Health Checkup Data Research Articles

Mortality risk and years of life lost for people with reduced renal function detected from regular health checkup: A matched cohort study

With the increasing threat of metabolic syndromes, a focus on maintaining kidney health from early- to mid-adulthood is necessary. This study elucidates mortality risk and years of life lost (YLLs) due to abnormal renal function. This was a retrospective, matched cohort study from health checkup data from 2000 to 2015. We identified 12,774 participants with abnormal renal function (eGFR < 60 mL/min/1.73 m2) and used propensity score matching to identify 25,548 participants with normal renal function (eGFR ≥ 60). YLLs were estimated using the life expectancy differences between the abnormal and matched normal cohorts. Cox models were used to estimate the adjusted mortality risk. The estimated life expectancy of participants with proteinuria and eGFR < 60 was 26.24 years, with a 95 % confidence interval of (23.96, 29.36), 17.62 (16.37, 18.78), and 11.70 (11.02, 12.46) for age groups of 30 – 54, 55 – 64, and 65 – 79 years, respectively. The estimated YLLs of participants with proteinuria and eGFR < 60, as compared with the matched normal cohort, were 17.86 (13.41, 20.36), 12.55 (11.41, 13.78), and 8.31 (7.47, 9.13) years for the three age groups, respectively. The Cox model estimates of mortality hazard ratios of participants having proteinuria and eGFR < 60 against matched referents were 5.29 (3.97, 7.05), 3.99 (3.34, 4.75), and 3.05 (2.62, 3.55) for the three age groups, respectively. Abnormal renal function shortens life expectancy, particularly in patients with proteinuria and in younger adults. Active health management of renal function can reduce the disease burden.

PS-C27-7: ELECTROCARDIOGRAPHIC LEFT VENTRICULAR HYPERTROPHY CAN PREDICT FUTURE DECLINE OF RENAL FUNCTION IN GENERAL POPULATION

Objective: Electrocardiographic left ventricular hypertrophy (LVH) has been shown to be associated with future renal function decline in patients with chronic kidney disease. However, this association remains unclear in general population. We aimed to elucidate the association between baseline LVH on electrocardiogram and future renal function decline using health checkup data. Design and method: We used a dataset derived from the annual health-screening program from 2010 to 2019. Patients with baseline estimated glomerular filtration rate (eGFR) &lt; 60 ml/min/1.73m2 were excluded. LVH was determined by the electrocardiographic voltage criteria at baseline. Association between LVH and renal function outcomes, specifically ≧ 30% decline in eGFR from baseline, eGFR decline less than 60 ml/min/1.73m2 and 45 ml/min/1.73m2, were analyzed using multivariate Cox proportional hazard models. Results: Among total of 19,825 participants, 1,263 had LVH and 18,562 no LVH. LVH group was older and more male and current smokers, and to complicated with more hypertension, and higher systolic (136.7 ± 20.9 vs 123.7 ± 19.1 mmHg, P &lt; 0.001) and diastolic blood pressure (81.8 ± 12.6 vs 74.6 ± 12.0 mmHg, P &lt; 0.001). Baseline eGFR and HbA1c were comparable between the groups. Incidence of ≧ 30% eGFR decline was significantly higher in LVH group (P &lt; 0.001, log-rank test). LVH was associated with ≧ 30% eGFR decline (HR 1.92, 95% CI 1.26–2.91, P = 0.002) and eGFR &lt; 45 ml/min/1.73m2 (HR 3.12, 95% CI 1.54–6.35, P = 0.002) after adjustment. LVH was not associated with eGFR &lt; 60 ml/min/1.73m2 in the overall population. However, when analyzed this association in subpopulation stratified by baseline SBP (&lt; 140 and ≧ 140 mmHg), LVH was associated with eGFR &lt; 60 ml/min/1.73m2 in persons with SBP ≧ 140 mmHg (HR 1.40, 95% CI 1.10–1.79, P = 0.007) but not in those with SBP &lt; 140 mmHg (HR 0.88, 95% CI 0.68–1.14, P = 0.32) (P for interaction between LVH and baseline SBP = 0.005). Similarly, LVH was significantly associated with ≧ 30% eGFR decline and eGFR &lt; 45 ml/min/1.73m2 in persons with SBP ≧ 140 mmHg but not in those with SBP &lt; 140 mmHg, although the interaction between LVH and baseline SBP was not significant in these renal outcomes. Conclusion: Electrocardiographic LVH was independently associated with the rate of renal function decline among general population. The association between LVH and future renal function decline might be stronger in people with SBP ≧140 mmHg.

