Impact of COVID-19 pandemic on healthcare service use for non-COVID-19 patients in Japan: retrospective cohort study

OBJECTIVES

We aimed to investigate the impact of the first and second waves of the COVID-19 pandemic on healthcare service use by non-COVID-19 patients.

DESIGN

Retrospective cohort study.

SETTING

Hospital-based claims database from anonymised hospitals in Japan.

PARTICIPANTS

Patients (n=785 495) who visited and/or were hospitalised in 26 anonymised hospitals in Japan between January 2017 and November 2020.

OUTCOME MEASURES

We compared changes in the monthly number of hospitalisations (overall or by diagnosis), outpatient visits, endoscopic fibrescopies (EFs), rehabilitations, outpatient chemotherapy treatments, maintenance haemodialysis treatments and outpatient prescriptions between pre-COVID-19 years and the same period in 2020.

RESULTS

The overall number of hospitalisations and outpatient visits decreased by 27% and 22%, respectively, in May 2020, of which the most substantial decrease was observed in the paediatrics department (65% and 51%, respectively). The number of hospitalisations for respiratory diseases, circulatory diseases, malignant neoplasms and digestive diseases decreased by a maximum of 55%, 32%, 10% and 26%, respectively, in 2020. The number of hospitalisations for non-COVID-19 pneumonia in patients aged <16 years, patients aged ≥16 years and patients with asthma decreased by 93%, 43% and 80%, respectively, in May 2020. EFs and outpatient rehabilitations decreased by >30%. In contrast, outpatient chemotherapy and maintenance haemodialysis treatments decreased by <10%, if at all. Outpatient prescriptions decreased by a maximum of 20% in 2020, with the largest decrease observed in drugs for obstructive airway diseases and cough and cold preparations.

CONCLUSIONS

The use of healthcare services by non-COVID-19 patients was most affected during the first wave of the COVID-19 pandemic in May 2020. The number of hospitalisations for respiratory diseases, particularly non-COVID-19 pneumonia and asthma, drastically decreased, while the number of hospitalisations and outpatient chemotherapies for malignant neoplasms or maintenance haemodialysis was less affected.

Outcomes of lactulose plus branched-chain amino acid infusion and lactulose alone for hepatic encephalopathy: A retrospective cohort study using a national inpatient database

AIM

Previous randomized controlled trials of branched-chain amino acid infusion for hepatic encephalopathy involved a small number of patients, and the effectiveness of branched-chain amino acid infusion has not been clarified. We evaluated whether branched-chain amino acid infusion in addition to lactulose treatment was associated with short-term outcomes in a large population of patients.

METHODS

Using the Japanese Diagnosis Procedure Combination database, we retrospectively identified inpatients with hepatic encephalopathy who received lactulose within 2 days of admission from July 2011 to March 2017. We divided the patients into those who received branched-chain amino acid infusion within 2 days of admission in addition to lactulose treatment and those who did not. We conducted propensity-score inverse probability of treatment weighting analyses to compare in-hospital mortality, consciousness at discharge, in-hospital complications, length of stay, and total hospitalization costs between the groups.

RESULTS

We identified 8051 patients with hepatic encephalopathy treated with lactulose, including 7560 patients who received lactulose plus branched-chain amino acid infusion and 491 who received lactulose alone. The lactulose plus branched-chain amino acid infusion group had a significantly lower mortality than the lactulose alone group (9.6% vs. 15.0%, odds ratio 0.60, 95% confidence interval 0.44-0.82). The lactulose plus branched-chain amino acid infusion group also had a lower proportion of patients with impaired mental status at discharge, and lower total hospitalization costs than the lactulose alone group.

CONCLUSIONS

Branched-chain amino acid infusion together with lactulose may improve the prognosis of hepatic encephalopathy.

A Machine Learning-Based Predictive Model to Identify Patients Who Failed to Attend a Follow-up Visit for Diabetes Care After Recommendations From a National Screening Program

OBJECTIVE

Reportedly, two-thirds of the patients who were positive for diabetes during screening failed to attend a follow-up visit for diabetes care in Japan. We aimed to develop a machine-learning model for predicting people's failure to attend a follow-up visit.

