;40(3):262–71. Patke A, Young MW, Axelrod S. Molecular mechanisms and physiological importance of circadian rhythms. Nat Rev Mol Cell Biol. 2020;21(2):67–84. Robinson I, Reddy AB. Molecular mechanisms of the circadian clockwork in mammals. FEBS Lett. 2014;588(15):2477–83. Johnson A, Bulgarelli L, Pollard T, Horng S, Celi LA, Mark R. MIMIC-IV [Internet]. PhysioNet; 2023. Available from: https://physionet.org/content/mimiciv/2.2/ Additional Declarations There is NO conflict of interest to disclose Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: revise 26 Nov, 2025 Review # 2 received at journal 20 Nov, 2025 Review # 1 received at journal 05 Nov, 2025 Reviewer # 2 agreed at journal 30 Oct, 2025 Reviewer # 1 agreed at journal 29 Oct, 2025 Reviewers invited by journal 27 Oct, 2025 Submission checks completed at journal 27 Oct, 2025 First submitted to journal 24 Oct, 2025 Unknown event 24 Oct, 2025 Editor assigned by journal 23 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7934062","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":535730892,"identity":"eb38b3f6-05ee-4dbc-8a78-98199cd436ec","order_by":0,"name":"Hong Xue","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApklEQVRIiWNgGAWjYBACAxAhUQHhSBCvxeIMyVoq20jRYi6RY/jg5rzaaIMDzAdv8xCjxbLnjLHhzG3HczccYEu2JkqLwfEeM2nJbceAWnjMpInTcpjH/PffOSAt/N+I1AK0hUGyoQZkCxuRWs4cK5aQOHYgd+ZhNmPLOURpuZG88YNETV1u3/HmhzfeEKMFCg4zMDCToBwE6khUPwpGwSgYBSMKAAAWUTP0e4oFVAAAAABJRU5ErkJggg==","orcid":"","institution":"George Mason University","correspondingAuthor":true,"prefix":"","firstName":"Hong","middleName":"","lastName":"Xue","suffix":""},{"id":535730893,"identity":"56c3b1f7-eb66-4164-8995-1eba9d67649b","order_by":1,"name":"Ge Song","email":"","orcid":"","institution":"George Mason University","correspondingAuthor":false,"prefix":"","firstName":"Ge","middleName":"","lastName":"Song","suffix":""},{"id":535730894,"identity":"2be775db-ea99-4431-9455-eb6772821cf3","order_by":2,"name":"Lawrence Cheskin","email":"","orcid":"https://orcid.org/0000-0003-0262-1030","institution":"George Mason University","correspondingAuthor":false,"prefix":"","firstName":"Lawrence","middleName":"","lastName":"Cheskin","suffix":""},{"id":535730895,"identity":"ba272e80-11af-492c-9466-8413f2031340","order_by":3,"name":"Hua Min","email":"","orcid":"https://orcid.org/0000-0003-2422-0043","institution":"George Mason University","correspondingAuthor":false,"prefix":"","firstName":"Hua","middleName":"","lastName":"Min","suffix":""}],"badges":[],"createdAt":"2025-10-23 16:31:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7934062/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7934062/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95499963,"identity":"7526cae7-0214-4879-b51e-ef0ad60bbc56","added_by":"auto","created_at":"2025-11-10 05:17:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":308756,"visible":true,"origin":"","legend":"","description":"","filename":"Figures10212025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/ec6b1bfafd93085189c3805f.docx"},{"id":95527682,"identity":"5cee990f-1725-4388-8559-e87ca5428492","added_by":"auto","created_at":"2025-11-10 10:14:29","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37102,"visible":true,"origin":"","legend":"","description":"","filename":"CircadianRhythmDisruptionandInHospitalMortalityinICUPatientswithObesity10212025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/bb347c8e3a0fbe4713119a32.docx"},{"id":95528079,"identity":"b7d060c9-1b74-4ee4-b44e-860052effe7c","added_by":"auto","created_at":"2025-11-10 10:15:32","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30207,"visible":true,"origin":"","legend":"","description":"","filename":"Tables10212025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/d1b18b89f14c5cc2c4bb2ae5.docx"},{"id":95528894,"identity":"02e196ba-24c3-45f2-ad44-b88c10efb48f","added_by":"auto","created_at":"2025-11-10 10:16:34","extension":"json","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6565,"visible":true,"origin":"","legend":"","description":"","filename":"2025IJO01880.json","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/229d06146fa9124591dbbacb.json"},{"id":95528688,"identity":"816c325d-c01b-41e9-bd8c-5eb671d584df","added_by":"auto","created_at":"2025-11-10 10:16:23","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91908,"visible":true,"origin":"","legend":"","description":"","filename":"2025IJO018800enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/313275e7c29625009e54f3f7.xml"},{"id":95499969,"identity":"9356f3ff-cc43-47c8-860d-c0e035378c11","added_by":"auto","created_at":"2025-11-10 05:17:37","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55894,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/188372fb1f79929f69941198.png"},{"id":95528853,"identity":"08ebe17a-d461-4a01-b4b9-a42a71c7fba2","added_by":"auto","created_at":"2025-11-10 10:16:32","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":92543,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/f1fcdc9224ffd23cbbc5d894.png"},{"id":95499971,"identity":"5992b7d5-fce4-47bf-9a4c-a52fea2e0174","added_by":"auto","created_at":"2025-11-10 05:17:38","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86664,"visible":true,"origin":"","legend":"","description":"","filename":"2025IJO018800structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/416ef7489bd6e6845321bfe1.xml"},{"id":95528407,"identity":"0a8902c0-4d5c-47d2-b86a-e29e4af86081","added_by":"auto","created_at":"2025-11-10 10:16:02","extension":"html","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96605,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/5225a3920e424aa29f5b1983.html"},{"id":95499961,"identity":"50a1b2ac-5527-43f3-8792-3724f52e235a","added_by":"auto","created_at":"2025-11-10 05:17:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":278082,"visible":true,"origin":"","legend":"\u003cp\u003eConceptual framework of circadian rhythms and cosinor analysis.\u003c/p\u003e\n\u003cp\u003eNote:\u003c/p\u003e\n\u003cp\u003e1. The suprachiasmatic nucleus (SCN), the central circadian clock, is entrained by light and other stimuli. The SCN coordinates the activity of peripheral clocks and is, in turn, affected by them. These circadian clocks regulate various physiologic processes.\u003c/p\u003e\n\u003cp\u003e2. Cosinor curve shows the mesor, amplitude, and acrophase from the cosinor model. The mesor is the midline estimating statistic of rhythm. The amplitude is the difference between the maximum and the average. The acrophase is the timing within a given cycle at which the peak of the cycle occurs\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/e9fa354de05b5528d84b9a7b.png"},{"id":95499966,"identity":"88af6d6a-c23c-4cb1-bd85-11b64769e4e1","added_by":"auto","created_at":"2025-11-10 05:17:37","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":394487,"visible":true,"origin":"","legend":"\u003cp\u003eStudy cohort flowchart\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/982c9fa720538d31bc435bf3.jpeg"},{"id":95531650,"identity":"b5ae14a3-2a4e-447b-aa11-5eb86d5d6aba","added_by":"auto","created_at":"2025-11-10 10:23:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1264655,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7934062/v1/d439494b-56e6-49d4-90d4-96ac704a9bc6.