S-05-1: ONSET OF LIFESTYLE-RELATED DISEASES AFTER THE GREAT EAST JAPAN EARTHQUAKE AMONG RESIDENTS IN FUKUSHIMA

Due to the accident at the Fukushima Nuclear Power Plant (NPP) following the Great East Japan Earthquake on March 11, 2021, residents of the coastal areas of Fukushima Prefecture were forced to evacuate due to the effects of radiation and their lifestyles have changed dramatically. Therefore, we conducted the Comprehensive Health Check (CHC) of the Fukushima Health Management Survey (FHMS) to assess the health status of residents in the evacuation area and to prevent lifestyle-related diseases. First, to examine the short- to mid-term effects of the NPP accident, pre- and post-disaster data on health checkup of the evacuation area residents were retrospectively analyzed. Next, to examine the mid- to long-term effects of the NPP accident, we prospectively examined health examination data during the post-disaster period (fiscal year (FY) 2011–2017). The subjects were men and women living in 13 municipalities in the area surrounding the NPP in Fukushima Prefecture. We analyzed 27,486 men and women (12,432 men and 15,054 women) aged 20 years or older for whom health checkup data were available before and after the disaster, and found that the proportion of obese people and the prevalence of hypertension, dyslipidemia, diabetes mellitus, metabolic syndrome, liver dysfunction, hyperuricemia, polycythemia, and atrial fibrillation were increased after the disaster (FY2011–2012), compared with those before the disaster (FY 2008–2010). This trend was more pronounced among residents who were forced to evacuate. The analysis of 27,536 people (12,254 men and 15,282 women) who could be followed for an average of 5.5 years from FY 2011 to FY 2017 showed that the proportion of obese people remained unchanged, while the proportion of people with liver dysfunction decreased. The prevalence of hypertension, dyslipidemia, and diabetes increased further. Furthermore, evacuees were shown to be at higher risk of developing diabetes, dyslipidemia, chronic kidney disease, hypertension, and hepatic dysfunction than non-evacuees. On the other hand, regular exercise was a significant preventing factor for liver dysfunction and dyslipidemia. Since residents in the evacuation areas, especially evacuees, continue to be at high risk of developing lifestyle-related diseases, especially cardiovascular diseases, it is necessary to management their health status over a long period of time and implement preventive measures against lifestyle-related diseases.

PS-BPP03-1: PREDICTIVE ABILITY FOR NEW-ONSET CHRONIC KIDNEY DISEASES BY MACHINE LEARNING ALGORITHM USING HEALTH CHECKUP DATA

Background: In Japan, many data on relatively healthy individuals exist because specific health checkups are conducted annually for individuals aged 40 and older for the purpose of early detection and intervention of lifestyle-related diseases. Efforts to effectively utilize these data to predict the future development of chronic kidney disease (CKD) are important from the perspective of improving prognosis and quality of life, as well as healthcare economics. Aim: We aimed to evaluate the predictive ability of machine learning algorithms for newly developed CKD using health checkup data. Methods: We included individuals aged 30–69 years who had a health checkups between April 2008 and March 2018 with baseline (year X) and 5-year (year X + 5) data. CKD was defined as &lt; 60 ml/min/1.73m2, and those with hypertension in year X were excluded. Three different machine learning algorithms, including Random Forest, Logistic Regression, and Neural Network, were evaluated to predict the development of new-onset CKD at year X + 5 from baseline (year X) data, and the AUCs of the ROC curve were used to assess the predictive ability of the models. Explanatory variables included physical measurements, blood pressure and pulse rate, laboratory values, lifestyle-related indicators, and medication history. Results: A total of 58,418 participants (age 53.8 ± 10.2 years, male 49.9%) were included in this study. Among them, 3,997 individuals (6.8%) had newly developed hypertension in year X + 5. The indices of predictive ability were as follows: Random Forrest, AUC 0.88, accuracy 0.93, recall 0.99, precision 0.93; Logistic Regression, AUC 0.89, accuracy 0.93, recall 0.99, precision 0.94; Neural Network, AUC 0.89, accuracy 0.93, recall 0.99, precision 0.93. The variables with the highest importance features were baseline creatinine level (0.49), age (0.24), systolic blood pressure (0.07), and anti-hypertensive medication (0.06). Conclusions: The AUCs of the model predicting the development of CKD after 5 years ranged from 0.88 to 0.89, which were generally superior to previously reported risk models using statistical methods. Baseline creatinine levels, age, and blood pressure were found to contribute significantly to new-onset CKD.