RESEARCH DESIGN AND METHODS

We conducted a retrospective cohort study of adults with newly screened diabetes at a national screening program using a large Japanese insurance claims database (JMDC, Tokyo, Japan). We defined failure to attend a follow-up visit for diabetes care as no physician consultation during the 6 months after the screening. The candidate predictors were patient demographics, comorbidities, and medication history. In the training set (randomly selected 80% of the sample), we developed two models (previously reported logistic regression model and Lasso regression model). In the test set (remaining 20%), prediction performance was examined.

RESULTS

We identified 10,645 patients, including 5,450 patients who failed to attend follow-up visits for diabetes care. The Lasso regression model using four predictors had a better discrimination ability than the previously reported logistic regression model using 13 predictors (C-statistic: 0.71 [95% CI 0.69-0.73] vs. 0.67 [0.65-0.69]; P < 0.001). The four selected predictors in the Lasso regression model were lower frequency of physician visits in the previous year, lower HbA1c levels, and negative history of antidyslipidemic or antihypertensive treatment.

CONCLUSIONS

The developed machine-learning model using four predictors had a good predictive ability to identify patients who failed to attend a follow-up visit for diabetes care after a screening program.

Association between nutritional guidance or ophthalmological examination and discontinuation of physician visits in patients with newly diagnosed diabetes: A retrospective cohort study using a nationwide database

AIMS/INTRODUCTION

Discontinuation of diabetes care has been studied mostly in patients with prevalent diabetes and not in patients with newly diagnosed diabetes, whose dropout risk is highest. Because enrolling patients in a prospective study will influence adherence, we retrospectively examined whether guideline-recommended practices, defined as nutritional guidance or ophthalmological examination, can prevent patient discontinuation of diabetes care after its initiation.

MATERIALS AND METHODS

We retrospectively identified adults with newly screened diabetes during checkups using a large Japanese administrative claims database (JMDC, Tokyo, Japan) that contains laboratory data and lifestyle questionnaires. We defined discontinuation of physician visits as a follow-up interval exceeding 6 months. We divided the patients into those who received guideline-recommended practices (nutritional guidance or ophthalmology consultation) within the same month as the first visit and those who did not. We calculated propensity scores and carried out inverse probability of treatment weighting analyses to compare discontinuation between the two groups.

RESULTS

We identified 6,508 patients with at least one physician consultation for diabetes care within 3 months after their checkup, including 4,574 patients without and 1,934 with guideline-recommended practices. After inverse probability of treatment weighting, patients with guideline-recommended practices had a significantly lower proportion of discontinuation than those without (17.2% vs 21.8%; relative risk 0.79, 95% confidence interval 0.69-0.91).

CONCLUSIONS

This study is the first to show that after adjustment for both patient and healthcare provider factors, guideline-recommended practices within the first month of physician consultation for diabetes care can decrease subsequent discontinuation of physician visits in patients with newly diagnosed diabetes.

Potassium Concentration in Initial Fluid Therapy and In-Hospital Mortality of Patients with Diabetic Ketoacidosis

CONTEXT

Guidelines worldwide recommend potassium replacement of 10 to 40 mmol/L in the initial fluid therapy for patients with diabetic ketoacidosis. However, evidence is lacking as to the association between infused potassium concentration and mortality.

OBJECTIVE

We aimed to determine the association between infused potassium concentration and in-hospital mortality.

METHODS

Using the Japanese Diagnosis Procedure Combination database, we retrospectively identified inpatients admitted for treatment of diabetic ketoacidosis from July 2010 to March 2018. Patients with kidney dysfunction or serum potassium abnormalities were excluded. We evaluated the association of the potassium concentration in the total infused solutions in the first 2 days of hospitalization with 28-day in-hospital mortality using multivariable regression analysis with a cubic spline model. We also assessed the association between potassium concentration and occurrence of hyperkalemia.