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Circadian Rhythm Disruption and In-Hospital Mortality in ICU Patients with Obesity","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCircadian rhythms are approximately 24-hour cycles that regulate most physiological processes and synchronize internal functions with environmental cues.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e In humans, the suprachiasmatic nucleus (SCN) acts as the central pacemaker, coordinating peripheral clocks and influencing sleep–wake patterns, hormone secretion, cardiovascular activity, and thermoregulation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e–\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Under normal conditions, circadian rhythms optimize physiological functions, providing a survival advantage by aligning with predictable daily tasks.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e However, in critically ill patients, these rhythms are frequently disrupted.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e While the underlying illness contributes to this dysregulation, intensive care unit (ICU)-specific factors, such as continuous artificial lighting, noise, mechanical ventilation, irregular feeding schedules, reduced mobility, and frequent interventions, exacerbate the loss of endogenous rhythmicity.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e Consequently, alterations in circadian patterns of sleep, heart rate (HR), core body temperature (BT), and blood pressure may correlate with increased disease severity and poor outcomes.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eObesity adds another layer of complexity. It is a recognized risk factor for adverse outcomes in critical illness and is associated with altered circadian biology. Studies showed obesity is associated with dysregulated hormonal and metabolic rhythms, affecting HR variability, blood pressure control, and thermoregulation.\u003csup\u003e\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e–\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e Individuals with obesity often display disrupted expression of clock genes, blunted hormonal oscillations, and reduced metabolic flexibility.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e This bidirectional interaction suggests that ICU patients with obesity may be particularly sensitive to circadian rhythm disruption, compounding physiological instability and clinical weakening.\u003c/p\u003e\u003cp\u003eAlthough emerging evidence links obesity and circadian rhythms to health outcomes\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, little is known about their combined influence on outcomes. Only a small number of studies have examined circadian variation in vital signs as prognostic markers\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, and none have systematically incorporated both circadian rhythm parameters and obesity into predictive models of mortality.\u003c/p\u003e\u003cp\u003eThis study addresses this gap by: (1) examining the impact of circadian rhythm parameters in vital signs (heart rate, body temperature, and mean arterial pressure) and obesity status on in-hospital mortality; (2) assessing whether obesity modifies the association between these circadian parameters and mortality; and (3) incorporating both elements into ML algorithms to enhance mortality prediction. We hypothesize that obesity amplifies the prognostic value of circadian disruptions and that their inclusion will improve model performance, with implications for personalized ICU management.\u003c/p\u003e\n\u003ch3\u003eRelated Works\u003c/h3\u003e\n\u003cp\u003eCircadian rhythms regulate a wide range of physiological functions, including sleep–wake cycles, endocrine activity, cardiovascular dynamics, and body temperature.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e In critically ill patients, these rhythms are often disrupted. Studies have shown that circadian patterns of core temperature and hormone secretion are diminished or lost in the ICU, and that the degree of disruption is linked to illness severity and outcomes.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e It is no surprise that the ICU environment itself, which is characterized by continuous light exposure, noise, sedation, irregular feeding times, and frequent interventions, further contributes to circadian misalignment.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eDespite these challenges, some circadian rhythmicity in vital signs can persist during ICU stays. Recent work by Davidson et al.\u003csup\u003e17\u003c/sup\u003e and Yang et al.\u003csup\u003e18\u003c/sup\u003e demonstrated that circadian rhythmicity in vital signs remains detectable during ICU stays, and these patterns may hold prognostic value. It suggests that, even during critical illness, crunching the numbers on these rhythms could sharpen risk assessments and possibly lead to better care.\u003c/p\u003e\u003cp\u003eOn the other hand, the relationship between obesity and circadian rhythms is gaining more attention lately. It has been associated with altered clock gene expression, blunted hormonal oscillations, and impaired metabolic regulation.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Individuals with obesity appear to be more sensitive to circadian misalignment, showing disturbances in cardiovascular control, thermoregulation, and metabolic flexibility.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e This interaction between obesity and circadian disruption has been implicated in insulin resistance, inflammation, and cardiovascular dysregulation.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eDespite the current evidence, research that integrates circadian parameters and obesity in predicting ICU outcomes is limited. To date, only one published study has included circadian metrics in a mortality prediction model for ICU patients.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e Comprehensive models that incorporate multiple vital sign rhythms, alongside obesity status, are still lacking.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eData\u003c/h2\u003e\u003cp\u003eData wew extracted from the Medical Information Mart for Intensive Care (MIMIC-IV), a large, publicly available database containing electronic health records of patients admitted to the intensive care units or emergency department of the Beth Israel Deaconess Medical Center between 2008 and 2019.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e The database includes detailed information on demographics, vital signs, laboratory values, medications, procedures, diagnoses, and outcomes. This study linked patient information, admission, ICD diagnoses, ICU stays, lab, and chart events.