scholarly journals Machine Learning Prediction Models for Chronic Kidney Disease Using National Health Insurance Claim Data in Taiwan

Healthcare ◽  
2021 ◽  
Vol 9 (5) ◽  
pp. 546
Author(s):  
Surya Krishnamurthy ◽  
Kapeleshh KS ◽  
Erik Dovgan ◽  
Mitja Luštrek ◽  
Barbara Gradišek Piletič ◽  
...  
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Health Insurance ◽  
Kidney Disease ◽  
National Health Insurance ◽  
National Health ◽  
Prediction Models ◽  
Research Database ◽  
Close Monitoring ◽  
Number Of Patients

Chronic kidney disease (CKD) represents a heavy burden on the healthcare system because of the increasing number of patients, high risk of progression to end-stage renal disease, and poor prognosis of morbidity and mortality. The aim of this study is to develop a machine-learning model that uses the comorbidity and medication data obtained from Taiwan’s National Health Insurance Research Database to forecast the occurrence of CKD within the next 6 or 12 months before its onset, and hence its prevalence in the population. A total of 18,000 people with CKD and 72,000 people without CKD diagnosis were selected using propensity score matching. Their demographic, medication and comorbidity data from their respective two-year observation period were used to build a predictive model. Among the approaches investigated, the Convolutional Neural Networks (CNN) model performed best with a test set AUROC of 0.957 and 0.954 for the 6-month and 12-month predictions, respectively. The most prominent predictors in the tree-based models were identified, including diabetes mellitus, age, gout, and medications such as sulfonamides and angiotensins. The model proposed in this study could be a useful tool for policymakers in predicting the trends of CKD in the population. The models can allow close monitoring of people at risk, early detection of CKD, better allocation of resources, and patient-centric management.

scholarly journals Machine Learning Prediction Models for Chronic Kidney Disease using National Health Insurance Claim Data in Taiwan

2020 ◽  
Author(s):  
Surya Krishnamurthy ◽  
Kapeleshh KS ◽  
Erik Dovgan ◽  
Mitja Luštrek ◽  
Barbara Gradišek Piletič ◽  
...  
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Health Insurance ◽  
Kidney Disease ◽  
National Health Insurance ◽  
National Health ◽  
Performance Metrics ◽  
Prediction Models ◽  
Research Database ◽  
Number Of Patients

ABSTRACTBackground and ObjectiveChronic kidney disease (CKD) represent a heavy burden on the healthcare system because of the increasing number of patients, high risk of progression to end-stage renal disease, and poor prognosis of morbidity and mortality. The aim of this study is to develop a machine-learning model that uses the comorbidity and medication data, obtained from Taiwan’s National Health Insurance Research Database, to forecast whether an individual will develop CKD within the next 6 or 12 months, and thus forecast the prevalence in the population.MethodsA total of 18,000 people with CKD and 72,000 people without CKD diagnosis along with the past two years of medication and comorbidity data matched by propensity score were used to build a predicting model. A series of approaches were tested, including Convoluted Neural Networks (CNN). 5-fold cross-validation was used to assess the performance metrics of the algorithms.ResultsBoth for the 6 month and 12-month models, the CNN approach performed best, with the AUROC of 0.957 and 0.954, respectively. The most prominent features in the tree-based models were identified, including diabetes mellitus, age, gout, and medications such as sulfonamides, angiotensins which had an impact on the progression of CKD.ConclusionsThe model proposed in this study can be a useful tool for the policy-makers helping them in predicting the trends of CKD in the population in the next 6 to 12 months. Information provided by this model can allow closely monitoring the people with risk, early detection of CKD, better allocation of resources, and patient-centric management

scholarly journals Changes in Prevalence and Health Checkup Coverage Rate of Chronic Kidney Disease (CKD) after Introduction of Prefecture-Wide CKD Initiative: Results of the Kagawa Association of CKD Initiatives

2021 ◽  
Vol 11 (11) ◽  
pp. 1121
Author(s):  
Tadashi Sofue ◽  
Taiga Hara ◽  
Yoko Nishijima ◽  
Satoshi Nishioka ◽  
Hiroyuki Watatani ◽  
...  
Keyword(s):  
Chronic Kidney Disease ◽  
Health Insurance ◽  
Kidney Disease ◽  
National Health Insurance ◽  
National Health ◽  
Coverage Rate ◽  
Medical Visit ◽  
Health Checkup ◽  
Insurance Database ◽  
Special Health