RESULTS

We identified 14 216 patients with diabetic ketoacidosis and observed 261 deaths. The quartile cut-points for potassium concentration were 7.7, 11.4, and 16.1 mmol/L. Within the range of approximately 10 to 40 mmol/L, potassium concentration was not associated with occurrence of hyperkalemia or death. Lower potassium concentrations were associated with higher 28-day in-hospital mortality; the odds ratio for patients receiving 8 mmol/L was 1.69 (95% CI, 1.03 to 2.78; reference: 20 mmol/L), and the odds ratio increased monotonically as potassium concentration decreased further.

CONCLUSION

Patients receiving potassium replacement at concentrations of 10 to 40 mmol/L had similar in-hospital mortality rates, whereas lower concentrations were associated with higher mortality.

Effect of Branched-Chain Amino Acid Infusion on In-Hospital Mortality of Patients With Hepatic Encephalopathy and End-Stage Kidney Disease: A Retrospective Cohort Study Using a National Inpatient Database

OBJECTIVES

Renal failure and hepatic cirrhosis are mutually aggravating factors. However, no specific therapeutic strategies for hepatic encephalopathy (HE) and end-stage kidney disease have been established. The coexistence, with an extremely poor prognosis, makes randomized controlled trials unfeasible. We evaluated whether an infusion of branched-chain amino acids was associated with mortality in patients hospitalized for HE and end-stage kidney disease.

DESIGN AND METHODS

Using the Japanese Diagnosis Procedure Combination database, we retrospectively identified patients with HE and end-stage kidney disease who received hemodialysis within 2 days of admission from July 2011 to March 2017. We divided the patients into those who received branched-chain amino acid infusion within 2 days of admission and those who did not. We conducted analyses using overlap weights based on propensity scores to compare in-hospital mortality between the groups. Sub-group analysis was conducted by stratifying patients by Child-Pugh class.

RESULTS

We identified 553 eligible patients, including 503 patients who received branched-chain amino acid infusion and 50 who did not. The patients who received branched-chain amino acid infusion had lower mortality than those who did not (10.2% vs. 20.1%, relative risk 0.51, 95% confidence interval 0.27-0.95). Sub-group analysis showed that branched-chain amino acid infusion was associated with decreased in-hospital mortality in patients with Child-Pugh class C (16.2% vs. 39.0%, relative risk 0.41, 95% confidence interval 0.23-0.76).

CONCLUSIONS

Branched-chain amino acid infusion may improve the prognosis of HE in patients with end-stage kidney disease, particularly those with lower liver function. Further research is necessary to provide a suitable treatment for HE in patients with end-stage kidney disease.

Modification Effects of Albuminuria on the Association Between Kidney Function and Development of Anemia in Diabetes

CONTEXT

Previous studies failed to adjust for estimated glomerular filtration rate (eGFR) in evaluating the association between albuminuria and anemia development, and we aimed to investigate whether albuminuria independently affects anemia development.

METHODS

We conducted a retrospective cohort study and retrospectively identified adults with diabetes from a Japanese nationwide clinical database (JMDC, Tokyo, Japan). To assess the modification effects of albuminuria on the association between eGFR and anemia development, we estimated prevalence of anemia, defined as hemoglobin < 13 g/dL in men and < 12 g/dL in women, using a modified Poisson regression and marginal standardization form of predictive margins, stratified by albuminuria severity after adjusting for eGFR. Hence, we revealed at which eGFR level this modification effect appeared and the extent to which this modification effect increased the prevalence of anemia.

RESULTS

We identified 327 999 data points from 48 056 individuals [normoalbuminuria: 186 472 (56.9%), microalbuminuria: 107 170 (32.7%), and macroalbuminuria: 34 357 (10.5%)]. As eGFR declined, anemia prevalence increased. Albuminuria severity modified this association induced by decreased eGFR among individuals with eGFR <30 mL/min/1.73 m2 after adjusting for multivariable factors, including age, sex, comorbidities, and medication use. Compared with the normoalbuminuric group, the macroalbuminuric group had a 5% to 20% higher anemia prevalence among individuals with eGFR of <30 mL/min/1.73 m2.