\u003c/p\u003e\u003ch3\u003eStudy Sample\u003c/h3\u003e\u003cp\u003eThe study sample consists of adult patients (18 years and older) with an ICU stay of at least 24 hours. For patients with multiple admissions, only the latest ICU admission was retained to avoid correlated observations. We required complete vital sign data for at least three daytime (06:00–19:59) and three nighttime (20:00–05:59) measurements of heart rate, body temperature, and mean arterial pressure (MAP) during the first 24 hours of ICU admission. Patients with do-not-resuscitate (DNR), do-not-intubate (DNI), or “comfort measures only” codes or similar orders were excluded to focus on those receiving active treatment interventions. Patients with measurements of vital signs outside of the broad physiological bounds (50 mmHg \u0026lt; MAP \u0026lt; 183 mmHg, 30 bpm \u0026lt; HR \u0026lt; 240 bpm, 34°C \u0026lt; T \u0026lt; 40°C) were also excluded.\u003c/p\u003e\u003ch3\u003eVariables\u003c/h3\u003e\u003cp\u003eThe primary outcome was in-hospital mortality, coded as a binary variable: died in the hospital (1) or survived (0).\u003c/p\u003e\u003cp\u003eThe primary independent variables were circadian rhythm parameters of three vital signs: body temperature (BT), heart rate (HR), and mean arterial pressure (MAP). Circadian rhythm parameters were estimated using cosinor analysis, which characterizes rhythmicity by three components: MESOR (midline estimating statistic of rhythm, representing the rhythm-adjusted mean), amplitude (half the difference between peak and trough, indicating variation around the MESOR), and acrophase (timing of the rhythm peak within the 24-hour cycle). Obesity status was defined by BMI-based categories: normal weight (\u0026lt; 25), overweight (25–29.9), class I obesity (30–34.9), class II obesity (35–39.9), and class III obesity (≥ 40).\u003c/p\u003e\u003cp\u003eCovariates included age, sex, race/ethnicity, and the Charlson Comorbidity Index (CCI), selected for their established influence on ICU outcomes and physiological regulation.\u003c/p\u003e\u003ch3\u003eAnalysis\u003c/h3\u003e\u003cp\u003eWe first applied cosinor analysis to each patient’s 24-hour vital sign records to derive MESOR, amplitude, and acrophase for HR, BT, and MAP. Cosinor analysis is a statistical technique commonly used in chronobiology to analyze rhythmic (cyclical) data. This method is particularly useful for assessing patterns in biological processes that demonstrate circadian rhythms, such as hormone levels, body temperature, and heart rate, which typically follow a 24-hour cycle. Cosinor analysis is used to fit a cosine wave to the time series, with the model specified as\u003c/p\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Y=M+A*\\text{c}\\text{o}\\text{s}\\left[2\\pi\\:\\right(T-\\varphi\\:)/24]$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003ewhere Y is the vital sign value (heart rate, body temperature, or MAP), M represents the MESOR (midline estimating statistic of rhythm), A the amplitude, T the time of day in hours, and ϕ the acrophase of the circadian rhythm (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). According to the cosinor model, the three parameters MESOR, amplitude, and acrophase characterize the circadian rhythm of heart rate (HR), body temperature (BT), and mean arterial pressure (MAP).\u003c/p\u003e\u003cp\u003eLogistic regression models were then used to examine associations between circadian parameters, obesity categories, and in-hospital mortality, adjusting for covariates. Stratified analyses were conducted to explore whether associations differed across obesity subgroups.\u003c/p\u003e\u003cp\u003eFor prediction, data were randomly split into a training set (70%) and a testing set (30%). Two sets of models were built. Baseline models include obesity status and covariates only. Proposed models include circadian parameters in addition to baseline variables. We evaluated four algorithms: logistic regression, random forest, support vector machine, and extreme gradient boosting. To address class imbalance in the training data, we applied the Synthetic Minority Oversampling Technique. Model performance was assessed in the testing set using the area under the receiver operating characteristic curve (AUC), accuracy, precision, and recall. Ten-fold stratified cross-validation was used to reduce overfitting and ensure model robustness.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eFrom 73 181 ICU patients in MIMIC-IV, 40 074 adults had at least one ICU admission lasting 24 hours. After applying exclusion criteria for incomplete or implausible vital sign records, 14,064 patients remained in the analytic cohort (Fig.\u0026nbsp;3). Of these, 1 938 (13.8%) died before hospital discharge.\u003c/p\u003e\u003cp\u003eBaseline characteristics are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Among the 12 126 patients who survived, the average age was 63.05 years (SD 17.86), and about 45.5% were women. For the 1 938 who died, the average age was higher at 70.40 (SD 14.59), with 47.8% women. Patients who died also had a higher Charlson Comorbidity Index score (4.28 on average versus 2.56 for survivors). The distribution of obesity categories was similar across groups, although a higher proportion of those who died had class III obesity (3.5% vs. 4.7%).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCharacteristics summary\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eN, (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSurvived\u003c/p\u003e\u003cp\u003e(N\u0026thinsp;=\u0026thinsp;12 126)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDied In Hospital\u003c/p\u003e\u003cp\u003e(N\u0026thinsp;=\u0026thinsp;1 938)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSex\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5 522 (45.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e926 (47.8%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6 604 (54.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1 012 (52.2%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge groups\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e18\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e682 (5.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27 (1.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e30\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e686 (5.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e33 (1.7%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e40\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1 275 (10.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e115 (5.9%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e50\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 146 (17.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e253 (13.1%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e60\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 527 (20.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e415 (21.