The National Health Insurance (NHI) special health checkup system in Japan targets the NHI population aged 40–74 years. Since 2015, the Kagawa NHI special health checkup was initiated in a prefecture-wide chronic kidney disease (CKD) initiative, including renal examination as an essential item in NHI health checkups. Here, we aimed to investigate the effects of the prefecture-wide CKD initiative. We conducted a retrospective cohort survey using the Kagawa National Health Insurance database created by the Kagawa National Health Insurance Organization. Results of the NHI health checkup (2015–2019) and prefecture-wide outcomes (2013–2019) were analyzed. The prevalence of CKD among examinees who underwent the NHI health checkup increased from 17.7% in 2015 to 23.2% in 2019. The percentage of examinees who completed a medical visit was 29.4% in 2015. After initiation of the initiative, the NHI health checkup coverage rate increased significantly, from a mean (standard deviation) of 40.8% (0.4%) to 43.2% (1.1%) (p = 0.04). After the start of the CKD initiative, we found an increase in the prevalence of CKD and the NHI health checkup coverage rate.

scholarly journals Income Levels and Mortality Rates in Patients with Chronic Kidney Disease in South Korea: a Population-Based Cohort Study Using National Health Insurance Data

2020 ◽  
Author(s):  
Young Choi
Keyword(s):  
Chronic Kidney Disease ◽  
Health Insurance ◽  
Kidney Disease ◽  
Mortality Rate ◽  
National Health Insurance ◽  
National Health ◽  
Mortality Rates ◽  
Dialysis Patients ◽  
Income Levels ◽  
Low Incomes

Background: To examine the association between income levels and mortality rates in patients with chronic kidney disease. Methods: We analyzed data obtained from 3,172 patients with chronic kidney disease obtained from the Korean National Health Insurance claims database (2003–2009). Each patient was monitored until December 2010 or until death, whichever came first. Individual income was estimated from the national health insurance premium. Information on mortality was obtained from the Korean National Statistical Office. Cox proportional hazard models were used to compare mortality rates between different income groups after adjusting for possible confounding risk factors. Results: A low income was significantly associated with a high mortality rate after adjusting for covariates (adjusted HR 1.298 [1.082–1.556]). In addition, dialysis patients who had low incomes were more likely to have higher mortality rates compared to those in dialysis patients who had high incomes (adjusted HR 1.528 [1.122–2.082]). Conclusion: The findings of this study indicate that chronic kidney disease patients with low incomes have the highest mortality risk. Promotion of targeted policies and priority health services for patients with low incomes may help reduce the mortality rate in this vulnerable group.

scholarly journals Prevalence of Chronic Kidney Disease and Variation of Its Risk Factors by the Regions in Okayama Prefecture

2022 ◽  
Vol 12 (1) ◽  
pp. 97
Author(s):  
Ryoko Umebayashi ◽  
Haruhito Adam Uchida ◽  
Natsumi Matsuoka-Uchiyama ◽  
Hitoshi Sugiyama ◽  
Jun Wada
Keyword(s):  
Risk Factors ◽  
Chronic Kidney Disease ◽  
Health Insurance ◽  
Kidney Disease ◽  
National Health Insurance ◽  
National Health ◽  
Health Problems ◽  
Ckd Progression ◽  
Increased Risk ◽  
Okayama Prefecture

Objective: The prevention of chronic kidney disease (CKD) progression is an important issue from health and financial perspectives. We conducted a single-year cross-sectional study to clarify the prevalence of CKD and its risk factors along with variations in these factors among five medical regions in Okayama Prefecture, Japan. Methods and Results: Data concerning the renal function and proteinuria as well as other CKD risk factors were obtained from the database of the Japanese National Health Insurance. The proportion of CKD patients at an increased risk of progression to end-stage renal disease (ESRD), classified as orange and red on the CKD heatmap, ranged from 6–9% and did not vary significantly by the regions. However, the causes of the increased severity differed between regions where renal dysfunction was predominant and regions where there were many patients with proteinuria. CKD risk factors, such as diabetes mellitus, hypertension, hyper low-density lipoprotein-cholesterolemia, obesity, smoking and lack of exercise, also differed among these regions, suggesting that different regions need tailored interventions that suit the characteristics of the region, such as an increased health checkup ratio, dietary guidance and promotion of exercise opportunities. Conclusions: Approximately 6–9% of people are at an increased risk of developing ESRD (orange or red on a CKD heatmap) among the population with National Health Insurance in Okayama Prefecture. The underlying health problems that cause CKD may differ among the regions. Thus, it is necessary to consider intervention methods for preventing CKD progression that are tailored to each region’s health problems.