CONCLUSION

We revealed that the severity of albuminuria modified the association between eGFR and anemia development among individuals with eGFR <30 mL/min/1.73 m2, highlighting the modification effect of albuminuria on the association between kidney function and anemia development in diabetes.

Diagnostic ability using fatty liver and metabolic markers for metabolic-associated fatty liver disease stratified by metabolic/glycemic abnormalities

AIMS/INTRODUCTION

Although several noninvasive predictive markers for fatty liver and metabolic markers have been used for fatty liver prediction, whether such markers can also predict metabolic-associated fatty liver disease (MAFLD) remains unclear. We aimed to examine the ability of existing fatty liver or metabolic markers to predict MAFLD.

MATERIALS AND METHODS

Participants in a high-volume center in Tokyo were classified into groups with and without MAFLD, based on the presence of metabolic abnormalities and fatty liver diagnosed through abdominal ultrasonography, between 2008 and 2018. The diagnostic abilities of three fatty liver markers: fatty liver index (FLI), hepatic steatosis index (HSI), and lipid accumulation product (LAP), and three common metabolic markers: waist-to-height ratio (WHR), body mass index (BMI), and waist circumference (WC), for predicting MAFLD, were evaluated. Analyses stratified by MAFLD subtypes were performed.

RESULTS

Of 92,374 individuals, 19,392 (36.1%) had MAFLD. The diagnostic performances for MAFLD prediction, measured as c-statistics, for FLI, HSI, LAP, WHR, BMI, and WC were 0.906, 0.892, 0.878, 0.844, 0.877, and 0.878, respectively. Optimal cutoff values for diagnosing MAFLD for FLI, HSI, LAP, WHR, BMI, and WC were 20.3, 32.7, 20.0, 0.49, 22.9, and 82.1, respectively. Analyses stratified by MAFLD subtypes, based on BMI and metabolic/glycemic abnormalities, suggested that FLI and HSI had acceptable (c-statistics >0.700) diagnostic abilities throughout all the analyses.

CONCLUSIONS

All six markers were excellent predictors of MAFLD in diagnosing among the general population, with FLI and HSI particularly useful among all sub-populations.

Major complications after percutaneous biopsy of native or transplanted liver in pediatric patients: a nationwide inpatient database study in Japan

Aim

Although major complication rates following percutaneous liver biopsy (PLB) have been reported to be higher in children than in adults, scarce data are available regarding pediatric patients stratified by native and transplanted liver. We aimed to assess the factors associated with major complications after percutaneous biopsy of native or transplanted liver using a nationwide inpatient database.

Methods

Using the Japanese Diagnosis Procedure Combination database, we retrospectively identified pediatric patients who underwent PLB between 2010 and 2018. We described major complication rates and analyzed factors associated with major complications following PLB, stratified by native and transplanted liver.

Results

We identified 3584 pediatric PLBs among 1732 patients from 239 hospitals throughout Japan during the study period, including 1310 in the native liver and 2274 in the transplanted liver. Major complications following PLB were observed in 0.5% (n = 18) of the total cases; PLB in the transplanted liver had major complications less frequently than those in the native liver (0.2% vs. 1.0%, p = 0.002). The occurrence of major complications was associated with younger age, liver cancers, unscheduled admission, anemia or coagulation disorders in cases with native liver, while it was associated with younger age alone in cases with transplanted liver.

Conclusions

The present study, using a nationwide database, found that major complications occurred more frequently in pediatric cases with native liver and identified several factors associated with its major complications.

Comparison of bleeding complications after pediatric kidney biopsy between intravenous sedation and general anesthesia: a nationwide cohort study

BACKGROUND

An increasing number of studies are evaluating the safety of intravenous sedation compared with that of general anesthesia; however, data on bleeding complications after pediatric percutaneous renal biopsy performed under intravenous sedation or general anesthesia are lacking. We aimed to examine differences in bleeding complications between intravenous sedation and general anesthesia in pediatric patients.