S-47-1: THE POTENTIAL OF AI IN HYPERTENSION TREATMENT

Deep learning, which has been the driving force behind AI in recent years, has been shown in various research reports to be as powerful as or more powerful than humans in the medical field. Specifically, deep learning techniques are being applied to images, time series data, biomolecular information, and natural language for medical images, clinical values, genomic information, and medical records in electronic medical records, respectively. In particular, AI for recognizing and diagnosing medical images has been shown to have high accuracy in various diseases including cancer, and it is expected that various types of AI for image diagnosis will be put to practical use in the medical field within the next five years. In addition to highly accurate diagnostic imaging, the benefits of AI for medicine include genomic diagnosis, decision support to reduce the burden on medical professionals, correction of regional disparities in medicine through AI, and self-medication by patients themselves using AI. Furthermore, AI is expected to make it possible to do things that are difficult to do with human knowledge and experience, such as predicting the future of a patient's medical condition, suggesting the best treatment for each individual, and quickly and appropriately extracting medical significant information from a large amount of medical knowledge around the world. As an application of AI and data science to the healthcare domain, we have worked on big data analysis of the Iwaki health checkup cohort in Hirosaki City, Japan and successfully developed AI to predict the early onset of 11 diseases, including hypertension, from the health checkup data. In addition to predicting disease onset, we have also developed an AI that estimates intervention points and proposes optimal plans to improve health and prevent disease onset. In this presentation, I would like to introduce the potential of big data and AI in hypertension by illustrating these our research.

PS-BPP05-2: TREATMENT STATUS OF HYPERTENSION DETECTED IN HEALTH CHECKUPS: AN OBSERVATIONAL STUDY OF JAPANESE EMPLOYEES

Objective: Despite the clear evidence for lowering blood pressure in hypertensive patients, control rates are insufficient, especially among younger adults. The first step in improving control rates is patient behavior to visit an outpatient when hypertension is detected at annual health checkups. The present study aims to explore the demographic and lifestyle characteristics related to the outpatient visits among employees who had untreated grade II or grade III hypertension. Design and method: An observational study was conducted using the data of annual health checkups and medical claims from 2012 to 2020 in an employee's health insurance society in Japan. We included participants if they were 20–69 years of age and had untreated grade II or grade III hypertension (blood pressure (BP)&gt; = 160/100 mmHg and no use of BP-lowering medications) at the time of health checkups (first measurement) and had BP &gt; = 140/90 mmHg in the previous years, to exclude transient hypertension. If more than one year of data met the criteria, the data from the first year was used to remove duplicates of individuals. Multilevel logistic regression with random effects of health checkup years was used to estimate odds ratios (ORs) with 95% confidence intervals (95%CIs) of demographics and lifestyle characteristics for newly outpatient visits (completing a clinic visit within 6 months). Results: The number of employees who had annual health checkups was approximately 10,000, and the number of applicable individuals for the present study was approximately 6% each year. In 2013–2020, after duplicate removal, 993 individuals (mean age 51.5 years, 63% women) had untreated grade II or grade III hypertension. Of these, 200 (20%) had an outpatient visit within 6 months. Among the independent variables, diabetes (ORs and 95% CI: 1.97, 1.19–3.26), body mass index (per 1-BMI increment: 1.04, 1.00–1.08), age (per 1-year increment: 1.03, 1.00–1.05), and systolic BP (per 1- mmHg increment: 1.04, 1.03–1.06) had significantly positive ORs for outpatient visits. The results of a similar analysis for antihypertensive medication after outpatient visits, no other factors besides systolic BP were associated with medication. Conclusion: Only 20% of individuals who detected grade II or grade III hypertension in annual health checkups visited outpatients within 6 months. The present findings suggested the necessity for more aggressive promotion of outpatient visits to younger adults without chronic illnesses or obesity.