Disparities in Kidney Transplantation Access among Korean Patients Initiating Dialysis: A Population-Based Cohort Study Using National Health Insurance Data (2003-2013)

2016 ◽  
Vol 45 (1) ◽  
pp. 32-39 ◽  
Author(s):  
Young Choi ◽  
Jaeyong Shin ◽  
Jung Tak Park ◽  
Kyoung Hee Cho ◽  
Eun-Cheol Park ◽  
...  
Keyword(s):  
Chronic Kidney Disease ◽  
Socioeconomic Status ◽  
Kidney Transplantation ◽  
Health Insurance ◽  
Kidney Disease ◽  
South Korea ◽  
National Health Insurance ◽  
National Health ◽  
Income Level ◽  
Kidney Transplants

Background: The socioeconomic status of a person has an impact on his or her access to kidney transplantation as has been reported in western countries. This study examined the association between income level and kidney transplantation among chronic kidney disease patients undergoing dialysis in South Korea. Methods: We analyzed data from 1,792 chronic kidney disease patients undergoing dialysis and listed in the Korean National Health Insurance Claim Database (2003-2013). The likelihood of receiving the first kidney transplant over time was analyzed using competing risk proportional hazard models on time from initiating dialysis to receiving a transplant. Results: Of 1,792 patients on dialysis, only 184 patients (10.3%) received kidney transplants. Patients with medical aid had the lowest kidney transplantation rate (hazard ratio 0.29, 95% CI 0.16-0.51). A lower income level was significantly associated with a low kidney transplantation rate, after adjusting for covariates, compared to patients in the high-income level group. Conclusions: Our findings indicate that in South Korea, the total number of kidney transplants is remarkably low and there exists income disparity with regard to access to kidney transplantation. Thus, we suggest that plans be implemented to encourage organ donation and increase organ transplant accessibility for all patients irrespective of their socioeconomic status.

scholarly journals Machine Learning-Based Cardiovascular Disease Prediction Model: A Cohort Study on the Korean National Health Insurance Service Health Screening Database

Diagnostics ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 943
Author(s):  
Joung Ouk (Ryan) Kim ◽  
Yong-Suk Jeong ◽  
Jin Ho Kim ◽  
Jong-Weon Lee ◽  
Dougho Park ◽  
...  
Keyword(s):  
Machine Learning ◽  
Health Insurance ◽  
Prediction Model ◽  
National Health Insurance ◽  
National Health ◽  
Prediction Models ◽  
Characteristic Curve ◽  
Health Screening ◽  
Gradient Boosting ◽  
Extreme Gradient Boosting

Background: This study proposes a cardiovascular diseases (CVD) prediction model using machine learning (ML) algorithms based on the National Health Insurance Service-Health Screening datasets. Methods: We extracted 4699 patients aged over 45 as the CVD group, diagnosed according to the international classification of diseases system (I20–I25). In addition, 4699 random subjects without CVD diagnosis were enrolled as a non-CVD group. Both groups were matched by age and gender. Various ML algorithms were applied to perform CVD prediction; then, the performances of all the prediction models were compared. Results: The extreme gradient boosting, gradient boosting, and random forest algorithms exhibited the best average prediction accuracy (area under receiver operating characteristic curve (AUROC): 0.812, 0.812, and 0.811, respectively) among all algorithms validated in this study. Based on AUROC, the ML algorithms improved the CVD prediction performance, compared to previously proposed prediction models. Preexisting CVD history was the most important factor contributing to the accuracy of the prediction model, followed by total cholesterol, low-density lipoprotein cholesterol, waist-height ratio, and body mass index. Conclusions: Our results indicate that the proposed health screening dataset-based CVD prediction model using ML algorithms is readily applicable, produces validated results and outperforms the previous CVD prediction models.

scholarly journals Hyperlipidemia Is a Risk Factor of Adhesive Capsulitis: Real-World Evidence Using the Taiwanese National Health Insurance Research Database

2021 ◽  
Vol 9 (4) ◽  
pp. 232596712098680
Author(s):  
Jr-Yi Wang ◽  
Chen-Kun Liaw ◽  
Chi-Chang Huang ◽  
Tsan-Hon Liou ◽  
Hui-Wen Lin ◽  
...  
Keyword(s):  
Risk Factor ◽  
Health Insurance ◽  
Propensity Score ◽  
National Health Insurance ◽  
National Health ◽  
Glenohumeral Joint ◽  
Adhesive Capsulitis ◽  
Small Sample ◽  
Research Database ◽  
Statin Use