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e70\u0026thinsp;\u0026le;\u0026thinsp;Age\u0026thinsp;\u0026lt;\u0026thinsp;79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 251 (18.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e472 (24.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u0026thinsp;\u0026ge;\u0026thinsp;80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 559 (21.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e623 (32.1)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRace\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNon-Hispanic White\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8 479 (69.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1 315 (67.9%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNon-Hispanic Black\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1 107 (9.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e179 (9.2%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHispanic\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e469 (3.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e53 (2.7%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAsian\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e322 (2.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e78 (4.0%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther Race\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1 749 (14.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e313 (16.2%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCCI, Mean (SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.56 (2.37)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.28 (2,65)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity class\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal weight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4 171 (34.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e690 (35.6%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverweight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4 317 (35.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e733 (37.8%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity I\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 280 (18.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e339 (17.5%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity II\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e788 (6.5%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e109 (5.6%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity III\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e570 (4.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e68 (3.5%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003eCCI, Charlson Comorbidity Index; SD, standard deviation.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eMultivariable logistic regression identified several predictors of mortality (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Having class III obesity was strongly associated with in-hospital death, with an odds ratio (OR) of 1.626 (95%CI 1.404\u0026ndash;1.883, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Circadian parameters of vital signs also carried prognostic value. A higher MESOR of heart rate was associated with greater mortality risk (OR 1.027, 95% CI 1.024\u0026ndash;1.030, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and larger amplitude of body temperature predicted higher odds of death (OR 1.092, 95% CI 1.059\u0026ndash;1.127, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Conversely, higher MESOR of mean arterial pressure was protective (OR 0.998, 95% CI 0.998\u0026ndash;0.999, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Other circadian parameters, including acrophase of heart rate, body temperature, and MAP, were not significant in the overall cohort.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eLogistic regression results\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKey independent variables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOdds ratio\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity class\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal weight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e\u003cp\u003eRef.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverweight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.911\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.092\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.747, 1.111)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.356\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity I\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.835\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.106\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.652, 1.070)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.155\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity II\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.809\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.154\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.556, 1.175)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.265\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity III\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.626\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.122\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(1.404, 1.883)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeart rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMesor\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.027\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(1.024, 1.030)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.004\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.003\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.998, 1.011)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.196\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.993\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.96,4 1.022)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.628\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBody Temperature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMesor\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.017\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.988, 1.047)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.261\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.092\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.017\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(1.059, 1.127)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.003\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.972, 1.034)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.871\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMean arterial pressure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMesor\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.998\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.998, 0.999)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(1.001, 1.002)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.980, 1.038)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.547\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eSE, standard error; CI: confidence interval.