PS-P10-2: NON-LINEAR RELATIONSHIP BETWEEN CKD PROGRESSION AND SYSTOLIC AND DIASTOLIC BLOOD PRESSURE; IKI EPIDEMIOLOGICAL STUDY OF ATHEROSCLEROSIS AND CKD (ISSA-CKD)

Objectives: The aim of the study is to assess the non-linear relationship between systolic and diastolic blood pressure and chronic kidney disease (CKD) progression. Design and Method: This is a cohort study using annual health check-up data from 2008 to 2018 in Iki City, Nagasaki, Japan. After excluding those who had health check-ups only once or those with the missing value for serum creatinine or urinalysis, a total of 5919 participants were included in the present analysis. The outcome of the present analysis was CKD progression, which was defined as worsening either the glomerular filtration rate stage or albumin creatinine ratio stage according to KDIGO classification at a visit and at the end of follow-up. Non-linear effects of SBP and DBP on the progression of CKD were evaluated by Cox proportional analysis with a restricted cubic spline controlling for confounding factors. Results: During a mean follow-up of 4.9 years, 1167(19.7%) CKD progression was observed. As each SBP and DBP were classified into quartiles, there was a linear relationship between SBP and CKD progression (p &lt; 0.001 for trend), on the other hand, there was no linear relationship between DBP and CKD progression (p = 0.534 for trend). For restricted cubic spline, we found a sigmoid curve-like relationship between SBP and CKD progression, in which CKD progression increased in accord with BP increase. For DBP, we found a U-like shape relationship between DBP and CKD progression, in which there was the lowest risk at nearly 70–80 mmHg, and then the risk for CKD progression increased to both lower and higher DBP. Conclusions: The findings of the present analysis showed a near-linear trend that higher SBP had a significant relationship with an increase in CKD progression. On the other hand, the relationship between DBP and CKD progression was not linear, and both lower and higher DBP could be related to CKD progression.

Collection and Standardization of Health Check-up Data for the Development of a Police Officer Health Management Platform

Collection and Standardization of Health Check-up Data for the Development of a Police Officer Health Management Platform

PS-P08-5: METABOLIC SYNDROME RISK PREDICTION EQUATION AND A SIMPLE SCORE MODEL DERIVATION AND VALIDATION IN JAPANESE GENERAL POPULATION

Objective: Predicting metabolic syndrome (MetS) is important for identifying high risk cardiovascular disease individuals and providing preventive interventions. We aimed to develop and validate an equation and a simple MetS score according to the Japanese MetS criteria using Japanese health checkup data. Design and Methods: In total, 54,198 participants (age, 54.5 ± 10.1 years; men, 46.0%), with baseline and five year follow up data were randomly assigned to Derivation and Validation cohorts (ratio: 2:1). Multivariable logistic regression analysis was performed in derivation cohort and scores were assigned to factors corresponding to beta coefficients. We evaluated predictive ability of the scores using area under the curve (AUC), then applied them to validation cohort to assess reproducibility. For equation model, all variables were treated as continuous except for sex, current smoking, and alcohol consumption, which were treated as binary variables. Results: The primary model (including blood tests) ranged 0 to 27 points had an AUC of 0.81 (sensitivity: 0.81, specificity: 0.81, cut-off score: 14), and consisted of age, sex, blood pressure (BP), body mass index (BMI), serum lipids, glucose measurements, current smoking, and alcohol consumption. The simplified model (excluding blood tests) ranged 0 to 17 points with an AUC of 0.78 (sensitivity: 0.83, specificity: 0.77, cut off score: 15) and included seven variables: age, men, systolic BP, diastolic BP, BMI, alcohol consumption, and current smoking. We classified individuals with a score &lt; 15 as low-risk MetS and ≧ 15 points as high risk MetS, respectively. The incidence of MetS increased constantly as the score increased. Furthermore, the equation model to predict MetS incidence from derivation cohort generated an AUC of 0.85 (sensitivity: 0.86, specificity: 0.55). Analysis of the validation and derivation cohorts yielded similar results. Conclusion: We developed a primary score, an equation model, and a simple score. The simple score is convenient, well validated with acceptable discrimination, and could be used for early detection of MetS in high risk individuals.

Occupational stress changes and new-onset depression among male Korean manufacturing workers.