METHODS

Data of pediatric patients aged ≤ 15 years undergoing percutaneous kidney biopsy for kidney disease between July 2007 and March 2019 were retrieved from a national inpatient database in Japan. We examined differences in bleeding complications after renal biopsy performed under intravenous sedation, defined by the absence of the record of general anesthesia with intubation but by the presence of intravenous sedation during biopsy, and general anesthesia, defined by the presence of the record of general anesthesia with intubation during biopsy, among pediatric patients admitted for percutaneous renal biopsy. We performed binomial regression using overlap weights based on propensity scores for patients receiving intravenous sedation. Analyses stratified by age or sex, a sensitivity analysis using generalized estimating equations considering cluster effects by hospital among a propensity score-matched cohort, and another sensitivity analysis using the instrumental variable method were performed to confirm the robustness of the results.

RESULTS

We identified 6,560 biopsies performed in 5,999 children aged 1-15 years from 328 hospitals and 178 events. Only three severe complications and no death were observed. No significant difference in the proportion of bleeding complications was observed between procedures performed under intravenous sedation and those performed under general anesthesia (unadjusted proportions, 2.8% and 2.3%; adjusted proportions, 2.5% and 2.2%), with an unadjusted relative risk of 1.21 (95% confidence interval, 0.80-1.81) and adjusted relative risk of 1.13 (95% confidence interval, 0.74-1.73). Both age- and sex-stratified analyses yielded similar results. The analysis using generalized estimating equation and the instrumental variable method showed relative risks of 0.95 (95% confidence interval, 0.48-1.88) and 1.18 (95% confidence interval, 0.74-1.89), respectively.

CONCLUSION

This retrospective cohort study using a national database revealed that the risk of biopsy-related bleeding was comparable between intravenous sedation and general anesthesia during pediatric percutaneous kidney biopsy, suggesting that intravenous sedation alone and general anesthesia may have a similar bleeding risk in pediatric percutaneous kidney biopsies.

Adding biomarker change information to the kidney failure risk equation improves predictive ability for dialysis dependency in eGFR <30 ml/min/1.73 m2

Background

Although the kidney failure risk equation (KFRE), a well-known predictive model for predicting dialysis dependency, is useful, it remains unclear whether the addition of biomarker changes to the KFRE model in patients with an estimated glomerular filtration rate (eGFR) <30 ml/min/1.73 m2 will improve its predictive value.

Methods

We retrospectively identified adults with eGFR <30 ml/min/1.73 m2 without dialysis dependency, and available health checkup data for two successive years using a large Japanese claims database (DeSC, Tokyo, Japan). We dichotomized the entire population into a training set (50%) and a validation set (the other half). To assess the incremental value in the predictive ability for dialysis dependency by the addition of changes in eGFR and proteinuria, we calculated the difference in the C-statistics and net reclassification index (NRI).

Results

We identified 4499 individuals and observed 422 individuals (incidence of 45.2 per 1000 person-years) who developed dialysis dependency during the observation period (9343 person-years). Adding biomarker changes to the KFRE model improved C-statistics from 0.862 to 0.921, with an improvement of 0.060 (95% confidence intervals (CI) of 0.043-0.076, P < .001). The corresponding NRI was 0.773 (95% CI: 0.637-0.908), with an NRI for events of 0.544 (95% CI of 0.415-0.672) and NRI for non-events of 0.229 (95% CI of 0.186-0.272).

Conclusions

The KFRE model was improved by incorporating yearly changes in its components. The added information may help clinicians identify high-risk individuals and improve their care.

Impact of the COVID-19 pandemic and COVID-19 downgrade on non-COVID-19 respiratory diseases in Japan

OBJECTIVES

A worldwide decrease in the incidence of respiratory diseases during the coronavirus disease (COVID-19) pandemic has been reported, largely due to non-pharmaceutical interventions (NPIs). However, the impact of lifting NPIs remains unclear. In Japan, NPIs were lifted rather drastically when COVID-19 was downgraded on May 8, 2023. This study aimed to evaluate the impact of the COVID-19 pandemic and its downgrade on non-COVID-19 respiratory diseases using nationwide databases.