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eStratified analyses by obesity category showed heterogeneity in these associations (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Heart rate MESOR consistently predicted higher mortality in all groups, with odds ratios ranging from 1.028 among patients with class II obesity (95%CI 1.022\u0026ndash;1.034, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) to 1.040 among patients with class III obesity (95% CI 1.022\u0026ndash;1.059, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Body temperature amplitude was significantly associated with mortality among patients with overweight (OR\u0026thinsp;=\u0026thinsp;1.146, 95% CI 1.076\u0026ndash;1.220, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and among patients with class I obesity (OR 1.108, 95% CI 1.009\u0026ndash;1.217, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), but not among patients with normal weight or higher obesity categories. MAP MESOR was inversely associated with mortality among patients with normal weight (OR\u0026thinsp;=\u0026thinsp;0.999, 95% CI 0.997\u0026ndash;1.001, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), among patients with overweight (OR\u0026thinsp;=\u0026thinsp;0.999, 95% CI 0.997\u0026ndash;1.001, p\u0026thinsp;\u0026lt;\u0026thinsp;0.1), and among patients with class I obesity (OR\u0026thinsp;=\u0026thinsp;0.998, 95%CI 0.996-1.000, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Among patients with class III obesity, both heart rate MESOR (OR\u0026thinsp;=\u0026thinsp;1.040, 95% CI 1.022\u0026ndash;1.059, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and body temperature acrophase (OR\u0026thinsp;=\u0026thinsp;1.219, 95% CI 0.979\u0026ndash;1.518, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were significantly associated with mortality, suggesting that both the magnitude and timing of circadian variation become clinically relevant among patients with severe obesity.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eStratified analysis by obesity status\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e\u003cp\u003eOR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e\u003cp\u003eObesity status\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eNormal weight\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eOverweight\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eObesity I\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eObesity II\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eObesity III\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeart rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMESOR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.028 (1.022, 1.034)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.026 (1.020, 1.032)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.026 (1.018, 1.034)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.021 (1.007, 1.035)***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.040 (1.022, 1.059)****\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.008 (0.996, 1.020)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.999 (0.989, 1.009)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.998 (0.982, 1.014)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.027 (0.997, 1.058).\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.999 (0.962, 1.037)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.009 (0.961, 1.060)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.964 (0.922, 1.008)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.044 (0.969, 1.125)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.963 (0.851, 1.090)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.911 (0.783, 1.059)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u003cp\u003eBody temperature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMESOR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.969 (0.926, 1.014)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.016 (0.966, 1.069)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.054 (0.973, 1.142)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.008 (0.857, 1.186)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.992 (0.830, 1.186)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.126 (1.066, 1.190)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.146 (1.076, 1.220)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.108 (1.009, 1.217)**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.926 (0.789, 1.087)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.073 (0.860, 1.339)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.006 (0.956, 1.059)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.972 (0.926, 1.021)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.000 (0.930, 1.075)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.054 (0.919, 1.209)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.219 (0.979, 1.518)**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u003cp\u003eMean arterial pressure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMESOR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.999 (0.997, 1.001)***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.999 (0.997, 1.001)*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.998 (0.996, 1.000)*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.997 (0.991, 1.003)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.999 (0.993, 1.005)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmplitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.001 (1.001, 1.001)**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.002 (1.000, 1.004)****\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.001 (0.999, 1.003)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.000 (0.994, 1.006)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.002 (0.996, 1.008)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAcrophase\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.999 (0.953, 1.047)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.023 (0.974, 1.074)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.969 (0.907, 1.036)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.026 (0.905, 1.163)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.099 (0.919, 1.314)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e****, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001; ***, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01; **, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05; *, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.1.\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003eCI, confidence interval.