Studies on the association between occupational stress and depression have been frequently reported. However, the cross-sectional designs of studies limited insight into causal associations. In this study, we investigated the longitudinal association between occupational stress and new-onset depression among employees in a single manufacturing plant. The annual health checkup data of employees at a manufacturing plant in Korea were collected. A total of 1,837 male employees without depression who completed a health checkup during two consecutive years were included. Occupational stress was measured using a short form of the Korea Occupational Stress Scale (KOSS-SF), and depression was assessed using a Patient Health Questionnaire-2. The association between occupational stress change over the two years and newly developed depression was investigated using two logistic regression models. Across all sub-factors of KOSS-SF, employees who reported increased occupational stress had a higher risk of new-onset depression. Newly developed depression was significantly associated with job demand (odds ratio [OR]: 4.34; 95% confidence interval [CI]: 2.37-7.96), job insecurity (OR: 3.21; 95% CI: 1.89-5.48), occupational climate (OR: 3.18; 95% CI: 1.91-5.31), lack of reward (OR: 2.28; 95% CI: 1.26-4.12), interpersonal conflict (OR: 2.14; 95% CI: 1.18-3.86), insufficient job control (OR: 1.93; 95% CI: 1.05-3.56), and the organizational system (OR: 1.84; 95% CI: 1.01-3.36). For every sub-factor of the KOSS-SF, occupational stress increase and persistent high stress were associated with the risk of developing new-onset depression. Among the seven sub-factors, job demand had the most significant effect. Our results show that occupational stress should be managed to promote employee mental healthcare.

Casual Serum Triglyceride Concentrations and New-Onset Chronic Kidney Disease in the General Japanese Population: The Iki City Epidemiological Study of Atherosclerosis and Chronic Kidney Disease Study

Introduction: Non-fasting triglyceride (TG) concentrations are useful for predicting various diseases, but most epidemiological studies investigated the association between fasting TG concentrations and chronic kidney disease (CKD). This study aimed to examine the association between casual (fasting or non-fasting) serum TG concentrations and new-onset CKD in the general Japanese population. Methods: We conducted a population-based, retrospective cohort study using annual health checkup data of residents of Iki City, Nagasaki Prefecture, Japan. Between 2008 and 2019, participants without CKD (estimated glomerular filtration rate <60 mL/min/1.73 m2 and/or proteinuria) at baseline were included. Casual serum TG concentrations were classified by sex as tertile 1 (men: <0.95 mmol/L; women: <0.86 mmol/L), tertile 2 (0.95–1.49 mmol/L; 0.86–1.25 mmol/L), and tertile 3 (≥1.50 mmol/L; ≥1.26 mmol/L). The outcome was incident CKD. Multivariable-adjusted hazard ratios and 95% confidence intervals were estimated using the Cox proportional hazards model. Results: 4,946 participants (2,236 [45%] men and 2,710 [55%] women; 3,666 [74%] fasting and 1,182 [24%] non-fasting) were included in the present analysis. During an average follow-up of 5.2 years, 934 participants (434 men and 509 women) developed CKD. In men, the incidence rate (per 1,000 person-years) of CKD increased with an elevation in TG concentrations (tertile 1: 29.4, tertile 2: 42.2, and tertile 3: 43.3). This association was significant, even after adjustment for other risk factors of age, current smoking habits, current alcohol intake, exercise habits, obesity, hypertension, diabetes mellitus, hyper-low-density-lipoprotein cholesterolemia, and use of lipid-lowering therapy (p = 0.003 for trend). In contrast, in women, TG concentrations were not associated with incident CKD (p = 0.547 for trend). Conclusion: Casual serum TG concentrations are significantly associated with new-onset CKD in Japanese men in the general population.

PS-C29-3: CHANGE OF PANCREAS VOLUME AND GLUCOSE METABOLIC INDEX WITH THE INCREASE OF WEIGHTS: ANALYSES OF COMPREHENSIVE HEALTH CHECKUP DATA