STUDY DESIGN

Retrospective cohort study.

METHODS

Monthly hospitalisation and prescription rates between January 2017 and October 2023 were collected from the JMDC insurance claims database covering 16,485,812 insured individuals. The monthly mortality rates in Japan were collected from an open data source published by Ministry of Health, Labour and Welfare of Japan. Interrupted time series analyses using seasonal autoregressive integrated moving average models were performed.

RESULTS

While hospitalisation rates for diseases such as pneumonia, asthma, and aspiration pneumonia decreased during the pandemic, a step increase in hospitalisations for these diseases and prescriptions for anti-asthma drugs was observed following the COVID-19 downgrade. The pandemic impacted all age groups; however, the impact of COVID-19 downgrade was more pronounced in children aged 0-5 years. Although mortality from non-COVID-19 respiratory diseases decreased during the pandemic, no immediate step increase in mortality was observed following the downgrade.

CONCLUSIONS

Although hospitalisations for pneumonia and prescriptions for anti-asthma drugs increased immediately after downgrading COVID-19, no step increase in mortality was observed presumably because older people were less affected than children.

Proteinuria screening and risk of bone fracture: a retrospective cohort study using a nationwide population-based database

Background and hypothesis

Proteinuria is associated with an increased risk of kidney function deterioration, cardiovascular disease, or cancer. Previous reports suggesting an association between kidney dysfunction and bone fracture may be confounded by concomitant proteinuria and were inconsistent regarding the association between proteinuria and bone fracture. Therefore, we aimed to evaluate the association using a large administrative claims database in Japan.

Methods

Using the DeSC database, we retrospectively identified individuals with laboratory data including urine dipstick test between August 2014 and February 2021. We evaluated the association between proteinuria and vertebral or hip fracture using multivariable Cox regression analyses adjusted for various background factors including kidney function. We also performed subgroup analyses stratified by sex and kidney function and sensitivity analyses with Fine & Gray models considering death as a competing risk.

Results

We identified 603 766 individuals and observed 21 195 fractures. With reference to the negative proteinuria group, the hazard ratio for hip or vertebral fracture was 1.10 [95% confidence interval (CI), 1.05-1.14] and 1.16 (95%CI, 1.11-1.22) in the trace and positive proteinuria group, respectively, in the Cox regression analysis. The subgroup analyses showed similar trends. The Fine & Gray model showed a subdistribution hazard ratio of 1.09 (95%CI, 1.05-1.14) in the trace proteinuria group and 1.15 (95% CI, 1.10-1.20) in the positive proteinuria group.

Conclusions

Proteinuria was associated with an increased risk of developing hip or vertebral fractures after adjustment for kidney function. Our results highlight the clinical importance of checking proteinuria for predicting bone fractures.

Modification effect of receipt of diabetes care on the association between COVID-19 infection and HbA1c level during the first year of the coronavirus pandemic using a nationwide population-based database

AIMS/INTRODUCTION

We assessed the modification effect of adherence to diabetes care on the association between HbA1c levels and the risk of coronavirus disease 2019 (COVID-19) among individuals with diabetes using a population-based database.

MATERIAL AND METHODS

We retrospectively identified individuals with diabetes during routine health checkups performed in 2019 in Japan using a population-based claims database (JMDC, Tokyo, Japan). We assessed the risk of COVID-19 infection in 2020 in relation to HbA1c levels during routine checkups, stratified by the presence/absence of follow-up for diabetes care in 2019. Several sensitivity analyses were performed.