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe training data includes 70% (9 844) of patients, and 30% (4 220) of patients are in testing data. Prediction model performance results are presented in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Incorporating circadian parameters consistently improved predictive accuracy compared with baseline models. In the proposed models, Random Forest achieved the best performance, with an area under the receiver operating characteristic curve (AUC) of 0.764, an accuracy of 85.8%, a recall of 88.2%, and a precision of 87.1%. Logistic regression also improved substantially with the addition of circadian variables (AUC 0.754 compared with 0.701 in the baseline model). XGBoost achieved an AUC of 0.753 with strong recall and precision, while support vector machine models showed modest but consistent improvements over the baseline. Together, these results demonstrate that circadian rhythm parameters enhance the prediction of ICU mortality, particularly when considered alongside obesity status.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePrediction model performances\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAccuracy\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBaseline model\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLogistic regression\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.701\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.682\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.680\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.817\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRandom forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.711\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.793\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.801\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.778\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExtreme gradient boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.707\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.753\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.775\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.790\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSupport vector machine\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.702\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.710\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.731\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.801\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eProposed model\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLogistic regression\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.754\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.684\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.681\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.835\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRandom forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.764\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.858\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.882\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.871\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExtreme gradient boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.753\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.765\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.807\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.810\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSupport vector machine\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.727\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.721\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.746\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.815\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eAUC: area under the ROC curve.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this large cohort of ICU patients, we found that circadian rhythm parameters of vital signs, together with obesity status, were independently associated with in-hospital mortality. Heart rate MESOR emerged as a consistent predictor across all BMI categories, while body temperature amplitude and mean arterial pressure (MAP) MESOR were significant in selected subgroups. Notably, among patients with class III obesity, both heart rate MESOR and body temperature acrophase were strong predictors, underscoring that the timing as well as the magnitude of circadian variation may play a clinically meaningful role among patients with severe obesity. Importantly, incorporating circadian parameters into machine learning models improved predictive performance compared with conventional models based only on demographics, comorbidities, and obesity status.\u003c/p\u003e\u003cp\u003eOur findings extend previous work on circadian disruption in the ICU. Prior studies have demonstrated that circadian rhythmicity persists in vital signs during critical illness, and that attenuated or phase-shifted rhythms are associated with adverse outcomes.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e However, most prior analyses were limited to small cohorts or single parameters such as body temperature. By applying cosinor analysis to multiple vital signs in a large ICU population, our study provides robust evidence that circadian metrics carry prognostic value beyond conventional clinical variables.\u003c/p\u003e\u003cp\u003eThe modifying role of obesity is also noteworthy. Experimental and epidemiologic studies have established that obesity alters circadian biology through changes in clock gene expression, hormonal oscillations, and metabolic regulation.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Our stratified results suggest that these disturbances amplify the prognostic importance of circadian rhythms in critically ill patients. For example, body temperature amplitude predicted mortality among patients with overweight and among patients with class I obesity. However, this relationship weakened among patients with higher obesity levels, where instead the timing of rhythms (acrophase) became more relevant. This pattern may reflect impaired thermoregulatory capacity in severe obesity, leading to reliance on timing shifts rather than amplitude changes as markers of physiological instability.\u003c/p\u003e\u003cp\u003eFrom a clinical standpoint, these findings highlight the potential of circadian parameters as simple, low-cost indicators for risk stratification in the ICU. Vital signs are continuously collected in most modern ICUs, and extracting circadian rhythm measures from these data could provide a dynamic, noninvasive means of assessing physiologic reserve and vulnerability. The consistent predictive improvement observed across multiple machine learning models supports the feasibility of integrating circadian metrics into real-time decision support systems. Such tools could help clinicians identify high-risk patients earlier, tailor interventions more precisely, and design ICU environments that minimize circadian disruption.