Background: Patients with adhesive capsulitis are evaluated for pain and progressive contracture of the glenohumeral joint. Whether endocrine, immune, or inflammatory processes are involved in its definite pathogenesis is still under debate. Some cross-sectional studies with a small sample size have noted that hyperlipidemia is a possible risk factor for frozen shoulders. Purpose/Hypothesis: The purpose was to conduct a longitudinal population-based study to investigate the risk of adhesive capsulitis among patients with hyperlipidemia. It was hypothesized that patients with hyperlipidemia would have a higher risk of adhesive capsulitis and that the use of statin drugs could reduce the rate. Study Design: Cohort study; Level of evidence, 3. Methods: Using data from the National Health Insurance Research Database (NHIRD) of Taiwan, the authors obtained the records of patients with hyperlipidemia who received a diagnosis between 2004 and 2005 and were followed up until the end of 2010. The control cohort comprised age- and sex-matched patients without hyperlipidemia. Propensity score matching was performed for the other comorbidities. A Cox multivariate proportional hazards model was applied to analyze the risk factors of adhesive capsulitis. The hazard ratio (HR) and adjusted HR were estimated between the study and control cohorts after adjustment for confounders. The effects of statin use on adhesive capsulitis risk were also analyzed. Results: The NHIRD records of 28,748 patients and 114,992 propensity score–matched controls were evaluated. A higher incidence rate of adhesive capsulitis was revealed in the hyperlipidemia cohort, with a crude HR of 1.70 (95% CI, 1.61-1.79; P < .001) and adjusted HR of 1.50 (95% CI, 1.41-1.59; P < .001). Patients with hyperlipidemia who used a statin still had higher crude and adjusted HRs compared with controls. Statin use did not exert protective effects on patients with hyperlipidemia. Conclusion: Patients with hyperlipidemia had a 1.5-fold higher risk of adhesive capsulitis than did healthy controls. Statin use did not provide protection against adhesive capsulitis in patients with hyperlipidemia.

scholarly journals Impact of comorbidity assessment methods to predict non-cancer mortality risk in cancer patients: a retrospective observational study using the National Health Insurance Service claims-based data in Korea

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Sanghee Lee ◽  
Yoon Jung Chang ◽  
Hyunsoon Cho
Keyword(s):  
Risk Assessment ◽  
Health Insurance ◽  
Cancer Patients ◽  
National Health Insurance ◽  
National Health ◽  
Cancer Mortality ◽  
Prediction Models ◽  
Assessment Methods ◽  
Chronic Obstructive ◽  
Mortality Risks

Abstract Background Cancer patients’ prognoses are complicated by comorbidities. Prognostic prediction models with inappropriate comorbidity adjustments yield biased survival estimates. However, an appropriate claims-based comorbidity risk assessment method remains unclear. This study aimed to compare methods used to capture comorbidities from claims data and predict non-cancer mortality risks among cancer patients. Methods Data were obtained from the National Health Insurance Service-National Sample Cohort database in Korea; 2979 cancer patients diagnosed in 2006 were considered. Claims-based Charlson Comorbidity Index was evaluated according to the various assessment methods: different periods in washout window, lookback, and claim types. The prevalence of comorbidities and associated non-cancer mortality risks were compared. The Cox proportional hazards models considering left-truncation were used to estimate the non-cancer mortality risks. Results The prevalence of peptic ulcer, the most common comorbidity, ranged from 1.5 to 31.0%, and the proportion of patients with ≥1 comorbidity ranged from 4.5 to 58.4%, depending on the assessment methods. Outpatient claims captured 96.9% of patients with chronic obstructive pulmonary disease; however, they captured only 65.2% of patients with myocardial infarction. The different assessment methods affected non-cancer mortality risks; for example, the hazard ratios for patients with moderate comorbidity (CCI 3–4) varied from 1.0 (95% CI: 0.6–1.6) to 5.0 (95% CI: 2.7–9.3). Inpatient claims resulted in relatively higher estimates reflective of disease severity. Conclusions The prevalence of comorbidities and associated non-cancer mortality risks varied considerably by the assessment methods. Researchers should understand the complexity of comorbidity assessments in claims-based risk assessment and select an optimal approach.

Sign in / Sign up

Export Citation Format

Share Document