Objective: Pancreas volume measured using imaging techniques has been shown to be increased in subjects with obesity and reduced in subjects with diabetes. Beta-cell mass of pancreas also was reported to be increased with obesity in autopsy samples. While, the increase in beta-cell mass of pancreas with obesity was reported to limit in Japanese autopsy studies, suggesting that Japanese have less compensatory changes of beta-cell associated with insulin resistance. Therefore, we aimed to explore the changes of pancreas volume and glucose metabolic index including beta-cell function with the increase of weights in analyses of comprehensive health checkup data. Design and method: We collected subjects to have the following characteristics: 1) had our health checkup between April 2014 and March 2015 at aged 20–80 years; 2) had our health checkup between April 2015 and March 2017 with an increase in body mass index (BMI) of at least 1 kg/m2 from the baseline period; 3) no diagnosis of diabetes (HbA1c &lt; 6.5%, fasting plasma glucose &lt; 126 mg/dl and no medication); 4) had the adequate quality in CT images to assess the pancreas volume. The pancreas was outlined by hand in each CT image and the pancreas volume was computed by summing the product of pancreas area (cm2) of each image and the CT section thickness. Homeostasis model assessment of insulin resistance (HOMA-IR) and beta-cell function (HOMA-beta) were calculated as indices of insulin resistance and beta-cell function, respectively. Paired t-test was used to analyze changes in parameters. Results: Thirty-five subjects (29 men, mean age 59 years, mean height 167 cm, mean weight 71 kg) were included in this analysis and mean follow-up was 21 months. BMI and HOMA-IR increased significantly from the baseline (25.5 kg/m2 vs. 27.0 kg/m2, P &lt; 0.001 and 1.58 vs. 1.89, P &lt; 0.05). While, there were no significant change in HOMA-beta and blood pressure from the baseline (56.4 vs. 65.7 and 122/79 mmHg vs. 126/82 mmHg, all P &gt; 0.05). Conclusion: The change in beta-cell function with the increase of weights and insulin resistance was modest, consistent with the change of beta-cell mass with obesity in Japanese autopsy samples. We report the change of pancreas volume evaluating with CT image in addition to these results.

여성 군 근무자 건강검진 결과의 10년간 추이분석(2009∼2018년)

Purpose: The purpose of this study was to identify the yearly trend of Korean female military workers’ health behaviors and health examination results. Methods: This study was a secondary analysis of 77,834 female military personnels'' health check-up data in 2009~2018 from the National Health Insurance Service. Results: The rate of smoking, high risk drinking and physical activity increased over time. In "category - normal B", prevalence of diastolic blood pressure, systolic blood pressure, hemoglobin, and high density lipoprotein-cholesterol decreased; however, that of fasting blood sugar, total cholesterol, low density lipoprotein-cholesterol, and triglyceride showed the increasing trend. In "category - suspected disease", the prevalence of anemia decreased. The prevalence of dyslipidemia, however, significantly increased over the past ten years. In addition, obesity and metabolic syndrome rate showed the increasing trend. Conclusion: We found that there are uniqueness in analysis results in female military workers compared to even females in the same age groups. Thus, it is necessary to establish a tailored health promotion policy considering the characteristics of female military workers. It would contribute to improving their health behaviors and preventing diseases.

Association between physical activity and risk of renal function decline and mortality in community-dwelling older adults: a nationwide population-based cohort study

BackgroundPhysical activity (PA) is an important risk factor associated with health outcomes. However, the relationship between PA and kidney function decline in older adults remains unclear. We examined the influence of PA on kidney function decline and mortality in community-dwelling older adults.MethodsAdults aged ≥ 65 years with an estimated glomerular filtration rate (eGFR) > 60 mL/min/1.73 m2 who had available health checkup data from 2009 to 2010 were included. The cohort was followed annually through December 2015 for anthropometric, sociodemographic, and medical information including outcomes and biennially for laboratory information from the health checkup. We divided these patients into three groups according to self-reported PA (Inactive group: no leisure-time PA, Active group: vigorous activity for at least 80 min/week or a sum of moderate-intensity activity and walking for at least 300 min/week, Low-active group: level of PA between the definitions of the other two groups). Associations between the intensity of PA and death, cardiovascular death, and ≥ 50% eGFR decline were investigated.ResultsAmong 102,353 subjects, 32,984 (32.23%), 54,267 (53.02%), and 15,102 (14.75%) were classified into the inactive, low-active, and active groups, respectively. The active group was younger, contained a higher proportion of men, and had higher frequencies of hypertension, diabetes mellitus, drinking, and smoking than the other groups. The active group had significantly lower incidence rates of mortality, cardiovascular mortality, and kidney function decline than the other groups (all p < 0.001). The active group also showed lower all-cause (hazard ratio [HR], 0.76; 95% confidence interval [CI], 0.70–0.82) and cardiovascular mortality (HR, 0.64; 95% CI, 0.53–0.78) and protection against ≥ 50% eGFR decline (HR, 0.81; 95% CI, 0.68–0.97) compared with the inactive group in the fully adjusted Cox proportional hazards regression model.ConclusionsHigh PA was an independent modifiable lifestyle factor for reducing mortality and protecting against declines in kidney function in older adults.