\u003c/p\u003e\u003cp\u003eHowever, some limitations should be acknowledged. First, the data were sourced from a single academic center, which may limit generalizability to other patient populations or care practices. Second, while cosinor analysis provides a robust framework for detecting circadian rhythmicity, ICU vital signs are influenced by multiple external factors, including sedation, mechanical ventilation, and therapeutic interventions, which may confound the rhythms observed. Third, despite adjusting for major covariates, residual confounding from unmeasured variables, such as medication timing, staff shift changes, or sleep disruption, cannot be excluded. Finally, the observational design precludes causal inference.\u003c/p\u003e\u003cp\u003eWe will extend our findings further to elucidate the role of circadian rhythms in critical care. Planned subgroup analyses will stratify patients by age groups (e.g., \u0026lt;\u0026thinsp;65 years, 65\u0026ndash;75 years, and \u0026gt;\u0026thinsp;75 years) to explore whether the associations between circadian parameters and mortality vary across the lifespan, potentially revealing age-specific vulnerabilities in circadian disruption among ICU patients with obesity. Additionally, incorporating established severity scores, such as the Sequential Organ Failure Assessment (SOFA) or the Acute Physiology and Chronic Health Evaluation II (APACHE II), into our predictive models will enable more comprehensive risk adjustment. This should provide a clearer understanding of how circadian parameters contribute independently, beyond the patient's initial level of illness upon entering the ICU.\u003c/p\u003e\u003cp\u003eIn summary, circadian rhythm parameters of vital signs were independently associated with in-hospital mortality, and their effects varied across obesity categories. Heart rate MESOR was a consistently significant predictor, while body temperature amplitude and MAP MESOR showed differential associations by BMI. Among patients with class III obesity, both the magnitude and timing of circadian variation emerged as important risk markers. Incorporating circadian metrics into machine learning models substantially improved predictive performance compared with conventional approaches.\u003c/p\u003e\u003cp\u003eThese findings underscore the value of circadian analysis in critical care and highlight the interaction between chronobiology and obesity in shaping outcomes. Routine ICU vital sign data can provide actionable circadian parameters, offering a low-cost, scalable approach to enhance risk stratification and guide individualized care. Future multicenter studies and interventional trials are needed to validate these observations and to determine whether preserving circadian alignment can improve outcomes in critically ill patients.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests\u003c/h2\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\u003cp\u003eGS was responsible for designing the study protocol, conducting data extraction and analysis from the MIMIC-IV database, performing analyses, interpreting results, and drafting the manuscript. LJC provided expertise on obesity and circadian rhythms and reviewed clinical interpretations. HM assisted with statistical modeling, including machine learning algorithms, and revised the manuscript for intellectual content. HX oversaw the overall project, contributed to methodology development, supervised data analysis, and interpreted findings. All authors approved the final version and agree to be accountable for all aspects of the work.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eThis study utilized publicly available data from the MIMIC-IV database and received no external funding or grants. The authors are solely responsible for the content.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe raw de-identified ICU data were obtained from MIMIC-IV (version 2.2) and are freely available to qualified researchers after completing the required training and data use agreement at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://physionet.org/content/mimiciv/2.2/\u003c/span\u003e\u003cspan address=\"https://physionet.org/content/mimiciv/2.2/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. All relevant raw data supporting the findings of this study are freely accessible for replication purposes without breaching ethical standards.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eRefinetti R. The circadian rhythm of body temperature. Front Biosci (Landmark Ed). 2010;15(2):564\u0026ndash;94.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGazendam JAC, Van Dongen HPA, Grant DA, Freedman NS, Zwaveling JH, Schwab RJ. Altered circadian rhythmicity in patients in the ICU. Chest. 2013;144(2):483\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLogan RW, Sarkar DK. Circadian nature of immune function. Molecular and Cellular Endocrinology. 2012;349(1):82\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMorrison SF, Nakamura K. Central mechanisms for thermoregulation. Annual Review of Physiology. 2019;81(Volume 81, 2019):285\u0026ndash;308.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcKenna H, van der Horst GTJ, Reiss I, Martin D. Clinical chronobiology: A timely consideration in critical care medicine. Crit Care. 2018;22(1):124.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKorompeli A, Muurlink O, Kavrochorianou N, Katsoulas T, Fildissis G, Baltopoulos G. Circadian disruption of ICU patients: A review of pathways, expression, and interventions. J Crit Care. 2017;38:269\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEgi M, Bellomo R, Stachowski E, French CJ, Hart G, Stow P. Circadian rhythm of blood glucose values in critically ill patients. Crit Care Med. 2007;35(2):416\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTelias I, Wilcox ME. Sleep and circadian rhythm in critical illness. Critical Care. 2019;23(1):82.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGao Y, Wang Q, Li J, et al. Impact of Mean Arterial Pressure Fluctuation on Mortality in Critically Ill Patients. Crit Care Med. 2018;46(12):e1167-e1174.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChaput JP, McHill AW, Cox RC, Broussard JL, Dutil C, da Costa BGG, et al. The role of insufficient sleep and circadian misalignment in obesity. Nature Reviews Endocrinology. 2023;19(2):82\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSumma KC, Turek FW. Chronobiology and obesity: Interactions between circadian rhythms and energy regulation. Adv Nutr. 2014;5(3):312S-9S.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFatima N, Rana S. Metabolic implications of circadian disruption. Pflugers Arch. 2020;472(5):513\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGaraulet M, Ordov\u0026aacute;s JM, Madrid JA. The chronobiology, etiology and pathophysiology of obesity. Int J Obes (Lond). 2010;34(12):1667\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHalder SK, Melkani GC. The Interplay of Genetic Predisposition, Circadian Misalignment, and Metabolic Regulation in Obesity. Curr Obes Rep. 2025;14(1):21.