RESULTS

We identified 65,956 individuals with an HbA1c ≥6.5% and fasting glucose ≥126 mg/dL in routine checkups, including 52,637 and 13,319 with and without at least one physician consultation for diabetes care in 2019, respectively. Although high HbA1c levels were associated with an increased risk of COVID-19 infection in a dose-dependent manner among individuals without diabetes care in 2019 (odds ratios, 1.53 and 2.17 in individuals with HbA1c of 7.0-7.9% and ≥8.0%, respectively) with a reference to HbA1c of 6.5-6.9%, individuals with diabetes care had no such trend in 2019 (odds ratios, 0.99 and 0.97 among individuals with HbA1c of 7.0-7.9% and ≥8.0%, respectively). Sensitivity analyses yielded consistent results when the variable definitions were changed and after multivariable adjustment with multiple imputation.

CONCLUSIONS

This population-based study suggests that adherence to diabetes care may modify the association between HbA1c levels and the risk of COVID-19 infection.

Diagnostic validity and solute-corrected prevalence for hyponatremia and hypernatremia among 1 813 356 admissions

Background and hypothesis

We aimed to evaluate the diagnostic validity of the International Classification of Diseases, 10th Revision (ICD-10) codes for hyponatremia and hypernatremia, using a database containing laboratory data. We also aimed to clarify whether corrections for blood glucose, triglyceride, and total protein may affect the prevalence and the diagnostic validity.

Methods

We retrospectively identified admissions with laboratory values using a Japanese hospital-based database. We calculated the sensitivity, specificity, and positive/negative predictive values of recorded ICD-10-based diagnoses of hyponatremia (E87.1) and hypernatremia (E87.2), using serum sodium measurements during hospitalization (<135 and >145 mmol/l, respectively) as the reference standard. We also performed analyses with corrections of sodium concentrations for blood glucose, triglyceride, and total protein.

Results

We identified 1 813 356 hospitalizations, including 419 470 hyponatremic and 132 563 hypernatremic cases based on laboratory measurements, and 18 378 hyponatremic and 2950 hypernatremic cases based on ICD-10 codes. The sensitivity, specificity, positive predictive value, and negative predictive value of the ICD-10 codes were 4.1%, 99.9%, 92.5%, and 77.6%, respectively, for hyponatremia and 2.2%, >99.9%, 96.5%, and 92.8%, respectively, for hypernatremia. Corrections for blood glucose, triglyceride, and total protein did not largely alter diagnostic values, although prevalence changed especially after corrections for blood glucose and total protein.

Conclusions

The ICD-10 diagnostic codes showed low sensitivity, high specificity, and high positive predictive value for identifying hyponatremia and hypernatremia. Corrections for glucose or total protein did not affect diagnostic values but would be necessary for accurate prevalence calculation.

Clinical Characteristics and Incidences of Benign and Malignant Insulinoma Using a National Inpatient Database in Japan

CONTEXT

Insulinoma is the most common pancreatic functional neuroendocrine neoplasm, yet little information on recent clinical practice in patients with insulinoma, especially malignant insulinoma, is available.

OBJECTIVE

This work aims to clarify the characteristics and practice patterns in patients with insulinoma using a national inpatient database.

METHODS

Using the Japanese Diagnosis Procedure Combination database, we retrospectively identified patients with insulinoma admitted between 2010 and 2018. We compared background characteristics and therapeutic interventions between patients with benign and malignant insulinoma. We also estimated the incidence of insulinoma using the number of patients with newly diagnosed insulinoma in 2012.

RESULTS

We identified 844 patients with benign insulinoma and 102 patients with malignant insulinoma. Patients with malignant insulinoma were younger (median, 55.5 vs 66.0 years, P < .001) and less likely to be female (55.9% vs 65.3%, P = .061) than patients with benign insulinoma. Analysis of therapeutic interventions revealed that patients with malignant insulinoma more frequently received medications (71.6% vs 49.6%, P < .001) but less frequently underwent pancreatic surgery (57.8% vs 72.0%, P = .003). Older patients were a smaller proportion of those undergoing surgery and a larger proportion of those managed with medications without surgery (P < .001). The incidence of insulinoma was estimated to be 3.27 (95% CI, 2.93-3.61) individuals per million Japanese adult population per year.