Comparison of Taste and Smell Test Results Before and After COVID-19 in Yakumo Residents Health Checkup Comparison between 2019 and 2022

Since August 2007, the authors have conducted health checkups for residents of Yakumo Town, Hokkaido, over three days yearly, with approximately 600 people. Taste and smell tests were conducted on the study participants, and the results have been reported. However, in 2020 and 2021, we were could not receive a health checkup data to the influence of the new coronavirus. But, in August 2022, we were finally able to obtain the results of taste and smell tests. Therefore, in this study, we compare the taste and smell test results obtained in August 2019 (before the COVID-19 epidemic) and in August 2022 (after the COVID-19 epidemic).Taste and smell were measured using a simple test kit, and height, weight, and blood pressure were also obtained.

Association between Indoor Temperature in Winter and Serum Cholesterol: A Cross-Sectional Analysis of the Smart Wellness Housing Survey in Japan.

Issuance of the WHO Housing and health guidelines has paralleled growing interest in the housing environment. Despite accumulating evidence of an association between outdoor temperature and serum cholesterol, indoor temperature has not been well investigated. This study examined the association between indoor temperature and serum cholesterol. We collected valid health checkup data of 2004 participants (1333 households), measured the indoor temperature for 2 weeks in winter, and divided participants according to whether they lived in a warm (average bedroom temperature ≥ 18℃), slightly cold (12-18℃) or cold house (＜12˚C). The relationship between bedroom temperature and serum cholesterol was analyzed using multivariate logistic regression models, adjusting for demographics, lifestyle habits and the season in which the health checkup was conducted, with a random effect of climate areas in Japan. The sample sizes for warm, slightly cold, and cold houses were 206, 940, and 858, respectively. Compared to those in warm houses, the odds ratio of total cholesterol exceeding 220 mg/dL was 1.83 (95%CI: 1.23-2.71, p=0.003) for participants in slightly cold houses and 1.87 (95%CI: 1.25-2.80, p=0.002) in cold houses. Similarly, the odds ratio of LDL/non-HDL cholesterol exceeding the standard range was 1.49 (p=0.056)/1.67 (p=0.035) for those in slightly cold houses and 1.64 (p=0.020)/1.77 (p=0.021) in cold houses. HDL cholesterol and triglycerides were not significantly associated with bedroom temperature. Besides lifestyle modification, improving indoor thermal environment through strategies such as installing high thermal insulation and appropriate use of heating devices may contribute to better serum cholesterol condition.

Artificial Intelligence-based Prediction of Diabetes and Prediabetes Using Health Checkup Data in Korea

ABSTRACT The economic burden of Type 2 Diabetes (T2D) on society has increased over time. Early prediction of diabetes and prediabetes can reduce treatment cost and improve intervention. The development of (pre)diabetes is associated with various health conditions that can be monitored by routine health checkups. This study aimed to develop amachine learning-based model for predicting (pre)diabetes. Our frameworks were based on 22,722 patient samples collected from 2013 to 2020 in ageneral hospital in Korea. The disease progression was divided into three categories based on fasting blood glucose: normal, prediabetes, and T2D. The risk factors at each stage were identified and compared. Based on the area under the curve, the support vector machine appeared to have optimal performance. At the normal and prediabetes stages, fasting blood glucose and HbA1c are prevalent risk features for the suggested models. Interestingly, HbA1c had the highest odds ratio among the features even in the normal stage (FBG is less than 100). In addition, factors related to liver function, such as gamma-glutamyl transpeptidase can be used to predict progression from normal to prediabetes, while factors related to renal function, such as blood urea nitrogen and creatinine, are prediction factors of T2D development.