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBishehsari F, Voigt RM, Keshavarzian A. Circadian rhythms and the gut microbiota: From the metabolic syndrome to cancer. Nat Rev Endocrinol. 2020;16(12):731\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoer-Martins L, Figueiredo VN, Demacq C, Martins LC, Consolin-Colombo F, Figueiredo MJ, et al. Relationship of autonomic imbalance and circadian disruption with obesity and type 2 diabetes in resistant hypertensive patients. Cardiovasc Diabetol. 2011;10:24.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDavidson S, Villarroel M, Harford M, Finnegan E, Jorge J, Young D, et al. Day-to-day progression of vital-sign circadian rhythms in the intensive care unit. Crit Care. 2021;25(1):156.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang Z, Xie X, Zhang X, Li L, Bai R, Long H, et al. Circadian rhythms of vital signs are associated with in-hospital mortality in critically ill patients: A retrospective observational study. Chronobiol Int. 2023;40(3):262\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePatke A, Young MW, Axelrod S. Molecular mechanisms and physiological importance of circadian rhythms. Nat Rev Mol Cell Biol. 2020;21(2):67\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRobinson I, Reddy AB. Molecular mechanisms of the circadian clockwork in mammals. FEBS Lett. 2014;588(15):2477\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJohnson A, Bulgarelli L, Pollard T, Horng S, Celi LA, Mark R. MIMIC-IV [Internet]. PhysioNet; 2023. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://physionet.org/content/mimiciv/2.2/\u003c/span\u003e\u003cspan address=\"https://physionet.org/content/mimiciv/2.2/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-obesity","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"ijo","sideBox":"Learn more about [International Journal of Obesity](http://www.nature.com/ijo/)","snPcode":"41366","submissionUrl":"https://mts-ijo.nature.com/cgi-bin/main.plex","title":"International Journal of Obesity","twitterHandle":"@intjobesity","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7934062/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7934062/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjectives\u003c/h2\u003e\u003cp\u003eCircadian rhythms are frequently disrupted in ICU patients, and obesity may exacerbate this misalignment, worsening outcomes. This study evaluates associations between circadian rhythm parameters of vital signs and in-hospital mortality, assesses whether obesity modifies these relationships, and determines if integrating circadian parameters with obesity status improves predictive model performance.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eRetrospective observational cohort study using MIMIC-IV database (2008\u0026ndash;2019). Included 14 064 adult patients with \u0026ge;\u0026thinsp;24-hour ICU stays after exclusions for short stays (n\u0026thinsp;=\u0026thinsp;33 107), incomplete/implausible vital signs (n\u0026thinsp;=\u0026thinsp;26 010), DNR/DNI orders, or multiple admissions (latest kept). Circadian parameters (MESOR, amplitude, acrophase) derived for heart rate (HR), body temperature (BT), and mean arterial pressure (MAP) using cosinor analysis from first 24-hour vital signs. Associations assessed using logistic regression, overall and BMI-stratified (normal\u0026thinsp;\u0026lt;\u0026thinsp;25, overweight 25\u0026ndash;29.9, class I 30\u0026ndash;34.9, class II 35\u0026ndash;39.9, class III\u0026thinsp;\u0026ge;\u0026thinsp;40 kg/m\u003csup\u003e2\u003c/sup\u003e). Machine learning models (logistic regression, random forest, SVM, XGBoost) trained on 70/30 split and 10-fold cross-validation.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eOf 14 064 patients, 1 938 (13.8%) died in-hospital. HR MESOR (OR 1.027, 95% CI 1.024\u0026ndash;1.030, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), BT amplitude (OR 1.092, 95% CI 1.059\u0026ndash;1.127, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), MAP amplitude (OR 1.001, 95% CI 1.001\u0026ndash;1.002, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), class III obesity (OR 1.626, 95% CI 1.404\u0026ndash;1.883, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) increased mortality risk; MAP MESOR protective (OR 0.998, 95% CI 0.998\u0026ndash;0.999, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). HR MESOR was significant across BMI groups (ORs 1.021\u0026ndash;1.040, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01 to \u0026lt;\u0026thinsp;0.001); BT acrophase in class III (OR 1.219, 95% CI 1.041\u0026ndash;1.428, p\u0026thinsp;=\u0026thinsp;0.014). The proposed random forest model achieved the best predictive performance (AUC 0.764; accuracy 85.8%; recall 88.2%; precision 87.1%), outperforming the baseline models.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eCircadian rhythm parameters were significantly associated with in-hospital mortality in ICU patients, with effects modified by obesity status, and their inclusion significantly improved predictive performance.\u003c/p\u003e","manuscriptTitle":"Circadian Rhythm Disruption and In-Hospital Mortality in ICU Patients with Obesity","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-10 05:17:33","doi":"10.21203/rs.3.rs-7934062/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2025-11-26T11:44:43+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-11-20T18:08:32+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-11-06T01:27:53+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-10-30T18:49:11+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-10-29T20:56:47+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2025-10-27T20:50:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-27T13:32:56+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Obesity","date":"2025-10-24T14:45:37+00:00","index":"","fulltext":""},{"type":"checksFailed","content":"","date":"2025-10-24T13:39:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-23T16:26:49+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-obesity","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"ijo","sideBox":"Learn more about [International Journal of Obesity](http://www.nature.com/ijo/)","snPcode":"41366","submissionUrl":"https://mts-ijo.nature.com/cgi-bin/main.plex","title":"International Journal of Obesity","twitterHandle":"@intjobesity","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"620a0706-5b1a-4a82-8276-801150047a98","owner":[],"postedDate":"November 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[{"id":56959477,"name":"Health sciences/Risk factors"},{"id":56959478,"name":"Health sciences/Diseases/Endocrine system and metabolic diseases/Obesity"}],"tags":[],"updatedAt":"2025-11-26T11:47:40+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-10 05:17:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7934062","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7934062","identity":"rs-7934062","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}