CONCLUSION

The present study using a nationwide database had a larger sample size than previous studies and revealed definitive differences in patient characteristics and therapeutic patterns between benign and malignant insulinoma.

Impact of body mass index on in-hospital mortality in older patients hospitalized for bacterial pneumonia with non-dialysis-dependent chronic kidney disease

BACKGROUND

Low body mass index (BMI) in older individuals with decreased kidney function is important because of its association with poor prognosis and frailty. Herein, we aimed to clarify the association between BMI and in-hospital mortality among older patients with non-dialysis-dependent chronic kidney disease (CKD) stratified by kidney function.

METHODS

Using data from the Medical Vision Database, this multicentre cohort study included people aged ≥ 60 years with an estimated glomerular filtration rate of < 60 ml/min/1.73 m² but without dialysis dependency, hospitalised for bacterial pneumonia during 2014-2019. We compared the risk of in-hospital death between patients with BMI categories based on the quartiles (low, medium-low, medium-high, and high) setting medium-high BMI as a reference. We further assessed the association with BMI using a cubic spline, setting BMI as a nonlinear continuous variable and a BMI of 22 kg/m² as a reference. We also evaluated the association between BMI and kidney function using a generalised additive model adjusted for interaction terms between nonlinear continuous BMI and kidney function.

RESULTS

We obtained data for 3,952 patients, with 350 (8.9%) in-hospital deaths. When compared with medium-high BMI, low BMI was associated with an increased risk of death and longer hospital stay, whereas the other two categories were comparable. Models using a cubic spline showing an association between BMI and in-hospital death showed an L-shaped curve; BMI < 22.0 kg/m² was associated with an increased risk for mortality, and at a BMI of 18.5 kg/m², the odds ratio was 1.43 with a 95% confidence interval of 1.26-1.61 when compared with a BMI of 22.0 kg/m². Analysis of the interactive effects of kidney function using the generalised additive model showed that a protective association of high BMI tapered along with decreased kidney function.

CONCLUSIONS

This cohort study suggests not only that lower BMI and low kidney function are associated with in-hospital mortality independently but also that the protective effects of high BMI weaken as kidney function decreases via the analysis of the interaction terms. This study highlights the necessity for the prevention of underweight and demonstrates the interaction between BMI and kidney function in older patients with non-dialysis-dependent CKD.

Steatotic liver index: An interpretable predictor of steatotic liver disease using machine learning with an enhanced shrinkage method

AIM

While the Fatty Liver Index (FLI) has been the most prominent among interpretable predictors for steatotic liver disease (SLD), we aimed to prepare a novel diagnostic/prognostic index better than FLI for SLD using a non-black-box and modified parsimonious machine learning method.

METHODS

We included individuals who participated in an annual health checkup in Tokyo, Japan, between January 2008 and December 2018. In the training set (randomly selected 80% of the sample), we developed a novel interpretable model, Steatotic Liver Index (SLI), using a modified method of least absolute shrinkage and selection operator regression focusing on parsimony and interpretability using as few variables as FLI, and confirming its superiority to FLI using the test set (the remaining 20%). The predictive performance of the constructed index was assessed for the diagnosis, development, and remission of SLD.

RESULTS

Among ultrasound data of 92 968 participants at the first health checkup, 20 380 (21.9%) had SLD. Using a modified method of least absolute shrinkage and selection operator regression, SLI was constructed with four variables: body mass index, waist circumference, alanine aminotransferase, and triglycerides. The C-statistic of SLI for SLD diagnosis was superior to that of FLI (0.909 vs. 0.892, p < 0.001). In participants without SLD, SLI was more accurate than FLI in predicting SLD development, whereas among those with SLD, SLI showed better accuracy in predicting SLD remission compared with FLI.

CONCLUSIONS

We developed SLI, a novel interpretable and parsimonious index for diagnosing SLD, which demonstrates superior predictive capability compared with FLI. Further studies are necessary to validate the diagnostic ability outside Japan.