Abstract 14630: Cigarette Smoking and Risk of Carotid Atherosclerotic Lesions Among Chinese Adults With or Without Cardiometabolic Disorders: A Longitudinal Study

Backgroud: Cigarette smoking is closely associated with the presence of carotid atherosclerotic lesions (CAL); however, whether such association differs in participants with or without cardiometabolic disorders (CD) is largely unknown. Methods: A dynamic retrospective cohort study of health checkup data was used from the Beijing MJ Health Screening Centers. A total of 14 410 participants were included among those who took both routine tests and carotid ultrasonography (CU) at least twice from 2008 to 2018 and had no CAL at the first time. Information on smoking status was collected from self-reported questionnaires. CAL was defined as increased carotid intima-media thickness, carotid plaques, or carotid stenosis. CD was defined with the presence of hypertension, diabetes, and/or dyslipidemia. The associations of smoking with CAL among participants with or without CD and the joint effect of smoking and CD were estimated by logistic regression. Results: Among 5810 and 8600 participants with and without CD at baseline, 2016 and 1386 developed CAL during follow-up, respectively. We found the status of CD significantly modified the associations between smoking and CAL ( P interaction &lt; 0.01). Among participants without CD, non-current smoking was independently associated with a decreased risk of CAL (odds ratio [OR] and 95% confidence interval [CI]: 0.80 [0.68-0.94]). Whereas, such association was not observed among CD patients. In joint analyses, compared with current smokers with CD (CD + &amp;smoking + ), the ORs (95% CIs) were 0.72 (0.61-0.84), 0.94 (0.82-1.07), and 0.56 (0.49-0.64) for participants in the CD - &amp;smoking + , CD + &amp;smoking - , and CD - &amp;smoking - group, respectively. Conclusion: Never and former smokers without CD might have a lower risk of CAL, but not the ones prevalent with CD, highlighting the carotid atherosclerotic benefits of early smoking cessation.

Association between serum ferritin levels and colorectal cancer risk in Korea.

The concentration of serum ferritin, a storage form of iron, may be associated with carcinogenesis in various cancers. There are only limited studies on the relationship between serum ferritin levels and colorectal cancer (CRC) risk, especially in the Asian population. This study aimed to analyze the association between CRC incidence and serum ferritin levels. This was a national cohort study that used health checkup and insurance claims data of the Korean population. CRC incidence according to the serum ferritin level was analyzed during 2008-2018 in 17,116 participants. The hazard ratio (HR) of CRC incidence decreased as serum ferritin levels increased (Q1: HR, 1.000 [95% confidence interval [CI], reference]; Q2: HR, 0.811 [95% CI, 0.558 to 1.178]; Q3: HR, 0.654 [95% CI, 0.442 to 0.968]; Q4: HR, 0.443 [95% CI, 0.285 to 0.687]; p = 0.0026). In subgroup analysis, 40 to 64 years of age, sex, body mass index of &lt; 25 kg/ m2, presence of metabolic syndrome, absence of diabetes mellitus, and absence of anemia had HRs of &lt; 0.5 (95% CI) in the highest quartiles compared with that in the lowest quartiles. This study shows an inverse association between serum ferritin and CRC risk. Serum ferritin measurement can aid in identifying young adults requiring active CRC screening.

Development and Evaluation of Machine Learning-Based High-Cost Prediction Model Using Health Check-Up Data by the National Health Insurance Service of Korea.

In this study, socioeconomic, medical treatment, and health check-up data from 2010 to 2017 of the National Health Insurance Service (NHIS) of Korea were analyzed. This year’s socioeconomic, treatment, and health check-up data are used to develop a predictive model for high medical expenses in the next year. The characteristic of this study is to derive important variables related to the high cost of domestic medical expenses users by using data on health check-up items conducted by the country. In this study, we tried to classify data and evaluate its performance using classification supervised learning algorithms for high-cost medical expense prediction. Supervised learning for predicting high-cost medical expenses was performed using the logistic regression model, random forest, and XGBoost, which have been known to result the best performance and explanatory power among the machine learning algorithms used in previous studies. Our experimental results show that the XGBoost model had the best performance with 77.1% accuracy. The contribution of this study is to identify the variables that affect the prediction of high-cost medical expenses by analyzing the medical bills using the health check-up variables and the Korea Classification Disease (KCD) large group as input variables. Through this study, it was confirmed that musculoskeletal disorders (M) and respiratory diseases (J), which are the most frequently treated diseases, as important KCD disease groups for high-cost prediction in Korea, affect the future high cost prediction. In addition, it was confirmed that malignant neoplasia diseases (C) with high medical cost per treatment are a group of diseases related to high future medical cost prediction. Unlike previous studies, it is the result of analyzing all disease data, so it is expected that the study will be more meaningful when compared with the results of other national health check-up data.